THE CATHODE RESISTOR METHOD

This is the method that is best for hobby techs and do-it- yourselfers. It is far and away the safest of the two methods, and can be successfully done with medium- and even low-quality test equipment. It is performed by reading the cathode current through each power tube. The cathode current is composed of the plate current *plus* the screen current. Plate current can be identical on two tubes (tubes are matched by plate current readings) while one tube is drawing more screen current; with this method, the readings will appear to indicate a mismatch when such is not actually the case. Since the cathode current will always be higher than the actual plate current, the readings obtained with this method will tend to make you set the tubes a little colder than your calculations will indicate that they are. This promotes slightly more conservative operation, which is beneficial to tube life. If you feel that the inaccuracy is significant, or if you’d just like to know how large it is for your particular amp, you can always read the voltage drop across each screen resistor, figure the screen current (I = E/R) and then subtract it from the cathode reading. Note that these instructions assume that your amplifier is biased by applying a negative voltage to the control grids; cathode-biased amplifiers cannot be adjusted other than by changing the value of the cathode resistor(s) so this method does not apply to them. BE AWARE THAT THE ACCURACY OF THE RESULTS YOU OBTAIN WITH *ANY* BIASING METHOD WILL BE DIRECTLY AFFECTED BY THE QUALITY OF YOUR TEST EQUIPMENT, AND YOUR SKILL IN USING IT. If any part of the following instructions doesn’t make sense to you, seek help from someone with more experience. Tech support is available from NBS Electronics AFTER 1:30 PM Denver (Mountain) time, Monday through Saturday. The number is 303-778-1156; ask for Lord Valve.

WHAT YOU WILL DO:

• Replace the ground wire on each power tube socket with a 1-ohm resistor.
• Read the voltage drop across this resistor (in millivolts) with your DMM.
• Read the plate voltage.
• Use the above readings to calculate the static dissipation wattage.
• Adjust the bias to obtain the best tone, while keeping the tubes within specifications.

A SUGGESTION: You may want to practice taking these readings and making these adjustments with your old tubes still in the amp, or with a spare (used) set. That way, you won’t fry your new tubes if you make a mistake.

On some sockets, the pins are numbered on the bottom (terminal) side; it is sometimes difficult to tell which pins the numbers go with. The best way to tell which pin you are looking at is to count clockwise from the notch on the locator “keyhole” in the center of the socket, with the first terminal clockwise from the notch being pin ONE. This assumes that you are looking at the sockets from the BOTTOM, or UNDERSIDE.

Most guitar amplifiers use output tubes which have the same (or very similar) basing. (”Basing” refers to the order in which the internal elements of the tube are connected to the pins on the bottom of the tube.) The 6L6, 7581A, 6V6, 6550, EL34, 5881, KT66, KT88, KT90, KT100, etc. are common guitar amplifier power tubes, and are all easily biased with this method. If you have an amp which uses power tubes which are not listed above, you will need to consult a spec manual (the RCA RC-30 Receiving Tube Manual is one of the best) for the basing, and adjust these instructions accordingly.

You’ll need a 1-ohm resistor for each power tube in the amp. All of the tubes listed above have their cathodes on pin EIGHT, which will be grounded to the chassis. On some amps, such as Marshalls, pin ONE will be tied to pin EIGHT, and both will be grounded. On older Fenders, pin ONE is usually used as a tie-point for the 1.5K grid-stopper resistor, and the negative bias voltage will be on this pin. DO NOT GROUND PIN ONE ON A FENDER AMPLIFIER, or you’ll get a big surprise. (Expensive, too.

REMOVE the ground wire from pin EIGHT on each tube, and REPLACE it with a 1-ohm resistor. On older Fenders, the ground wire is a piece of copper braid; unsolder it from the socket pin but *don’t* cut it off where it attaches to the chassis. Solder the resistor to pin EIGHT, and attach the free end of the resistor to the ground braid you unsoldered from pin EIGHT. Repeat this for all of the power tube sockets. I prefer to use 2-watt resistors because they have thicker leads which will take more abuse, but half-watters will work just fine. The accuracy of your measurements will be directly related to the tolerance of these resistors; precision 1% (or better) types are suggested.

A WORD ABOUT TAKING READINGS: It is vital that your probe tips make *good* contact with the pins you’re attempting to read. Tube-socket terminals often have a residual coating of non- conductive flux on them, and it is necessary to push the probe against the terminal hard enough to break through this coating. Most test probes supplied with today’s meters are fairly blunt; if you can come up with a set of “insulation piercing” test probes, these will help solve this problem. Don’t go overboard when pressing the probes against terminals, however…the probe tip may suddenly slip off the terminal and slide down against the chassis while the side of the metal prong is still touching the terminal. This will result in a dead short from that terminal to ground, and if you’re reading plate or screen voltage the resulting spark (and loud popping noise) might make you jerk back reflexively, pulling the chassis off your workbench and into your lap, injuring you or breaking your tubes. BE CAREFUL! Also be aware that amp chassis surfaces are usually dirty or corroded, so the advice above goes double for touching a probe to the chassis (ground). You may even want to take a small flat file and scrape a nice shiny spot on the chassis in a convenient place, to aid you in making a good ground contact.

Turn your amp on, but leave it on STANDBY. Set your DMM to the highest DCV scale, ground the black probe to the chassis, and take a reading from pin FIVE of any power tube socket. You should see a negative voltage in the -35 to -50 volt range if the amp has EL34s, or in the -45 to -60 volt range if the amp uses 5881s, 6L6s, or KT66s. KT88s, 6550s, KT90s, and KT100s can have bias ranges that go as high as -100 volts. Amps which use 6V6s will usually have bias supplies which produce voltages that are similar to EL34 amps…but not always. Note that you should *not* have any power tubes installed in your amp yet.

First, locate the bias trimmer. (Possibly a little square blue thingy with a screwdriver-adjust slot in the center, or a round black thing that stands on three legs, or, for an old Fender, a full-sized pot with a screwdriver-adjust slot on both sides; newer PCB-type Fenders use three-leg horizontal trimpots, if they have a bias-adjust pot at all.) Next, adjust the bias control until you have MAX NEGATIVE voltage on pin FIVE. (In other words, rotate the bias trimmer until you obtain the highest negative voltage that the bias supply is capable of delivering.) Note that on some really old amps, the bias supply may be controlled by the standby switch; if you see *no* negative voltage on pin FIVE, you may have one of these amps. This is a poor design; see a tech about having the bias supply moved to the “hot” side of the standby switch. Install your tubes (the amp is still on STANDBY, remember) and wait a few minutes for them to warm up. Take the amp off STANDBY and make sure your DMM is still set to the highest DCV scale; take a reading between the chassis (ground) and pin THREE on any power tube socket. Remember, the BLACK probe always goes on the CHASSIS. Write this voltage down; you’ll need it later.

Now, set your DMM to the lowest DCV scale (usually 200 mV) and take a reading across the 1-ohm resistor(s). (This reading can be interpreted directly in milliamperes, because one millivolt across one ohm equals one milliamp. Ohm’s law says so, and you ain’t gonna argue with *that*, are ya? It’ll be pretty low, because you have the bias trimmer set to max neg voltage.

Adjust the bias trimmer (”pot”) until you get a reading across the 1-ohm resistor(s) somewhere in the 30-40 mV range, for everything but 6V6s or EL-84s. For 6V6s, you’ll want to start out at around 20 mA and work upward from there. Note that the polarity of this reading is unimportant; only the numerical value means anything. (If you put the black probe on the side of the resistor that is grounded to the chassis, you will get a POSITIVE reading.)

MULTIPLY the voltage you read on pin THREE earlier by the reading you just obtained from the 1-ohm resistor. (Example: 450 Volts times 35 milliamps, or .035 Amperes.) This will give you the STATIC DISSIPATION WATTAGE at which the tube is operating. (It’ll be wrong, but more on that later.) The above example gives a static dissipation of 15.75 WATTS, which is well within specs for an EL34 (fairly cold, in fact) or a 5881/6L6. See TABLE “A” (at the end of this article) for suggested static dissipation wattages for most of the common octal-based tubes discussed here. To sum up what this calculation is, PLATE VOLTAGE times CATHODE CURRENT equals STATIC DISSIPATION (IDLING) WATTAGE. It is important not to exceed the tube manufacturer’s specification for this parameter, because tube life will be shortened. At extreme settings, tube life will be measured in MINUTES…be advised.

Take another reading from pin THREE (remember to set your meter on the HIGHEST DCV scale before you do!) and write it down. This new reading should be LOWER than the first reading you took, because the tubes are drawing more current now and the plate voltage will drop somewhat. Multiply this new reading by the value you measured across the 1-ohm resistor(s); this will give you the idling (static) wattage. The cooler you run the tubes, the longer they’ll last. If you dig the way the amp sounds when the tubes are idling at only 12 watts, fine…don’t worry about it. (6V6s, though, will be running fairly *hot* at 12 watts.)

Remember, each time you adjust the bias control, you’ll have to take a new reading from BOTH the 1-ohm resistor *and* the plate (pin THREE) and multiply them to see how hot the tubes are running. You can play your guitar through the amp each time you adjust the bias, and see how you like it. You can even adjust the bias by ear, and then take readings as outlined above to see if the tubes are being operated within their ratings. If you find that you only like the tone when the tubes are operating near their limits, you may decide to trade some tube lifetime for the tone you seek. If you like the tone with the tubes running cold, you’ll obtain significant extra tube life that way. It’s *your* call.

If you see a few milliamps difference between the readings on the 1-ohm resistors, don’t sweat it; this could be due to poor matching (not a factor if you bought ‘em from *me* , differences in screen current between the tubes, or differing leg impedances in the output tranny’s primary. (All of those things are fairly common in guitar amps.) Note that for an amplifier which uses four (or more) power tubes, balance between the two sides is more important than having identical readings from socket to socket. You should add the readings for each pair; if the left pair is close to the right pair, things are fine. If the left pair reads 32 and 34 milliamps (total = 66) and the right pair reads 35 and 31 milliamps (total = 66) then you’ve got a nicely balanced output stage, even though some of the tubes are running slightly hotter or colder than others. Having the currents balanced on the two legs of the tranny helps eliminate 120 Hz power-supply ripple from the output. Note that you can swap the tubes around to obtain the best current balance, since you can take individual readings on each socket. If you see a large difference between them (say, 8-12 milliamps) this means you need to find out why this difference exists. One thing you can do is SWAP the tubes into the opposite sockets and take new readings. If the bogus readings are consistent on the SOCKETS, then you’ll need to look at the circuitry to find out the cause. If the readings MOVE with the TUBES, you can be fairly sure you have a poorly-matched pair/quad.

Once you have everything adjusted to your taste and you’re sure the tubes are being operated within specifications, leave the amp fully powered up for three or four hours. Eyeball the tubes every fifteen minutes or so, to make sure the plates aren’t turning red. You are doing this to let the tubes “settle” into their new operating con- ditions; at the end of the settling period, take a final set of readings to make sure everything is still OK. If any readings have drifted significantly, readjust the bias accordingly. Note that the incoming line voltage directly affects all of the voltages in the amp; you may want to read the line voltage occasionally to see if this is happening. Line voltage will drop a bit around supper time (lots of juice being used for cooking) and also after sunset. If the line was 120VAC when you completed your biasing procedure and it’s 117VAC when you take your final readings after the settling period, expect to see a corresponding small drop in your measurements.

You may decide to purchase a “bias-probe” type device; this is a gizmo which consists of an “interruptor” socket/plug assembly that goes between the tube(s) and the amp’s socket(s). This test adaptor will have a couple of testleads hanging out through a hole in the side, for connection to your meter(s). If you do get one of these, there is no need to install the 1-ohm resistors on the tube sockets as outlined above. You can use the readings obtained from the adaptor sockets in place of the readings normally taken across the one-ohm resistors. BE AWARE THAT THERE ARE TWO TYPES OF THESE ADAPTORS COMMONLY AVAILABLE. One type *breaks* the cathode connection, and instructs you to connect the testleads to the CURRENT jacks on your meter. The other type contains a one-ohm resistor in series with the cathode pins, with the testleads connected to either side of the resistor; this type instructs you to connect the testleads to the VOLTAGE jacks on your meter. It has been my experience that some amps (especially old Marshalls) do not react well to having several feet of wire inserted in series with the power tube cathodes, and will oscillate like crazy. Therefore, if you decide to get a set of these test adaptors, get the ones which use an internal 1-ohm resistor.

REMEMBER…THERE ARE VOLTAGES PRESENT INSIDE EVEN THE SMALLEST TUBE AMPLIFIER WHICH WILL KILL YOUR ASS JUST AS DEAD AS A HAND GRENADE WILL!! If you’re not familiar with high-voltage safety, seek guidance from someone who is. BTW, an oven mitt or a pot-holder (real men like me use welding gloves) will come in handy for handling hot power tubes if you need to switch sockets; you don’t want to let the tubes cool off too much while you swap them before taking new readings.

THE OUTPUT TRANSFORMER SHUNT METHOD

This is the way many pro techs measure plate current. A *good* quality DMM is required for this measurement. (When it comes to good DMMs, you have three choices…Fluke, Fluke, and Fluke.) This section assumes that you know a bit more about your amp, and how to use your testgear. If any of it is unclear, DON’T TRY THIS.

NOTE… Marshall amps have output transformers which have a very low DC resistance in the primary winding. If your meter’s internal current-measuring shunt resistor is a relatively high value (~ 10 ohms, for instance) it will induce significant error into a transformer shunt measurement. This is because when such a meter is connected in parallel with half of the output transformer’s primary, a significant portion of the current is not flowing through the meter, and can’t be read. For this reason, unless you’re *sure* you have a meter with a low internal current-sensing resistor (~ 1 ohm) the shunt method is *not* recommended for use on Marshall (and other low DCR) output transformers. Fairly good results can be achieved on Fenders, though.

WHAT YOU WILL DO

• Read the current flowing through each leg of the output transformer’s primary.
• Read the plate voltage.
• Use the above readings to calculate the static dissipation wattage.
• Adjust the bias to obtain the best tone, while keeping the tubes within specifications.

For this particular reading, you’ll need to change your test leads to the CURRENT input jacks, and select the 200 mA DC range. The two probes are applied to the center tap and either of the ends of the output transformer’s primary. (On a Fender, for instance, the center-tap is RED, and the two plate wires are BLUE and BROWN. On a Marshall, the center tap is usually BROWN, and the plate leads are usually RED and WHITE.)

On some amplifiers, the easiest way is to put one probe on pin THREE of either socket (or of either of the two sockets on each side) and the other on the center-tap, which will be located at some distance from the socket. Some amps (like the Marshall JCM 900 series, for instance) have all the wires soldered to terminals on the bottom of the output transformer, conveniently sticking up right where you can reach them.

The current that would normally flow through half of the transformer’s primary winding is “shunted” through the meter, and thus measured. A small amount still flows through the part of the winding you are shunting, but the transformer’s resistance is much higher than your meter’s internal resistance. (See “NOTE” above.) Nearly all of the current flows through the meter.

SAFETY ADVISORY

BE WARNED…for all practical purposes, a meter set to measure CURRENT is equivalent to a STRAIGHT WIRE. This means that as soon as you touch either probe to the high voltage circuitry, THE OTHER PROBE NOW CARRIES THE SAME VOLTAGE. If you drop the probe and it lands on your arm or leg, you could be electrocuted. If it lands on the chassis (or anything else that is at earth or circuit ground potential) a huge spark will be generated, along with a noise like a small firecracker. (Please don’t ask how I know this. The probe tip will be partially melted, and at the very least, the meter’s internal fuses will blow. At worst, the meter will be history. Shorting the HV to ground isn’t especially good for the amp either, and may blow the amp’s fuse or damage the circuitry. You can easily kill a rectifier tube this way. BE ESPECIALLY CAREFUL NOT TO LET A PROBE SLIP OFF A TERMINAL AND HIT THE CHASSIS WHILE YOU ARE TAKING A READING! BE *EXTRA* CAREFUL TO MAKE SURE YOUR FINGERS DON’T SLIDE DOWN THE PROBE AND COME INTO CONTACT WITH THE METAL TIP!! And make DOUBLE DAMN SURE you know which two points in the circuit you are supposed to touch the probes to, because if you accidentally touch the bias supply and the plate supply at the same time, you won’t *believe* what happens. IF YOU’RE NOT *SURE* WHAT TO PROBE, *DON’T* PROBE IT!!
Once you’ve obtained the current readings from both sides of the output transformer’s primary, you’ll need to take a plate voltage reading so you can calculate the static dissipation wattage (as outlined above in the CATHODE RESISTOR method) and decide whether you need to increase or decrease the plate current. Note that if you are using the OPT shunt method with an amplifier which uses more than one tube per side on the transformer, you will need to divide the current reading on each side by the number of tubes used. Example: you read 88 mA on one side of a Twin Reverb’s output tranny; that’s 44 mA per tube, since there are two on each side. (4 total.)

REMEMBER TO REMOVE THE TEST LEADS FROM THE CURRENT MEASURING JACKS, AND TO SET THE METER TO THE HIGHEST DC VOLTAGE RANGE BEFORE YOU TRY TO READ THE PLATE VOLTAGE!! If you attempt to read the plate voltage with your meter still set up for a current reading, the results will be spectacular (as outlined above.) Since you may need to take several plate CURRENT and several plate VOLTAGE readings before you are finished setting the bias, you will need to be extremely vigilant about changing the meter settings (and the test leads) each time you take the different readings. Most pro techs use TWO METERS for this procedure, leaving one set up for current and one for voltage. (I use a handheld meter for the voltage reading, and a bench meter for the current.)

Once you have the necessary readings, the procedure is the same as for the CATHODE RESISTOR method: read, multiply, listen, adjust, read, multiply, listen, adjust, read, multiply, etc. Don’t neglect the “settling” period, either. BE CAREFUL!!

TYPES OF (FIXED) BIAS CIRCUITS

Many amps which use “fixed” (negative grid) bias have provisions for adjusting the negative grid voltage upward or downward. Making the grids LESS negative will cause MORE current to flow through the tubes. Some amplifiers don’t have a bias-adjusting control (pot) but instead use a fixed resistor to set the voltage. If you encounter one with a fixed resistor, the best thing to do is convert it to an adjustable type. Most of the time, the fixed resistor will be in parallel with the bias capacitor; the lower this resistor’s value is, the lower the bias voltage will be. If you can locate and identify this resistor, you can replace it with a simple network consisting of a (lower value) resistor in series with a potentiometer. What you’ll be shooting for is a range of adjustment that goes from LESS voltage to MORE voltage than is set by the (existing) fixed resistor. Take the value of the fixed resistor and divide by two; pick the closest standard value to your result, and put it in series with a pot which is as close to the original resistor’s value as you can find. Example: the existing resistor is 33K; use a 15K resistor in series with a 25K pot to replace it. The original resistor was 33K; you now have the ability to adjust the value from 15K to 40K. This should provide you with sufficient adjustment range to set any plate current you wish. If not, use a different value pot or resistor.
Some amps have a “balance” type bias adjustment, which allows you to vary the negative grid voltage between the two halves of the output stage; this makes a “matched” set of tubes less crucial to good performance, although it can’t compensate for tubes that are wildly different. If you encounter this circuit, the easiest way to adjust it is to simply “tune” the control for minimum 120Hz hum on the output. This type can be modded to the *best* type, which is not only variable from side-to-side, but adjustable up-and-down, too. Usually, this circuit will have the “balance” pot’s wiper connected to a resistor which is grounded at the other end. You can replace this resistor exactly as outlined above (half the value, add a pot, etc.) and have the best of both worlds.

If the simple mods outlined above (and the reasons for making them) don’t seem perfectly clear to you, DON’T TRY THEM. A schematic (and the expertise with which to interpret it) will go a long way towards helping you do them correctly. You can have the mods performed by a tech, and then do your own biasing from then on, if you wish.

TABLE A

Suggested MAX static dissipation wattages for common guitar amplifier tubes. You can exceed these (although I wouldn’t do it with a 6V6) at the cost of some tube lifetime. The colder you run ‘em, the longer they will last. Remember, as long as you don’t run the tubes hot enough to damage them, there are *no* rules about how much current to set them for. If you like the way your amp sounds when your 6L6s are only pulling 14 watts, bully for you… you probably won’t need to retube it for 10 years. I know that many sources for biasing information just specify plate (or cathode) current settings; telling you to bias your 6L6s at “35 milliamps” is nonsense. Unless you take the plate voltage into consideration, a current specification is meaningless. For instance, 40 mA at 250 volts is 10 watts; the same 40 mA at 500 volts is 20 watts… TWICE as much. In both cases, the current is the same. Amps vary; two identical amps can have plate voltages which differ by as much as 20%. Just because you have a schematic that specifies the plate voltage in your amp as being at 450VDC, don’t expect to see that voltage when you take a measurement. TAKE the reading, don’t assume the voltage will be as specified. Trust your meter. Most of these suggested MAX wattages have been arrived at through my own experience. NOTE THAT THESE FIGURES ARE NOT TARGETS, BUT MAXIMUMS. This is important. You are not looking to set your static plate dissipation to the values listed here, but to set it at a level which produces the tone you are looking for without exceeding them.

KT88, KT90, KT100 can be treated as 6550s, although all three of these tubes are supposed to be able to take more current. The ultimate test is to view the tubes’ plates IN THE DARK, after they have been powered up for 15-20 minutes. If you see any red spots, back the current off a bit. One exception to this is the NOS 6V6; some of these will show a slight red “stripe” down the center of the plates even when they’re set fairly cold. I’ve seen them run for years in this condition. *Large* red blotches, or even the entire plates turning red, is what you want to watch out for.

Entire contents copyright Lord Valve, 08/24/01

There is a lot schematics on the internet, but finding a good one is pretty hard. I want to build something like that:

• Good power: 10 Watts seems to be a good starting point.
• Something simple. I want to build an amp, I can fix by myself. So I decided to build a Single Ended one. (Not a push-pull) Push Pull provide far more power, but need more components and harder to build.
• Low cost. The best issue for this kind on amp, is the price. You need to buy some output transformers, big power supply transformer and choke. And of course, when we talk about HIFI this cost a lot of money.
• Look’s good. Something which is important is the way it looks. So I will have to take really care of that.

For people who have never seen this kind a stuff here a little pic I ripped out.

Tubes

For this design, I decided to use some EL34. Check out DrTube[0] for additionnals data. They offer good output power for a decent price.
The reflected load permits to use a KT88 instead if this doesn’t provide enought power.

For the preamp, I think I gonna use a 6SN7 in SRRP. This provide a good gain, and very good linearity.

Output transformers

EL34 has a reflected load of 2KOhm, so I need to find coresponding transformers. Of course this cost a lot of money, Hammond HIFi (not guitar) output transformer for 20w cost something like 99$ each. And
even worst, I unable to find 2.0K only 2.5 are currents.

I found this one:

• EuphoniaAudio[1] (205$ the pair)
• Hammond[2] (99$ each)
• Electra Sud Ouest (86Euros)
• Elettronocanovarria[3] (75Euros)

Despite, Elett.. offer good price, I unable to buy them here, cause the vendor speack only italian.. and that sucks. Electra sud has a good price too, so I guess I will buy them here if I don’t find anything else.

Update: I found a matched pair of 2.5K / 20W transformators on eBay.

Power Supply

This part isn’t really clear for me right now. Singled Ended amps doesn’t need regulated PSU because they work in pure class A, so the current is fixed (throught the tube bias) but It need to be really stable to avoid distortion.I don’t want to buy a large transformer because they cost a lot of money and doesn’t feet with my needs. EL34 use a quite low voltage (~250v) but need quite a bit of current ~100ma. I think, a good isolating transformer can do the job. Something like 230*230 at 200Va doesn’t cost a lot of money.

Schematic

This is really a hard part. But I found 2 good example:

• first[4]
• and second[5]

To be continue…

Links

• [0] DrTube: http://www.drtube.com/tubedata.htm
• [1] Elettronocanovarria: http://www.elettronicanovarria.it
• [2] Hammond: http://www.tubetown.de/ttstore/index.php/cat/c18_Hammond-SE.html
• [3] EuphoniaAudio: http://euphoniaaudio.netfirms.com/ea/nfoscomm/catalog/product_info.php?products_id=28
• [4] http://perso.orange.fr/michel.terrier/radiocol/detail2004/ampli-el34-se-stereo.htm
• [5] http://web.mac.com/scress1958/iWeb/Steve%27s%20Tube%20Trials/EL34%20Amp.html

In electronics, a vacuum tube (U.S. and Canadian English) or (thermionic) valve (outside North America) is a device generally used to amplify, or otherwise modify, a signal by controlling the movement of electrons in an evacuated space. For most purposes, the vacuum tube has been replaced by the much smaller and less expensive transistor, either as a discrete device or in an integrated circuit. However, tubes are still used in several specialized applications such as guitar amplifiers (also called a valve amp outside the U.S.) and high power RF transmitters, as a display device in television sets and in microwave ovens.

Operation

Vacuum tubes, or thermionic valves, are arrangements of electrodes in a vacuum within an insulating, temperature-resistant envelope. Although the envelope was classically glass, power tubes often use ceramic and metal. The electrodes are attached to leads which pass through the envelope via an air tight seal. On most tubes, the leads are designed to plug into a tube socket for easy replacement.

The simplest vacuum tubes resemble incandescent light bulbs in that they have a filament sealed in a glass envelope which has been evacuated of all air. When hot, the filament releases electrons into the vacuum: a process called thermionic emission. The resulting negatively-charged cloud of electrons is called a space charge. These electrons will be drawn to a metal “plate” inside the envelope if the plate (also called the anode) is positively charged relative to the filament (or cathode). The result is a current of electrons flowing from filament to plate. This cannot work in the reverse direction because the plate is not heated and cannot emit electrons. In it’s simplest form a vacuum tube can be created to operate as a diode: a device that conducts current only in one direction. A third element called a “control grid” can be added to the design which provides the ability to amplify a signal. Other configurations are also possible including the Pentode, a tube with 5 active elements providing an additional amplification factor. There are a large number of tube varieties and uses. This is only a very brief overview and we suggest consulting additional resources if you are interested in additional information. Some Information provided by wikipedia.org.

Diode	Triode

Why Use Tubes in Guitar amps ?

Most good guitar amplifiers use tubes rather than solid-state components. Why tubes ? The amplifier is a critical element in achieving the sound the musician desires. Tubes provide the tone that musicians want. Tube amps are warmer, richer and have a more desirable tone than solid-state amps. The distortion and speaker-damping characteristics of a tube amp with an output transformer matched to the speaker load is hard to replicate with solid-state devices. Tube amps are particularly popular with serious musicians. Many musicians prefer to play vintage Fender, Marshall and Gibson amps. Replacement tubes and transformers are readily available for these amps however there are many boutique amp manufacturers making new tube amps with a vintage sound.

Amplifier classes

Amplifier circuits are classified as A, B, AB and C for analog designs, and class D and E for switching designs. For the analog classes, each class defines what proportion of the input signal cycle (called the angle of flow) is used to actually switch on the amplifying device.

What’s a Class A Amp ?

In a class A amp 100% of the input signal is used. The amplifier is passing current at all times even when you are not playing. The instant you strike a note it’s immediately fed to the speakers resulting in a “fast” sound. Class A is very inefficient but usually gives very low distortion and is generally a better sounding amp at low volumes. Class A amps are often more expensive boutique amps. Some of our Divided by 13 amps are Class A.

What’s a Class B Amp ?

A class B amp uses 50% of the input signal. Class B is different from Class A in that there is no current flowing when the output is at idle and turn on from zero current when a signal is present. In a push-pull Class B amp design each of the output circuits produce one half the audio waveform with each circuit not producing any current flow when the other circuit is operating. Class B designs tend to have more crossover distortion and require a less beefy power supply. Many popular guitar amps use class B designs including Fender and Gibson amps.

What’s a Class AB Amp ?

As the name implies class AB amps exhibit some characteristics of class A amps and some of class B amps. In a class AB amp design, more than 50% but less than 100% of the input signal is used . If an amp uses class A mode for a portion of it’s output then has to apply additional circuitry for the remainder of it’s output then it is considered a class AB Amp. Class AB amps are also more efficient than a straight class A therefore does not require as large a power supply.

What is an Amp Transformer ?

A transformer is an electrical device that transfers energy from one circuit to another by magnetic coupling with no moving parts. It consists of a minimum of two coils, the primary and the secondary, wound on the same core. An alternating current in one winding creates a time-varying magnetic flux in the core, which induces a voltage in the other windings. Transformers are used to convert between high and low voltages, to change impedance, and to provide electrical isolation between circuits. This is useful in converting the voltage from a wall outlet, typically 120 or 240 volts, into a higher voltage required by tubes in tube amps . typically 400V or more, and a lower voltage for the tube filament, typically 6.3 or 12.6V. There are several transformers used in tube amps. Some information provided by http:// en.wikipedia.org /wiki / Transformer

Output transformer – An output transformer is used to match the low impedance of a speaker voice coil to the high impedance of a tube output stage. Output transformers consist of at least two windings, a primary and a secondary. Some output transformers have multiple impedance taps on the secondary side, to allow matching to different speakers, typically 4, 8, and 16 ohms

Power transformer – A Power transformer converts the incoming line voltage to a higher or lower value for use in the guitar amplifier. Typically, the power transformer will have at least one primary, but sometimes two or more, to allow use at 120V or 240V. In an amp the power transformer will generally increase the voltage to 400 volts or more for the tube plate. There will also usually be a 6.3V filament winding. There is also sometimes a 5V. winding for use with a tube rectifier.

Choke – Another term used for an inductor, most commonly an inductor used as a power supply filter.

Fender Transformer Chart

Dating Fender Amps – Using the Fender Tube Chart

Look inside the amp (but don’t stick you hand in there, even after being unplugged the amp may retain a dangerous electrical charge), there should be a tube chart on most amps. On this chart there is a hand stamped date code consisting of 2 letters. For example AD would be April 1990 and DG would be July 1954.

Dating Marshall Amps

In 1969 Marshall introduced a date coding system. Some of the older Marshall amps have an inspection sticker on the top of the chassis which usually has the day, month and year the amp was actually made or inspected. Here’s a chart with date codes for Marshall amps. Note, “A” Date Code ran for 18 months (July 1969 to December 1970) so the “B” date Code was never used and has been omitted. Use the serial number to determine the date code. The serial number is generally located on the back of the chassis but from 79 to 80 it was on the front panel. From 1969 to 1983 the date code was after the serial number. From 1984 to 1992 the model number was first, then the date code, then the serial number.

A=1969      A=1970
C=1971      D=1972
E=1973      F=1974
G=1975      H=1976
J=1977      K=1978
L=1979      M=1980
N=1981      P=1982
R=1983      S=1984
T=1985      U=1986
V=1987      W=1988
X=1989      Y=1990
Z=1991      Z=1992

Early Marshall Model Codes (approx Mid 69 to Late 83)

/A = 200 Watt
SL/ = 100 Watt Super Lead
SB/ = 100 Watt Super Bass
SP/ = Super PA
ST/ = 100 Watt Tremolo
S/ = 50 Watt
T/ = 50 Watt Tremolo

Later Marshall Model Codes (approx Early 84 to Late 92 )

A/ = 200 Watt
SL/A = 100 Watt Super Lead
SB/A = 100 Watt Super Bass
SP/ = Super PA
ST/A = 100 Watt Tremolo
S/A = 50 Watt
T/A = 50 Watt Tremolo
RI = Reissue

Early Marshall Date Code Example
EXAMPLE: SL/A 25353 E
SL/A = Model Code
24523 = Serial Number
E = Date Code
This amp would be a 100 Watt Super Lead 1973

Later Marshall Date Code Example
EXAMPLE: S/A S 24523
S/A = Model Code
S = Date Code
24523 = Serial Number
This amp would be a 50 Watt 1984

Common Tubes used in Vintage Amps and modern replacements

Six JAN (joined army navy) tubes, after some 40 years of rest, finally make music…

I started this project with the idea to build a simple single-ended tube amplifier. I didn’t have any experience with the construction of tube amps so I was looking for a beginner’s project. Such a project should be simple (for easy debugging) and it should also be cheap. Nobody likes to blow up a component of 300 dollars by a small lapse of attention…

As far as simplicity is concerned, nothing beats the “Darling” series of amplifiers which are described on the website of New Jersey resident Bob Danielak 1. They employ power triodes type 1626 which can be obtained cheaply from various sources (I bought several for 3 US dollars apiece). Old readers of this page may remember that the 1626 was employed in various transmitters for the American airforce, e.g. the AN/ARC-5 series (BC-457, 458, 459 and 696). Since they were used in military airplanes, 1626s are rugged valves which can stand shocks, vibration and electrical insults. Once I had excessive power dissipation in two 1626s because of a wiring error. The plates became fiery red and blazingly hot and remained in this condition for two minutes before I discovered my mistake. Yet these tubes still sound and measure OK. Apparently, Darling amplifiers make excellent projects for dummies like me!

I decided to make Jeremy Epstein’s DC-coupled variant of the Darling amp with two parallelled 1626s per channel 2. The schematic is shown below:

For my first try, I followed Jeremy’s schematic to the letter, although I used a GZ34 rectifier in stead of a 5U4G. The Hammond 125E output transformers, 1.5 H choke plus the 8532 and 1626 tubes were from Antique Electronic Supply in Tempe, AZ. A NOS power transformer (Prova) supplied 2 x 345, 5 and 6.3 Volts. I acquired this tranny and the GZ34 rectifier from Frits Meuris Electronics (Sittard, Holland). The 12.6 Volts for the 1626 heaters came from an obscure transformer from my junkbox which I bought for one Dutch guilder from Conrad Electronic.

I built the amp in the old style, hardwired on an aluminum chassis in a wooden frame with all tubes, chokes and trannies on top and all electrical wires below. Special parts were not used. I took what was in my junkbox (carbon resistors and styroflex capacitors from the former DDR, electrolytics from Czechia, wirewound resistors from Vitrohm in Germany). After I had corrected my wiring error, all voltages were within a few percent of Jeremy’s values and 1 kHz square waves looked OK. No trace of microphonics. So on for the listening tests!

After a burn-in period of a few hours, I listened to live recordings from twentieth-century classical music (Schцnberg Ensemble), using the multimedia speakers which are described elsewhere on this website. The amp sounded wonderful, as if you were in a very good seat in the concert hall. I was really excited about the “SE magic”. But after a while I became disappointed. The amp sounded excellent with some programme material (chamber music, small combos), but it didn’t sound good at all with large orchestra, grand piano, church organ, or big drums. The presentation seemed then “anaemic”. Especially piano recordings sounded terrible. Measurements indicated what was wrong:

The frequency response of this version of the amplifier was 75Hz…14 kHz (-3 dB). With orchestral music and grand piano, I was missing the lower two octaves. For Jeremy this didn’t cause problems, for he employs his own Darling as the treble amp in a bi-amped system. But as a stand-alone amp, the results were not satisfactory.

Jeremy thought the cheap Hammond transformers were limiting the bandwidth. So I decided to try more expensive output trannies. I replaced the Hammonds with Lundahl LL1664s from Sweden which I purchased from Aqua Blue in Belgium. The (almost fivefold) difference in price certainly resulted in improved specs (see below).

With the Lundahls, the frequency response was flat from 8 Hz…55 kHz (-3 dB). This was a major improvement! Grand piano now had weight and sounded like the real thing. Sibilants were somewhat emphasized in the Hammond 125E version, but sounded natural with the Lundahl LL1664.

So I was very happy for a while. But the perfectionist bug hit me once again. IMHO, the amp still had a major flaw. There was audible hum in the output. With the multimedia speakers, the hum was not objectionable during the reproduction of music. But in silent passages, it became annoying. Additional measurements indicated that the signal-to-hum ratio was only -55 dB with shorted inputs!

First I thought that I had made a grounding error, but this proved not to be the case. Then I thought: maybe I should use DC heating. However, there was no improvement when I hooked up the 8532 heaters to a DC power supply. I noticed that the hum had a frequency of 100 Hz rather than 50 Hz. I concluded that it originated from the B+ supply.

After running some simulations on Ben Duncan’s PSU designer program, I decided to modify the B+ circuit. After the GZ34, I maintained the 47 uF electrolytic capacitor. The stereo channels are connected to this cap via individual LC filters (10 H chokes and 200 µF/500V electrolytics). This modification resulted in a 29 dB improvement of the signal-to-noise ratio. The capacitive load to the GZ34 remains within safe limits. And hum and noise are now completely inaudible.

Specifications of our modified Darling:

Output power: 2 x 1.5 Watts in 8 Ohms

Voltage gain: 8.6 x

Input sensitivity: 400 mV

Frequency response: 8 Hz…55 kHz (-3 dB)

Signal-to-noise ratio (guesstimated): -84 dB (related to 1.5 W output)

This is an excellent amp. It sounds lively, agile, and natural. Makes you tap your feet and sing along with the music.

Verdict after two years of use: We listen daily to this amp on the large Jericho horns, so you can guess that we are pleased with its sound. However, the NOS power transformer (Prova) suddenly broke down (catastrophic damage to the insulation of the B+ windings) and we had to resurrect our killed Darling, using a custom-made transformer (purchased from Kent Electronics). We now used a metal chassis (Hammond) and we replaced the 1000µF 250V electrolytics with 47µF 400V polypropylene capacitors (Audyn-Cap, Intertechnik). This slightly improved the sonics. With the electrolytics, the sound could occasionally be somewhat “hard”, with the foil caps this never happens, not even in difficult passages. This tube amp sounds sweeter than any of my solid-state products. Finally, we added volume controls (100k log). Now, the amp can be driven directly by line-level sources (CD players, FM tuners). A preamp is no longer necessary.

Finally, a picture:

Darling (front view). A protective cover was mounted over the transformer after the testing phase.

References

¹ Bob Danielak, “Darling” and “DC Darling” SE 1626 amps”

² Jeremy Epstein, “My experiments with Bob Danielak’s ‘darling’ 1626 amplifiers”

So I sat down with a calculator and the tube datasheets, and came up with some circuit values. But I forgot about the loading of the output stage, so the driver stage didn’t work correctly the first time I tested it. The driver stage clipped before the output stage… But I recalculated the driver stage, and the circuit then worked. I built my “prototype” on a piece of wood. The lethal result can be seen below.

The Hammond transformer didn’t have a heater winding for the 2A3, but it did have a center-tapped 5V rectifier winding, so I used half of that for the 2A3 heater. But then I didn’t have a heater winding for the rectifier… So I found a nice “junk” tube solution – the 6BY5GA! A dual rectifier or damper tube with a 6.3V heater. The 6BY5GA has a very low voltage drop, but it is not meant for use with large input capacitors. So I used a 5H choke for a choke input filter. The resulting B+ voltage is about 285-290V from a 600Vct winding!

The amp sounds great, even with the small output transformer. The dual 6SN7 input/driver stage makes the amp very sensitive – a bit too sensitive probably… I have replaced the Chinese 2A3 with a Sovtek. Both tubes sound good, but the Sovtek is a bit more HiFi’ish. I have bought some better output transformers, but I haven’t installed them yet. When something sounds good I don’t mess too much with it. I also have a Hammond chassis waiting to make the amp a bit more cat safe…

The schematic can be seen here:

I seldom design SE amp more than 2 stages, but my client request for higher gain. 6SL7 can give enough gain with only 2 stages but less current, I prefer higher current, and the sound quality of 6SN7 but less gain, that’s why I’m using 3 stages in this design. The 2A3 working at plate current of 60mA with cathode-plate voltage of 300V. The EH’s 2A3 can handle this 18W of dissipation. I had tried to change it to chinese 2A3 with better sound quality but it can’t take that high dissipation, need to reduce to 14W. Since my client love the EH’s sound, I leave the EH tubes on. I use DC filament supply from 5V ac, using a bridge rectifier and a 0.25 ohm resistor to get the 2.5V DC. If I build 2A3 for myself, I will use AC as filament supply, and use tube rectifier and choke to supply B+, that will sound even better, I think.

Nothing much to talk about the sound quality. As expected, 2A3 always put a grin on my face every time I I had build one and listen to it for the first time. So far I haven’t hear anyone says that he or she don’t like the sound of 2A3 SE amp. Tell me which 2A3 SE amp sound bad if you could find one……….

I’m designing another 2A3 SE amplifier (for my own use) that using 717A door knob tube, and will put it on here when finished. Now I’m still trying to find out the characteristic of 717A tube, and optimized it to drive 2A3. I can’t get complete data sheet for 717A, if anyone of you found the data sheet, I would like to have a copy!

The schematic of this 2A3 will be on soon, I’m quite lazy to draw it using any design software, I will draw it manually and scan as jpg file. Stay tuned for the schematics and the new 717A driven version.

The simple but messy layout, I build it without a schematic in hand, The 2 big rectifier is DC for filaments. I don’t like using DC for 2A3’s filaments. Again, customer’s request that I must follow. I really NOT enjoy building this 2A3! can’t follow what I like ,such as rectifier tubes, AC 2.5V filaments, oil caps etc .

Single-Ended Triode (a.k.a SET) amplifier producing 1…1.5W into 8 ohm load. Uses two Russian deflector combo 6f5p tubes wired as triode.

Introduction

6F5P is Russian TV deflection triode/pentode combo tube, somewhat like ECL85/6GV8. If the pentode section is connected in triode mode, the tube provides a great playground for experiments with 2-stage SET designs on a “dime”. It is no joke, however. The plate currves are very linear, reminding 45 or 2A3, input capacitance is low and the plate resistance is < 800 Ohm.

The schematic is as studied with SPICE in [4], with the following changes:

(1) output bias resistors were adjusted for specific tubes, for best results, assuming the bias voltage ~30 V and current 35-40mA; (2) bypass caps are 47uF; (3) instead of 2k driver stage autobias resistors were used series connected 1.6k constant resistor and 2.2k pot in rheostat wiring – one leg to the 1.6 resistor, two other legs – shorted and connected to to the ground (4) driver stage load is 2 resistors in series, 47k and 30k, with a switch shorting the 30k resistor. (5) power stage bypass caps can be disconnected with a switch.

The power supply uses ultra fast diodes. RC filters a separate for each channel.

For those preferring tube rectifiers – it makes sense to organize the supply around the theme of paper and oil capacitors and filter chokes:

This supply needs a ~600V CT transformer with a winding for the heater of EZ80. EZ81 (6CA4) can be used too; it is more robust but needs more heater power. A good choise is 6X5. 5AR4/GZ34 can be used, too (different pinout filament supply: 5V, 3A), but it is an overshoot. In case of EZ80 and 81 heater voltage is 6.3V and may be taken from the same coil that feeds 6F5Ps, although it is preferable to use a separate coil. If the same coil is used, pins 3 and 4 must be disconnected. One 6X4 per channel can be used as well.
The chokes are 5…9H. An excellent choice is to use 2 TW3 output transformers as chokes.

This snubber circuit is optional. It can help to reduce hum or hiss that some tubes may produce due to heater-to-cathode leakage.

Parts

One of the aims of this project is reproduceability. All parts are easy to find and inexpensive. Except for large-value capacitors, all parts are NOS soviet production, still available in Russia or Ukraine. Thus, the output transformers, TWZSH (aka TW-3SH), are available in Ukraine at electronic markets and so are 6f5p tubes. The power transformer is TAH6, rated at 56W, offering 2 6.3V, .9A windings and a 520V, 70mA, center-tapped coil. The rectifier is solid state – partly for simplicity, partly because TAH6 does not have enough filament power to drive a tube rectifier. Rectification is full-wave, performed by SF36 – super fast (50ns reverse impulse) low noise 3A, 400VRMS diodes. The PS filter is Pi-type, formed by a common 100uF electrolytic capacitor, and a 1.2k, 94uF filter for each channel. All resistors are carbon, 1/2 to 2 W. Signal capacitors are paper BMT. Paper & oil K40-y9 is another choice. Electrolytics are Samsung, rated at 400V. Rubicon capacitors is another and better choice.

Construction

The chassis is constructed from aluminum L-shaped extrusions. The front extrusion is the the “front panel”, hosting the double volume pot and two “mood” controls (driver bias pots). These pots also allow to adjust balance by +-3db. Behind that the input jacks are mounted, close to the tubes. Behind them two output transformers are located at 90 degrees to the power transformer. The speaker posts are placed immediately after the output transformers.

The rear extrusion was drilled to host power cord socket, power switch and fuse holder. The rectifier diodes are soldered directly onto the power transformer lugs.

The power supply is visibly detached from the signal circuit. The space between the power transformer and the output transformers can be occupied by paper and oil bypass, ultrapath or B+ filter caps – these are all possible upgrades.

Another way to use this space is to host a small filament transformer, providing current sufficient for a tube rectifier, which can be located in the space by the power transformer side.

Soldering is almost universally point-to-point and directly onto the tubes, with the exception of the bias resistors and bypass caps of the power stage which are mounted on bias cap switch. The channels are wired symmetrically. Each channel has its own ground bus, short in length, connected to the input socket and not connected to the aluminum chassis.

Most parts are soldered on the bus in close proximity (< 20mm) of these points thus forming “star grounding”. The power supply is also symmetrical (common rectifier, split filters) and has its own ground bus. The power supply bus is connected to the chassis and via separate wires – the grounds of each channel.

Two wooden blocks are attached to the side L extrusions. Those shown on the picture are prototypes. The final version will sport more exotic wood, with access to the binding posts cut through the wood. Also, missing on the pics are the top cosmetic covers The output transformers are supposed to be covered with metal boxes, and so is the power transformer. the remaining top surface can be covered with small anodized aluminum panels, ceramic, copper, nickel alloy, etc.

Measurements

We loaded the amp with 8 ohm, 2W resistors and obtained distortion data and frequency sweeps. The results are below. All data were obtained with 0db level corresponding to the output signal amplitude at which the amplifier starts clipping, which is the amplitude that gives raise to lots of high order harmonics. Unless specified, the output level is -3 db under the level of clipping.
First, the spectrum data for various settings of the bias, load and bypass controls. Some combinations offer interesting possibilities for tone experiments.

These patters are mirrored by the other channel as well.

The frequency response plot into 8 ohm resistive load, with the bypass cap, at full power is 0..-3dB in 30Hz-20kHz region; with bypass cap disconnected: 0..-2dB in 30Hz-20kHz region.

More detailed data is in [5]

Variants

This design was repeated, with some variations, by me and by friends. Here are a few picks.

San Francisco prototype

Viktor Dodatko, Dnepropetrovsk:

Orest Baidan, Mountain View:

http://www.geocities.com/dmitrynizh

Copyright © Dmitry Nizhegorodov
dmitrynizh@hotmail.com

Designing a SE output transformer is the imposition of the transformer’s gods’ fury…. look at the formula’s for AC flux density vis-a-vis DC flux density. What we find is that if we want to keep the AC flux density low then (all other things equal) we would want a large number of primary turns. But, conversely, if we want the DC flux to be low (all other things being equal) we would want to decrease the number of primary turns.

In part of the design process for SE… you must calculate what your AC flux Consumption will be (at full power and lowest frequency of interest) and then subtract this amount from the maximum flux capacity for the specific core material that you are using. The amount left over tells you what the maximum DC flux could be…. if you wish to stay below the knee. And this, provided only as a one approach to design, assumes that your willing (or feel it smart) to use all of the flux available. And there are good arguments against this.

Another approach as explained in detail in Rueben Lee’s book on magnetics is that to maintain a high degree of inductive consistency… then your DC flux should be(I think it was… short of looking up the reference) 3 to 4 times the magnitude of your AC flux. And remember in a magnetic component with both an AC and a DC flux component the effective perm is a function of the absolute values of each as well as the relative (to each other)ratio of fluxes. As just one example… on one of our part’s (NLA in the DIY community btw) the DC flux was approximately 9850 gauss while the AC flux was approximately 3000 gauss. Rueben Lee demonstrates that if these conditions hold then the deviance in primary induction will be approximately 20% or so.

Push-pull coils can (and perhaps should be) mirror-imaged around a centerline (both geometric and electrical)… so that each half has the same leakage inductance and each half has the same capacity. And given the “equal but opposite” (at least in theory)nature of the two halves of a PP transformer the coil design and it’s geometry will be (or do I mean should be?) different than a SE output transformer design would be.

The SE output has one end of it’s winding at AC ground potential and the other end at “AC HIGH”…. for push-pull we have two ends at “AC HIGH” and (let us say informally) the center (the B+) is at AC ground. This is a fundamental difference…. if I am making this clear enough in my writing….. so if we follow the AC voltage gradients in a PP transformer we will find that they are “equal but opposite” in a mirror imaged way between the two halves of the coil.

SE outputs don’t offer us this “degree of symmetry” (or at least not on first blush)…. here one end of the winding is at AC ground and the other end at AC HIGH. So that as we traverse the coil, through each primary winding turn, the AC voltage potential is changing… at our interstices (junctions between primary and secondary windings) we then have certain voltage gradients… the gradients will be different across each interstice… and the capacities different. So if you have a PP output that was optimized for a different voltage gradient and a different voltage potential at the interstices… then using it as a SE device the unit will not achieve the intended optimization that the transformer designer had wished to achieve.

There is a way to do a SE output transformer as a quasi-symmetrical design…. wherein you design it as a PP coil geometry but then reverse wind one half of the primary and put it in parallel with the other half. Using this technique a design can achieve the same level of “symmetry” in terms of the capacities being equal about a coil geometric centerline….

POST 2

>Mike, please excuse my technical misunderstandings in advance. In your
>post, considering symmetric winding geometries for SE transformers, you
>described a technique of taking two primary windings, as used in a PP,
>reversing one winding, and connecting them in parallel. Question: are
>the two windings not bucking each other, if paralleled out of phase?
>If an ordinary PP transformer was reconfigured this way, would DC
>currents in the paralleled windings be nulled out?

Several of the responses to my post thought that the example used a bilfilar wound primary winding. Although perhaps it could, the illustration does not rest on it being bifilar wound. And the language I used was imprecise enough to lead to this confusion.

In the example…. what is happening is that you wind the whole primary twice… using three guages smaller than normal. for instance suppose the primary needed 2,000 turns of #30. then what you would do it wind 2.000 turns of #33 in P1 and P3 and another 2000 turns in P2 and P4. The example I had in mind derives from the Acrosound patent… and is a modification
of that patent in that P2 and P4 are reverse wound and then placed in parallel with P1 and P3.

Barry concerns about phasing… remember, and I always have to think this through and on some coils with tons more interleaving it becomes a real puzzle to keep the reverses straight…. when you reverse a winding the physical start becomes the electrical finish. And the reverse physical finish becomes an electrical start. so if you go through these reverses…. and their placement in parallel with the non reversed windings (say P1 an P3) then the electrical
polarities of the reversed windings… you’ll see (I hope) that they are not bucking….

>Question: Can an SE
>winding symmetrically configured give true symmetrical cancellation of
>stray reactances, as reflected to the secondary? Its not clear to me how
>symmetrical, out of phase components could be set up in an SE winding,
>in order to effect a true cancellation. I hope there’s a cogent question
>in here.

I used the term quasi-symmetrical moreso as an analogy to describe the physical configuration of the windings and their electrical polarities in response to a question about the capacitances within a coil and etc. In a single ended output…your right…. no differential phases (the push and the pull) are present…

So the strategy in illustrating what I called a quasi-symmetrical coil geometry in a SE output is to achieve a minimum number of different voltage gradients across the interstices of primary to secondary and to achieve a “sameness” of voltage potentials across these interstices. Wish you (and everyone who is interested in this topic) had a copy of the Acrosound patent and perhaps a diagram showing the winding sequences and etc. Then it would be easier to explain…

Just another note…. the illustrations of coil geometries should not be construed (on the basis of my choosing it as an illustration) as the method that I use on our own products. I actually use several different strategies for interleaving in our own products… but to fend off any misconceptions or speculations… I just wanted to put this disclaimer in here….

POST 3

I have been threatening Joe Roberts to do some more “Core Issues” in Sound Practices for quite some time and this is one of the articles that I would like to do.

Anyone have access to the old SP where I did an article on “how to pick a power transformer”? In it there was a discussion of current capability and how to evaluate or, minimally, establish a context for assessing qualitatively the numbers (the current ratings) which mfgr’s spec out.

Part of the problem in power trans is that the “current” number must be provided in some sort of context if it is to have meaning as a useful basis of discriminating btwn products. My article tried to show that to compare “raw numbers” from one company against “raw numbers” from another was often like comparing “apples to oranges” as opposed to “apples vs. apples”. My sense is that the same difficulty applies to especially the comparison of raw DC current numbers which manufacturers use for SE output transformers. But…. least I get a head of myself…

Steve, I don’t mind putting up some of my ideas and research on
“AC versus DC characteristics in SE and PP” as you stated above…. but I want to figure out how to structure the response so that it goes in some logical order. And I would be grateful for constructive comments and criticisms since I want to turn some of this stuff (that I have done over the past week or so) into an article for Sound Practices or Positive Feedback.

so give me a day or two or three or four or….. yep…. probably never get it done… so let me throw out a teaser and see if any of you RAT’s can “tell” me about this formula and what it means and what the variables are and what is unique about some of the variables… it is one of the building blocks of designing a SE output trans…. a focus and understanding of it…. leads, IMO, to clearer sight on some of the other issues that come after it.

energy factor equals L times (Isubdc) squared divided by core volume

core volume equals L times (Isubdc) squared divided by energy factor

L is primary inductance — unit measure is henries

Isub dc is the magnitude of dc current (and it is squared) — unit of measure is ADC

core volume is the volume of the magnetic core being used — unit of measure is cubic centimeters

Excerpt from Rec.Audio.Tubes newsgroup
By Mike LaFevre

This article is the second part of my examination of the output impedance (Zout) of triode single ended amplifiers using no negative feedback. Here I intent to examine carefully the effects at low frequencies, which I believe can help you understand the sometimes puzzling bass performance of SE amplifiers. With the help of the Thielle/Small theory, I will show how the high output impedance modifies the the system response.

Although the T/S theory is more familiar to speaker designers, I must use it to arrive at some conclusions relevant to the design of SE amplifiers. I will briefly describe the terms related to the theory as they appear but you may wish to consult the original papers (ref. 1,2,3,4) or a work such as Vance Dickason?s The Loudspeaker Design Cookbook (ref. 5) where these terms are much better presented.

I will include as examples only closed and vented box loudspeakers, because they are by far the most common but you can use the same methods with any other form of correctly modeled bass loading.

T/S PARAMETERS

The low-frequency design of most loudspeakers today is based on the justly famous Thielle/Small theory. At the low frequencies this theory uses an acoustical analogous circuit that allows you to calculate the actual acoustical output of the speaker driver in a box.

If you consider the Zout a simple resistive value, as I did in my first article, it is easy to use the T/S theory to analyze the behavior at low frequencies. Some available formulas take into account the output resistance of the amplifier and the cables. Thielle?s original paper addresses this point, and the work of Small (ref.2) show you very clearly what to do. If you know the basic Thielle/Small parameters of the woofer you?re using in the system (or better yet, if you measure them), you can use the equations (21) and (22) of ref.2 to calculate new parameters for your system. Look at these equations in a slightly different way: it is as if you had a “new” driver with a different Qes and Qts connected to a zero Zout amplifier. It will produce the same results.

	Qes’ =  Qes  (Rg + Re) 	     (1)                         Re

	Qts’ =     Qms Qes’ 	     (2)
                 Qms + Qes’

where,

Rg = output resistance of amplifier + cables
Re = DC resistance of the loudspeaker coil
Qes = Q of driver at resonance due to electrical resistances
Qms = Q of driver at resonance due to nonelectrical resistances
Qts = Q of driver at resonance due to all resistances
Qes’= electrical Q of “new” driver
Qts’= total Q of “new” driver

You should make the calculations for Rg equal to the Zout of the SE amplifier which is typically around 3 to 4 ohms, as I showed in my previous article. If you have significant resistance in the crossover and cables you should add it. You will then have two sets of parameters, each with a different Qts. Using these parameters you can now easily calculate the system response with any loudspeaker design software (or by working through the tables and formulas). Then you can compare the two response curves.

Zout VALUES

All this is not suited for a quick analysis of a loudspeaker low frequency response, because it requires a certain amount of work as well as access to the loudspeaker design parameters which may not be available. But I have done it, looking for any general patterns that might appear.

I used this approach with the loudspeaker referred in the past article which has a KEF woofer with the following parameters: fs = 22.5 Hz Qts = 0.274 Qes = 0.286 Qms = 6.547 Vas = 160 and Re = 6.8 Ohm . The crossover resistance is a high 1.6 Ohm . Table 1 shows the parameters change with different values of Zout, and the final results for a system composed of this driver and a 70l closed box that is lightly stuffed.

When the finite value of the primary inductance starts to show its effects, the output impedance begins to drop at the low frequencies. Which value of Zout should we use? There is no easy answer. You must consider what happens if you use a low frequency model of the output stage, which includes the primary inductance of the transformer.

Although the original T/S papers include a provision for taking high Zout values in account, they apear to consider only resistive values. I could find nothing about the effect of reactive Zout. The nearest I found was Benson?s work (ref.6) which looks at filters that are directly connected before the loudspeaker but he apparently restricts himself to the case of a zero output resistance amplifier.

Here I will use an acoustical analogous circuit of the whole system, composed of output stage, driver and box. I developed it by connecting the model for low frequencies of the SE output stage (fig.2d of the previous article) with the speaker?s electrical equivalent circuit (see reference1 or 2). The result is circuit a of fig.3.

I have used the T/S assumptions and methods to arrive at an acoustical circuit that permits calculation of the system?s low frequency acoustical response (circuit b in fig.3). In this circuit, the primary inductance appears as an acoustic compliance represented by a capacitor symbol. The relevant new parameters you need are rp, R1, Lp, R2, and n?, which are already defined in my GA 3/97 article and repeated in the figure.

The assumptions needed to make this circuit valid are that the output stage low frequency response is limited by the transformer?s primary inductance, that the iron core losses will be very small, and that the loudspeaker is working as a piston – plus all the other assumptions of the T/S theory.

FURTHER QUESTIONS

In fig.3 also appear the acoustical circuits of the output stage and loudspeaker in a closed box and in a vented box (circuits c and d). Several questions arise. Will the transformer plus loudspeaker be a third order system? If we put the speaker in a closed box will we also have a third order system ? Also will fourth order vented systems became fifth order systems?

The answer to all these questions is yes, but depending mostly on the value of Lp, you can ignore it. For suitably high Lp values the acoustical circuit will, with some use of circuit theory, revert to the original T/S circuits with the values of (rp+R1)/n? and R2 added to form the amplifier output resistance (Rg).

But what are the values of Lp? In attempting to answer this question I wrote a program to calculate the response of the complete system based on the models in fig.3. You can use this program for closed and vented boxes by supplying the driver parameters, the box data and the amplifier values of rp, R1, R2, n? and Lp.

It calculates the system response, computing not only the frequency response of the driver in a box with the variations introduced by the high (and complex) Zout but also accounting for the low frequency response of the output transformer.All the phase responses are accounted for, as well.

I have tested the model and the program with other speakers and transformers, and they have worked fairly well. With the program, I have simulated countless loudspeakers designs with many different SE output stage parameters. From these simulations, I have selected some that may give a good picture of the overall effect you can expect with a SE amplifier connected to a loudspeaker.

Figures 8 to 15 show these simulations of classical bass alignments. Each figure has four different responses corresponding to four different output stage parameters, as detailed in Table 3.

I believe the first important fact that emerged from all the simulations I have done was the confirmation that the values of Lp/n? found in normal SE transformers has little effect in the system response. The use of Lp/n? permits the comparation of the low frequency performance of transformers designed for different tubes with different impedance.

The specifications listed by some transformers manufactures show that the smallest value for Lp/n? for different type of transformers is usually around 35mH refered to the 8 ohm tap. This sort of value means that only if you have a loudspeaker with very good low frequency extension (something like f3 below 30 Hz) will you need to consider carefully the primary inductance values of traditionally designed SE transformers.

If you are using in your projects off the shelf SE output transformers with a gap, the assumption of resistive behavior for the Zout is sensible. This allows you to use the formulas (1) and (2) from the T/S theory to account for the high Zout, as long as the output transformer holds its specified value of Lp at the bias current you are using.

The second fact is an interesting effect: low values of Lp could help in some cases! The initial effect of a low inductance is to reduce the usual hump, bringing the loudspeaker response a little closer to the intended one. It is not just the reduction you would expect from the falling low frequency response of the amplifier measured into resistive load, because the reduction is greater for higher impedances, as in the resonance peak, where the high Zout introduces the hump.

This is why I have included in the simulations the unusually low Lp value of 3H. A low value of Lp may in some instances bring the speaker?s low frequency response a little closer to that expected by its designer. This is well ilustrated by Fig. 8.

MODIFYING LOW-FREQUENCY PERFORMANCE

High Zout amplifiers always modify the bass performance of the loudspeakers connected to them in a way that depends on the loudspeaker alignment. Although it is very difficult to describe what is a typical loudspeaker load at low frequencies, most of them are very far from a constant 8 ohms resistance, and usually have one or two impedance peaks.

Therefore, it may be possible to optimize the SE amplifier performance at low frequency by balancing the effects of the high Zout with those of the transformer?s primary inductance.For SE amplifiers used with normal loudspeakers there may be a possibility for optimizing the system performance at low frequencies by balancing the effects of the high Zout with the effects of the primary inductance of the transformer. This may work with at least some loudspeakers, and since it pointa to lower values of Lp, you may derive other related benefits such as better high frequency response, better power bandwidth, lower cost and smaller size.

Another possibility is to optimize the design of the whole system when designing both the amplifier and loudspeaker. I am exploring this route right now, and the effect of the variation of the Lp with power levels and frequency remains to be verified more carefully.

Although transformers operating with DC in the primary and with air gap exibhits much less Lp variation than those designed for push pull operation, the sensitivity of the system to these variations may be significant. With high relative values of Lp the difference between its effect and that of an infinite value is very small for frequencies above 20 Hz. But if you depend on its (low) value for an alignment you may need to control it more tightly than usual.

DO SE AMPLIFIERS HAVE POOR L.F. RESPONSE?

It is often taken for granted that tube amplifiers, specially single ended ones of the type I am discussing here, do not have good bass response. One of the main problems seems to be that loudspeakers not designed to work with the high output impedance of these amplifiers will develop a hump in the low frequency response and suffer degradation in their transient response.

But this picture can change if you design the loudspeaker to work with a high Zout. This can be done quite accurately using the same formulas (1) and (2) for the low frequency design and you may even get some additional benefits.

When you design loudspeakers, the efficiency, the box volume and the low frequency response are interrelated in such a way that for any two of these parameters that you fix the third one is already fixed for a given bass alignment. A high output impedance in the amplifier is one way to change this balance. If you model it correctly and use it to your advantage, it can give you more efficiency with the same low frequency response and box volume.

To explain more fully, the speaker efficiency is fixed whenever you select a driver. Choosing the box type and alignment defines the box size and the low frequency extension. With the same box type, alignment and size you can have an efficiency gain for the same low frequency response with amplifiers of high output impedance by choosing another driver.

MAGIC FORMULAS

This result also comes from formulas (1) and (2) as I will show in the following example. If you design a box for a loudspeaker to have a f3 of 40Hz in a vented B4 alignment you will need a woofer with Qts = 0.405. When you use this system with a high Zout amplifier, it will develop the already expected hump.

To correct this you could use another speaker driver with all the same parametersexcept for Qes and Qts. The Qes of the new driver will need to be the Qes of the old driver divided by (Zout+Re)/Re. When this new driver is used with the SE amplifier it will have again a Qts of 0.405.

Also from the T/S theory , the efficiency of a driver is given by equation (33) in ref.2:

KfsіVas ,
  Qes

where                    _10
K = constant = 9.64 x 10 for Vas in liters
fs = resonant frequency of driver
Vas = Equivalent Compliance Volume in liters

Since K, fs and Vas can remain unchanged, and the Qes of this new driver will be smaller, its reference efficiency will be greater in dB by 10log((Zout+Re)/Re). For typical values of 3.7 ohms for Zout and 6 ohms for Re you will have a gain of 2.1 dB for any type of bass alignment.

You can see this in fig.17, where I have simulated four different responses. Of course it may be difficult to find exactly this “new” driver but this gain will always exist and may appear translated into more efficiency, a smaller box volume, a lower f3 or any combination of the three.

You need not use single ended tube amplifiers to get high Zout. You could as well use solid state amplifiers with current feedback to get this same characteristic. It may be hard to have the same mids and highs, but this is another debate and as much as the high output impedance is inherent in a single ended amp design, you could use it in your favor. Surprisingly, it is one way to get a better compromise between efficiency, box size and low frequency response for any kind of bass alignment.

CONCLUSIONS

This and my former article show that the high Zout can produce important changes in the loudspeaker frequency response, and I believe that you cannot talk about meaningful listening tests for SE amplifiers without considering it. Probably the only thing that usually mask part of the high Zout effects is the room interaction with the loudspeaker, which may be responsible for even larger variations of the frequency response in the listening position.

I have shown how the bass alignment of typical loudspeakers is altered by a high Zout and also the influence of the finite value of the primary inductance of the output transformer. I did this last with the help of an acoustical model which includes the typical output stage of a SE amplifier. This article deals with the small signal behavior of the system composed of output stage, speaker and box, and although large signal considerations are also needed for the pratical use of the information contained here I believe that it can be used for tree basic purposes:

1. To help in the design of SE amplifiers more suitable to drive an existing normal loudspeaker. This could even include the use of much smaller than usual primary inductance to partially compensate the high Zout effects, and this in turn could lead to several benefits in other areas of transformer performance such as size, low frequency power response, high frequency extension and cost.

2. To help design loudspeakers that not only work correctly with common SE amplifiers, but can achieve an even better bass performance with improved relationship between size, efficiency and low frequency response.

3. To design the amplifier and loudspeaker work with one another as a system optimized to use all the above possibilities. If you can keep the bad effects of the high output impedance under control and design the loudspeaker to take advantage of it, you can actually benefit from it.

Finally, the high output impedance might have even more subtle effects on a system composed of amplifier, cable and loudspeaker which I have not been able to figure out, but I have tryed to detail and understand the more easily measurable effects it produces. Next time I intend to describe an amplifier and loudspeaker that were designed using some of the above ideas.

FIGURES

REFERENCES

ref. 1 – A. N. Thielle – Loudspeakers in vented boxes -Part 1 & 2 – Loudspeaker Anthology Vol.1 pg.181 & pg.192.

ref. 2 – R. H. Small – Direct Radiator Loudspeaker System Analysis – Loudspeaker Anthology Vol.1 pg.271.

ref. 3 – R. H. Small – Closed-Box Loudspeaker Systems – Part 1 & 2 – Loudspeaker Anthology Vol.1 pg.285 & pg.296.

ref. 4 – R. H. Small – Vented-Box Loudspeaker Systems – Part 1,2,3 & 4 – Loudspeaker Anthology Vol.1. pg.316, pg.326, pg.333 & pg.339.

ref. 5 – Vance Dickason – The Loudspeaker Design Cookbook – available from Old Colony Sound Lab, PO Box 243, Peterborough, NH 03458

ref. 6 – J. E. Benson – An Introduction to the Design of Filtered Loudspeaker Systems – Loudspeaker Anthology Vol.1 pg.365.

ACKNOWLEDGMENT

I would like to thank Mr. Manuel M. Pereira for all the discussions about output transformers and other related issues. I’m also indebted for his help and patience with all the changing requirements on the transformers he designed and made for this work.

(published in Glass Audio 6/97)

Eduardo B. E. de Lima

(c) Copyright 1997 Audio Amateur Corp.

Much has been said about triode SE amplifiers with no negative feedback – how good they sound, how lively they present a musical performance and so on. Two easily measurable characteristics that make this type of amplifiers sound different are the amount of distortion produced with its particular spectrum and the high output impedance. This raises the question of whether they sound so good and musical because of these characteristics or in spite of them.

The first time I listened to such an amplifier with no feedback was through a loudspeaker that apparently was not much affected by the high output impedance of these amplifiers. Since I was really impressed with the sound I heard, I decided to do some measurements trying to understand what was going on.

After making these measurements, I became more deeply involved with the subject, thinking about the design of a SE amplifier and the loudspeakers to use with it. Based on this work, my intention here is to examine the output impedance (Zout) and its role in the interface between SE amplifiers and the loudspeaker.

LISTENING TO AN SE AMPLIFIER

After finishing a 300B SE amplifier kit, I connected it to my system and was really impressed with what I heard. I had never heard the violins and the voices sound so natural Although I have other speakers and amplifiers around, the amplifier I was using in my system at that time was a Quad II with some parts upgraded and the speaker was one I had designed and built in the mid 80s, using a Kef bass unit, a Peerless midrange, and a JVC ribbon tweeter as drivers. This loudspeaker has only suffered minor changes and parts upgrade during the years. Although an old design, for me it still sounds good. It is not a high efficiency loudspeaker but with the 300B SE amplifier I have been able to listen at levels I’m used to without any problems.

After some days I decided to look at the whole system trying to find clues explaining the differences between the amplifiers. I measured the Zout of the amplifier and found it to be almost 4 ohms. I knew the loudspeaker very well, and from looking at the notes I made at the time I designed it, I remembered that it presented an almost purely resistive input impedance, with a magnitude between 6.8 and 10.1 ohms from 100Hz to 20Khz (fig. 1). It is a closed box design with about 87 dB/W/m average sensitivity and a highly damped resonance (Q=0.58) at 39 Hz.

In Fig.1,you can also see the loudspeaker frequency, valid from 300 hz up, measured with a low Zout transistor amplifier. With this speaker, driven by a SE amplifier with almost 4.0 ohm of Zout, the big difference I should be hearing because of this characteristic ought to be a stronger and less controlled bass caused by a less damped resonance. From mid bass up the expected differences in frequency response should all fall inside a 1.1dB total range (-0.5 dB / +0.6 dB referred to the nominal 8 ohm impedance).Even taking into account the drop of about 1dB in the high frequency response of the SE amplifier at 20 kHz, the differences I had heard were much greater than what might have been expected from these curves, and surprisingly, the most dramatic differences were in the mid range, with violins and vocals sounding remarkably open, The bass was stronger, as expected, but this was a minor difference compared to the midrange sound. I have to say again that all these thoughts and measurements occurred several days after the initial listening tests, when I noticed the differences in sound.

My reaction to this listening experience seems to be much like most of the described reactions to SE amplifiers. The fact that the loudspeaker I used was fairly insensitive to the high Zout made me conclude that I have to look elsewhere to find the reason for the amplifier’s good sound . This is why I started to look more closely at this characteristic of SE amplifiers in general. If we could lower the Zout without destroying the other aspects of the SE sound, we probably could use these amplifiers with more predictable results with many more loudspeakers.

TYPICAL SE OUTPUT IMPEDANCE

Fig. 2 shows a practical equivalent circuit for single ended output stage, with simplified circuits at mid, low and high frequencies (fig. 2c, d, e). These models are based on the description given by Terman (ref.1). I have simply changed the values to what they are as seen at the transformer’s secondary, instead of at the primary, as in the book. Also, to calculate Zout, RL (the load resistance) is omitted. These equivalent circuits can be improved, especially in the high frequencies, by considering the capacitance in the primary and secondary, but I believe they are suitable as they are for my purpose here After measuring the required parameters for one transformer I used the complete practical equivalent circuit (fig. 2b) with a circuit simulator to calculate the magnitude and phase of the Zout. I followed this with the actual measurements at some frequencies(Fig 3). Although the incremental primary inductance (LP) is not a very constant parameter, the measured results were close to the simulation values. Only at the highest frequencies there was there some appreciable difference – probably the result of not using the distributed capacitances in the equivalent circuit. Measuring other transformers has confirmed that the Fig 3 curves are fairly typical of the simulation of this kind of output transformer using these equivalent circuits.

The decrease in the Zout at low frequencies coincides with a rapid rolloff of the transformer’s low frequency response. One very interesting example appeared in the review of a commercial 300B SE amplifier with a 2.5 ohm Zout at 1Khz, 2.7 ohm at 20Khz and 0.76 ohm at 20Hz (ref. 2).

This low value of Zout at low frequencies may look like a good thing, but it probably happens only because the primary inductance of the output transformer is probably much lower than it should be for extended low frequency response.

This is confirmed by the frequency response plot shown in the review. The amplifier is 9dB down at 20Hz. The low value of the output impedance at 20Hz shows only that the limiting low frequency factor is the output transformer.

RESISTIVE Zout BEHAVIOR

SE Amplifiers with good output transformers should have the Zout with the behavior shown in the fig. 3 or better, with a more extended region of flat impedance. The phase plot shown means that except for the frequencies extremes, you can consider the Zout to be resistive. The region of flat impedance with resistive behavior corresponds to the region where the equivalent circuit of fig. 2c is valid.

I will use this extremely simplified fig. 2c model in all the following analysis. This is an important assumption that must be made to simplify the first visualization of the effects of the high Zout. I will try to show when you should expect this most simplified model to fail, making necessary the use of the other equivalent circuits .

As a first approximation in calculating the output impedance of a tube amplifier with no negative feedback, you can simply divide the plate resistance by n? (n being the ratio between the output transformer’s turns ratio between primary and secondary turns).

Several books state that as a practical rule, output triodes should be loaded with an impedance of about three times their plate resistance for maximum undistorted power output (although some theoretical calculations find a ratio of two to one). Following this rule, the Zout of SE amplifiers with any triode will always be about the same.

At first glance, its value should be one third of the nominal load impedance of the output tap of the transformer, or around 2.7 ohms for an 8 ohm tap. But we should look more carefully using the simplified model of fig. 2c and take in account the transformer’s resistance. As an example, you can do some quick calculations with a typical 300B SE amplifier.

According to the WE manual, the 300B has a 700 ohm plate resistance. Using an output transformer with a 2.5K primary reflected impedance you can calculate the output impedance. You need to measure or estimate its primary and secondary DC winding resistance and use it in the following formulas, which are based on formula (2) on page 206 of the Radiotron Designer’s Handbook.

Zout = (rp+R1) + R2 ( 1 )
          n?

where n? = Zp-R1 ( 2 )
           R2+RL

Zout = Output impedance
rp = Output tube plate resistance
RL = Nominal load resistance
R1 = DC Resistance of the primary of the output transformer
n = Ratio of primary to secondary turns (it can be measured directly)
R2 = DC Resistance of the secondary of the output transformer
Zp = Reflected primary impedance

If you estimate R1 to be 200 ohms and R2 to be 0.5 ohms, for Zp of 2500 ohms and a RL of 8 ohms. Using formula (2) yelds n?=270. Then, formula (1) gives the value of 3.8W for Zout.

RESULTS WITH MULTIPLE TRANSFORMERS

I measured several parameters of four output transformers for use in SE applications and calculated its Zout when used with 300B tubes. The results are summarized on table 1, along with the measured Zout. These Zout measures were taken at 1Khz. All the results are within 10% of the calculated values. Using a digital multimeter to measure low resistances like 0.6 ohms may introduce a sizable error. Although I have not further investigated the differences between measured and calculated Zout, I believe they can be explained mainly by this factor and by the fact that I didn’t account for the Rp of the particular 300B tube used for the measurements, relying instead on the 700 ohms value given by the manual. All the transformers were measured with the same 300B tube, and changing it during one measurement gave just slightly different results.

USING HIGHER TURN RATIOS

To see what happens using higher turn ratios, you can calculate the Zout for an hypothetical transformer the same way as before. Use a 5K primary and estimate R1 of 150 ohms and an R2 of 0.4 ohm. For the 8 ohm tap, n? will be 577. You end up with a Zout of 1.87 ohms. Since the characteristics of this hypothetical transformer are those of one that?s optimized for low Zout, I believe that 2 ohms should be about the lowest you could expect from a 300B SE amplifier with no feedback, but still with reasonable output power. Also for any other output triode – and even for tubes used in parallel – you probably cannot get much lower Zout, unless you deviate even more from the old rule about load impedance for maximum undistorted power output (”undistorted” here usually meaning 5% distortion). Everything I have said above applies to the nominal 8 ohm tap. The 4 ohms tap will usually have half of the Zout of the 8 ohm tap. Actually what I have described above for the 5K transformer is like using the 4 ohm tap in an amplifier with a 2K5 output transformer. Therefore when using speakers rated as 8 ohms, if you are willing to sacrifice the output power you can see how it will sound with a lower output impedance using this tap..It should be clear, however, that an amplifier designed to reflect a 5K impedance from 8 ohms should have details in the output transformer, the biasing and the power supply such that it would be optimized for the right impedance, extracting more power and being a better overall solution if done right. And, of course, you could have still a lower output impedance using its 4 ohm tapThe issue as to which output tap to use and of specifying them nominal loudspeaker load is not simple. With tube amplifiers using the tap that better reflects the average loudspeaker impedance will usually result in more output power. To get lower output impedance, the 4 ohm (or lower) tap is always the better option, regardless of the loudspeaker impedance. It’s difficult to balance not only this points, but also the change in frequency-response extension and the different amounts of distortion for different power levels. This is why listening to all tap options is sometimes the only way to decide which one is better for a loudspeaker with a particular amplifier.

EFFECTS OF HIGH OUTPUT IMPEDANCE

As you have seen, the lower limit on the value of the Zout of the SE amplifiers is high makes it important to consider its effects on the system response. Driving a loudspeaker with a high Zout seems to be a bad idea. In general, loudspeakers are designed with the assumption that they will be used with amplifiers that closely resemble a perfect voltage source, which means Zout close to zero. Although the resistance of contacts and cables usually make reducing the Zout of an amplifier below same point like 0.1 ohm a useless effort, in the SE amplifiers, you are dealing with values around 4 ohms. Such a Zout value interacting with typical loudspeakers input impedance will produce a deviation from the loudspeaker intended frequency response.

From the midbass up you can calculate the probable maximum deviation from the original frequency response if we have the amplifier’s Zout value and an input impedance curve of the loudspeaker. You can more or less tell by looking at the imput impedance curve what will happen with the loudspeaker response as a kind of “modulation” of the frequency response curve.

In trying to visualize it you should keep in mind that because of the difference in the effect of the peaks and dips of the input impedance of the loudspeaker and also because of the complex nature of the impedances, the peaks will appear broader and the dips sharper in the frequency response curve than in the original input impedance curve.

THE RESPONSE CHANGE

To calculate the change in the response we should use the following formula:

D(f) = 20log __Zin (f)__ (3)
              Zin(f)+Zout

where, D(f) = deviation in dB at frequency f
        Zin(f) = input impedance of the loudspeaker at frequency f

For any resistive value of Zout that is not zero, this formula will always give a negative value which corresponds to a loss. But you are looking for the difference between the loss at the maximum Zout and the minimum, which will give the range of deviation from the intended frequency response curve of the loudspeaker.

Ideally we should use the complex values of Zout and Zin but the output transformer model we are using implies a resistive behavior for Zout. Also the peaks and dips in the magnitude of the loudspeaker input impedance usually correspond to points where its phase angle is 0 or very close to it. Because of these facts using just the magnitude of the input impedance and assuming a resistive behavior, will give you the most probable maximum deviation from the intended frequency response of the loudspeaker.

I went through all the 1995 issues of Stereophile and looked at 31tested loudspeakers whose input impedance curves were published. The maximum values for input impedance had to be estimated since the curves are limited to a maximum of 20 ohms. Considering only the frequencies well above the low frequency resonance of the system, the average loudspeaker had an input impedance varying from 4.3 ohms to 15.0 ohms. with the phase going from -28 to +29 degrees. Very seldom a loudspeaker had phase angle of more than 45 degrees or less than -45 degrees (only 2 cases, being the worst a 55 degrees angle). This is fortunate because the error introduced assuming resistive behavior will be small.

MAXIMUM DEVIATION RANGE

Based on the above assumptions, tables 2 and 3 give the estimated range of maximum deviation from the intended frequency response of the loudspeaker for different values of input impedance and of amplifier’s Zout. Table 2 is for loudspeakers which would be rated as 8 ohms and table 3 as a table for loudspeakers rated as 4 ohms. I chose the range of input impedance so that, as much as possible, this nominal impedance would split the range in half. Hence, in table 2 a 5.00 dB range corresponds to ± 2.50 dB referred to the level corresponding to the 8 ohms nominal impedance.

For these tables the error introduced by considering Zin and Zout resistive is always less and most of the time much less than 0.7 dB for any of the values in the tables for a difference in phase angle below 45 degrees. But loudspeakers can have all kinds of input impedance curves and, although it doesn’t happen frequently , if the phase difference gets higher you could have larger errors. The intent of these tables is to show how the Zout affects the system response, allowing us to look at the broad pattern. I did not intend to exhaust all the possibilities.

HIGH FREQUENCY EFFECTS

Now look at what will happen at the very high frequencies. Should the equivalent circuit of fig.2e be used here? With dynamic loudspeakers the voice coil inductance of the tweeters will usually have an effect on the phase of the loudspeaker input impedance that corresponds to an inductive behavior. This is similar to the effect of the transformer’s leakage inductance on the phase of the amplifier Zout. This keeps the difference in phase angle small and this reduces the need to use the model of fig.2e for evaluating the effects of the high Zout at high frequencies with dynamic loudspeakers. As long as the change in the magnitude of the Zout at high frequencies is reasonable we should have no surprises here and can still consider the Zout resistive when using formula (3). This also holds even when impedance compensating networks (Zobels) are used. Only with output transformers with very limited high frequency response driving loudspeakers with capacitive behavior (like piezo tweeters) we should need to take a closer look at these frequencies, because of the probably high phase difference.

All the considerations above are valid through the frequency region that’s far from the resonant frequency of the system’s woofer. Near the system’s resonant frequency you can still try to use the same approach as a rough guess but it may give wrong results. You cannot simply use the idea of “modulation” of the frequency response curve by the loudspeaker input impedance curve without further care.

Not only does the falling frequency response make it difficult to visualize the effect of the output impedance, but in vented systems the change in the alignment produced by the high Zout will also change both the value and the frequency of the impedance peaks. There are also other factors – including the primary inductance of the output transformer – that may affect the frequency response of the system, making the use of the equivalent circuit of fig. 2d necessary to understand what happens. But this deserves an article of its own.

REFERENCES

ref. 1 – Terman – Electronic and Radio Engineering (1955) – pg. 341

ref. 2 – Stereophile – March/1995 – pg. 120

ref. 3 – Reich – Theory and Application of Electron Tubes (1944) – pg. 228 – 232

ref. 4 – Dammers, Haantjes & Van Suchtelen – Application of the Electronic Valve in Radio Receivers and Amplifiers. Vol II (1951) – pg. 97-101

ref. 5 – Langham – High Fidelity Techniques (1950) – pg. 38 – 41

(published in Glass Audio 3/97)

? Copyright 1997 Audio Amateur Corp.