ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Choosing an OT: Theory Q's

8/13/1999 12:58 PM
Mook	Choosing an OT: Theory Q's A pair of 6V6s in fixed bias want to see an impedance load of about 4K on the OT primary. However, the same pair of 6V6s in cathode bias want to "see" about 8K on the OT primary. What if you have an OT where the primary has a higher or lower impedance than the "nominal" ratings. That is... 1) What if you used an 8K OT primary for FIXED bias? Will you get more "crunch", will it sound "darker"? 2) What if you used a 4K OT primary for CATHODE bias? What will it sound like? In general, what are the pros and cons for mis-matching OT primary impedances? Mook

8/13/1999 12:58 PM

Mook

Choosing an OT: Theory Q's

A pair of 6V6s in fixed bias want to see an impedance

load of about 4K on the OT primary.

However, the same pair of 6V6s in cathode bias want to

"see" about 8K on the OT primary.

What if you have an OT where the primary has a higher or

lower impedance than the "nominal" ratings. That is...

1) What if you used an 8K OT primary for FIXED bias? Will you get

more "crunch", will it sound "darker"?

2) What if you used a 4K OT primary for CATHODE bias? What will it sound like?

In general, what are the pros and cons for mis-matching OT primary impedances?

Mook

8/13/1999 1:10 PM
R.G.	Try it and see. Mismatching the nominal primary impedance of an OT is exactly the same as mismatching the secondary impedance with speakers. OT's have no "impedance" of their own, only a reflection ratio. You can cause the plates to see a 2:1 change at their plates by changing the speakers 2:1. Be aware that putting speakers in series with lower the damping factor and loosen bass all on its own, though, so you really should use single speakers of different nominal impedances. That makes it harder to hear tonal differences of just the OT change. Hmmm. Maybe the thing to do is to set up an amp with a quad of 6V6's and plug/unplug two of them, then readjust the volume level. That should have the same effect.

8/13/1999 1:10 PM

R.G.

Try it and see. Mismatching the nominal primary impedance of an OT is exactly the same as mismatching the secondary impedance with speakers. OT's have no "impedance" of their own, only a reflection ratio. You can cause the plates to see a 2:1 change at their plates by changing the speakers 2:1. Be aware that putting speakers in series with lower the damping factor and loosen bass all on its own, though, so you really should use single speakers of different nominal impedances. That makes it harder to hear tonal differences of just the OT change.

Hmmm. Maybe the thing to do is to set up an amp with a quad of 6V6's and plug/unplug two of them, then readjust the volume level. That should have the same effect.

8/13/1999 1:15 PM
SpeedRacer	To RG's reply I'd only add that the idea of wanting to "see" a specific imp load is only a design spec that's dictated by the tubes ratings (plate diss) and the level of distrotion we'll accept in the output.. there's no one spec that is the right one for a given tube. The "rightness" is purely a function of your design parameters, eg you sacrifice output for low distortion, etc. Like he said, try stuff out! I've run 6V6's from 4K to 8K (biased both ways) and there's no hard and fast rule..

8/13/1999 1:15 PM

SpeedRacer

To RG's reply I'd only add that the idea of wanting to "see" a specific imp load is only a design spec that's dictated by the tubes ratings (plate diss) and the level of distrotion we'll accept in the output.. there's no one spec that is the right one for a given tube. The "rightness" is purely a function of your design parameters, eg you sacrifice output for low distortion, etc.

Like he said, try stuff out! I've run 6V6's from 4K to 8K (biased both ways) and there's no hard and fast rule..

8/13/1999 4:31 PM
Doc	The 8k p-p load is normally seen in tube manuals as a suitable load for 6V6s run in cathode bias, but I don't know where you found the "4k p-p in fixed bias" recomendation for those tubes. Anyway, it's tough to be able to state some hard and fast rules in choosing the "optimum" load resistance for a certain tube type, just based on one parameter like how it's biased. Some of the parameters that are juggled to select a load resistance are: maximum plate supply voltage for the circuit, max plate current, tolerated distortion level, maximum signal power output, rated plate dissipation, class of operation, and of course tube type. In choosing an optimum load resistance for the performance characteristics desired, it's an iterative process, involving graphs & equations. In triode circuits, the maximum power transfer occurs when the load is twice the ac plate resistance. In pentode (or beam tube) circuits, higher loads will usually extract more power than lower value loads, based on the product of the change in plate current and voltage across the load. But there's a safe limit with respect to max plate dissipation. For instance, in the typical 50w 6L6GC class AB2 circuit, such as seen in fender amps, raising the load resistance above the optimum design value will increase distortion and may take the tube out of the safe operating area for plate dissipation. (This stuff doesn't matter at lower than maximum power levels, where the tube's max dissipation isn't approached.) Class AB2 amps are usually set up for the maximum power capability of a particular tube, so you'll see that a relatively low value of load resistance will be specified. Many circuits for lower power tubes must be run at lower anode supply voltages, and usually need a higher load resistance to extract safe maximum power at that operating point. Your cathode biased 6V6 circuit would be a good example. Anyone interested in seeing what's involved in choosing a power amp load should take a look at the front section of an RCA tube manual. One I have in front of me now is an RC-23, but most of the info there is similar or repeated in other versions (other printing dates). You can obtain a reprint of the RC-19 version from various sources, but AES (tubesandmore.com) has it for $12.95. It's ok to experiment, and pick the load that sounds right for the application. Erring on the low side seems to be safe. Erring on the high side, in maxed out circuits, could be troublesome. Fortunately, high power transformers with high impedance windings are very costly, and therefore not usually available. So unless you try to use, say, something like a 50w p-p 6L6 (6.6k primary) output transformer on a maxed out class AB 6V6 or 6BQ5 circuit with an 8ohm speaker connected to the 4ohm tap, and running it flat out, it should be fairly safe (IMHO) to experiment varying load resistance up or down within reasonable limits, to arrive at the tone you're after.

8/13/1999 4:31 PM

Doc

The 8k p-p load is normally seen in tube manuals as a suitable load for 6V6s run in cathode bias, but I don't know where you found the "4k p-p in fixed bias" recomendation for those tubes.

Anyway, it's tough to be able to state some hard and fast rules in choosing the "optimum" load resistance for a certain tube type, just based on one parameter like how it's biased.

Some of the parameters that are juggled to select a load resistance are: maximum plate supply voltage for the circuit, max plate current, tolerated distortion level, maximum signal power output, rated plate dissipation, class of operation, and of course tube type. In choosing an optimum load resistance for the performance characteristics desired, it's an iterative process, involving graphs & equations.

In triode circuits, the maximum power transfer occurs when the load is twice the ac plate resistance. In pentode (or beam tube) circuits, higher loads will usually extract more power than lower value loads, based on the product of the change in plate current and voltage across the load. But there's a safe limit with respect to max plate dissipation. For instance, in the typical 50w 6L6GC class AB2 circuit, such as seen in fender amps, raising the load resistance above the optimum design value will increase distortion and may take the tube out of the safe operating area for plate dissipation. (This stuff doesn't matter at lower than maximum power levels, where the tube's max dissipation isn't approached.) Class AB2 amps are usually set up for the maximum power capability of a particular tube, so you'll see that a relatively low value of load resistance will be specified. Many circuits for lower power tubes must be run at lower anode supply voltages, and usually need a higher load resistance to extract safe maximum power at that operating point. Your cathode biased 6V6 circuit would be a good example.

Anyone interested in seeing what's involved in choosing a power amp load should take a look at the front section of an RCA tube manual. One I have in front of me now is an RC-23, but most of the info there is similar or repeated in other versions (other printing dates). You can obtain a reprint of the RC-19 version from various sources, but AES (tubesandmore.com) has it for $12.95.

It's ok to experiment, and pick the load that sounds right for the application. Erring on the low side seems to be safe. Erring on the high side, in maxed out circuits, could be troublesome. Fortunately, high power transformers with high impedance windings are very costly, and therefore not usually available. So unless you try to use, say, something like a 50w p-p 6L6 (6.6k primary) output transformer on a maxed out class AB 6V6 or 6BQ5 circuit with an 8ohm speaker connected to the 4ohm tap, and running it flat out, it should be fairly safe (IMHO) to experiment varying load resistance up or down within reasonable limits, to arrive at the tone you're after.

8/14/1999 2:03 PM
Mark Knapp	Doc, This is a very good, simple explanation of a complicated topic. I am saving it for future reference. Great job! Mark Knapp

8/16/1999 8:09 PM

Ken

Doc said:

quote:
"In triode circuits, the maximum power transfer occurs when the load is twice the ac plate resistance."

I don't mean to be nit-picky, but this isn't perfectly true. Maximum power TRANSFER actually occurs when the load impedance is conjugate to the generator impedance. Assuming pure resistances for simplicity's sake, that means max power transfer occurs when Rl=Rp.

Now, whether or not that makes max power OUTPUT depends on the ability of the driver circuit to increase it's voltage swing output. Assuming that this is true, then max power OUTPUT for triodes occurs at Rl=2Rp, as Doc points out.

Note that this is NOT due to more efficient transfer of power, but rather because the load line is tilted in the counterclockwise direction, enabling a more negative excursion on the grid before the tube pinches off. Likewise, that means that there is also a higher permissible positive excursion possible to retain symmetry about Vg at idle.

There is more power output because we are GENERATING more power, not because the transfer is more efficient.

Nit mode off ;-)

Page 1 of 1