Some of us do not know the particulars of a Pilot SA232.
Do you have an easily Readable schematic?
At least some things might be determined from the schematic: B+ voltage; output tube bias voltage / or cathode current; output tube types; output tube mode . . . UL, Pentode/Beam Power, or Triode / Triode wired outputs; global negative feedback or not; Williamson circuit or not; etc.
And, with those factors, it might be easier to determine a suitable output transformer.
Nobody is going to have all of the specifications of the output transformers, some important factors will be missing.
Important things like primary inductance; maximum DC un-balanced current; leakage inductance from primary to secondary;
primary distributed capacitance, etc.
Without the exact output transformers, if there is global negative feedback, you probably will have to adjust the Rs and Cs in the global negative feedback circuit.
When is a Clone not a Clone? When you do not have the original DNA of the output transformers.
Sounds like fun!
Have fun researching, building, and listening.
Do you have an easily Readable schematic?
At least some things might be determined from the schematic: B+ voltage; output tube bias voltage / or cathode current; output tube types; output tube mode . . . UL, Pentode/Beam Power, or Triode / Triode wired outputs; global negative feedback or not; Williamson circuit or not; etc.
And, with those factors, it might be easier to determine a suitable output transformer.
Nobody is going to have all of the specifications of the output transformers, some important factors will be missing.
Important things like primary inductance; maximum DC un-balanced current; leakage inductance from primary to secondary;
primary distributed capacitance, etc.
Without the exact output transformers, if there is global negative feedback, you probably will have to adjust the Rs and Cs in the global negative feedback circuit.
When is a Clone not a Clone? When you do not have the original DNA of the output transformers.
Sounds like fun!
Have fun researching, building, and listening.
That seems like a fairly standard EL84 amplifier, with the push pull output stage in pentode mode. A reasonable quality 8K plate to plate primary should work well.
The 12AX7 paraphase inverter triodes have gains of about 23, and their cathodes are un-bypassed (no low frequency pole there, since there are no bypass caps).
The global negative feedback only wraps around the output transformer, 12AX7, the 2 coupling caps; and with the inconsequential common cathode EL84 self bias resistor bypass cap.
Low frequency poles of the coupling caps, and the output transformer primary inductance are the only significant low frequency poles.
For the most part, the EL84 cathode bypass cap does not count, becasue it sees the opposite phases from the 2 EL84 cathodes.
On the other hand, a Williamson circuit would have at least one more low frequency pole, which is often problematic (lucky you, no Williamson circuit).
The 12AX7 gains at 23
The EL84 gains are perhaps about 40.
An 8k to 8 Ohm voltage reduction is about 32.
(23 x 40) / 32 = voltage gain of about 29 from 12AX7 grid to 8 Ohm output tap, which is [also] about 29 dB.
29 dB gain inside the global negative feedback loop.
With perhaps 12dB of global negative feedback, that should make the negative feedback fairly easy to compensate.
Have Fun!
The 12AX7 paraphase inverter triodes have gains of about 23, and their cathodes are un-bypassed (no low frequency pole there, since there are no bypass caps).
The global negative feedback only wraps around the output transformer, 12AX7, the 2 coupling caps; and with the inconsequential common cathode EL84 self bias resistor bypass cap.
Low frequency poles of the coupling caps, and the output transformer primary inductance are the only significant low frequency poles.
For the most part, the EL84 cathode bypass cap does not count, becasue it sees the opposite phases from the 2 EL84 cathodes.
On the other hand, a Williamson circuit would have at least one more low frequency pole, which is often problematic (lucky you, no Williamson circuit).
The 12AX7 gains at 23
The EL84 gains are perhaps about 40.
An 8k to 8 Ohm voltage reduction is about 32.
(23 x 40) / 32 = voltage gain of about 29 from 12AX7 grid to 8 Ohm output tap, which is [also] about 29 dB.
29 dB gain inside the global negative feedback loop.
With perhaps 12dB of global negative feedback, that should make the negative feedback fairly easy to compensate.
Have Fun!
Last edited:
One moderately odd aspect of the transformer specs on the SAM's shown above is the primary impedance at 6k which is a bit out of the ordinary for a PP EL84 amp.
Many Thanks for the quick response with the information I was looking for! Now to search current options. A quick look at the Lundahl list on Erhard Audio's site shows there is an OPT in their line w/ a 6K and a PP rating of 24W - LL2766. https://www.erhard-audio.com/OutputTransformers.html. And yes, there are likely more 8Ks options out there - the OPTs in mt ST-35 are 8K
Looking at 6k versus 8k transformer primaries . . .
A speaker that is rated '8 Ohms', often has a minimum impedance of 6 Ohms.
With that loudspeaker on the 8 Ohm tap of an 8k transformer, the 8k primary will reflect 6k Ohms minimum impedance.
Do you see how that works?
Design an amplifier to work with a precision non-inductive load resistor.
Then . . .
Design an amplifier to work with a loudspeaker. It has minimum impedance, rated impedance, and maximum impedance, which is completely different than a load resistor.
A speaker that is rated '8 Ohms', often has a minimum impedance of 6 Ohms.
With that loudspeaker on the 8 Ohm tap of an 8k transformer, the 8k primary will reflect 6k Ohms minimum impedance.
Do you see how that works?
Design an amplifier to work with a precision non-inductive load resistor.
Then . . .
Design an amplifier to work with a loudspeaker. It has minimum impedance, rated impedance, and maximum impedance, which is completely different than a load resistor.
Well I'm plowing through the sticky thread of the Tubes section looking for resources on how to understand the R&C adjustments that would need to be made for an alternate OPT - say the Lundahl w/ 6K primary - any suggestions on an education path there would be appreciated. Unless I can learn the how-to's required, I'll just set this aside and be content with kits or well documented modification articles I guess. Thanks
You could start with a 50pF cap, 100pF cap, then a 250pF cap.
With a decent output transformer, that circuit might work without any capacitor in that negative feedback network.
The optimal negative feedback adjustment is dependent on the circuit topology; and is dependent on 6k versus 8k,
and it is dependent on the gain/phase of the output transformer.
It also is dependent on the loudspeaker that loads the amplifier.
A lot of amplifiers are working and being listened to and enjoyed, even though their negative feedback adjustment is not optimal.
If it did not work properly in pentode mode, then you should be able to at least listen to it and enjoy it in triode wired mode (lower power),
with, or without, global negative feedback; until someone could help you sort out the problem to make it work in pentode mode.
Triode wired mode only requires a pair of 100 Ohm resistors.
Disconnect the screen from B+, and connect the screen through a 100 Ohm resistor to the same tube's plate.
And it is Easy to change back when you want to.
With a decent output transformer, that circuit might work without any capacitor in that negative feedback network.
The optimal negative feedback adjustment is dependent on the circuit topology; and is dependent on 6k versus 8k,
and it is dependent on the gain/phase of the output transformer.
It also is dependent on the loudspeaker that loads the amplifier.
A lot of amplifiers are working and being listened to and enjoyed, even though their negative feedback adjustment is not optimal.
If it did not work properly in pentode mode, then you should be able to at least listen to it and enjoy it in triode wired mode (lower power),
with, or without, global negative feedback; until someone could help you sort out the problem to make it work in pentode mode.
Triode wired mode only requires a pair of 100 Ohm resistors.
Disconnect the screen from B+, and connect the screen through a 100 Ohm resistor to the same tube's plate.
And it is Easy to change back when you want to.
Last edited: