• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Global Feedback Source Point

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
In looking at a heap of PP circuits I have noticed that they invariably take the global feedback from the highest impedance tap of the secondary of the output transformer. I this something that is one of those "default standards" i.e. we do it that way because everyone else does - or is there a good technical reson for it?

I can see that a higher impedance tap will have higher (source)voltage and therefore use a higher value feedback resistor. Hence less shunt of signal/bias current at the point where the feedback is applied BUT was wondering if there was something in transformer theory which might account for this choice.

I've always used the tap to which my speakers connect (as I've used the same nominally 6 Ohm speakers for 10 years this is always the 4 Ohm tap) and wonder if I might have been missing a trick?

Any ideas??
Joined 2004
I don't think there's anything particularly in favour of always using the highest impedance tap and, in any case, it's not always done that way.

For example, in the Mullard 5-10 and 5-20 a table of possible values for FB resistors was given, depending on the speaker impedance. This was because Mullard didn't manufacture the amplifiers, so they didn't know what OPTs and speakers people would use and, therefore, couldn't assume what secondary taps would be available.

An amp manufacturer, on the other hand, knew what OPT would be used and so took the easier approach of standardizing on one particular tap for NFB.
I've changed the feedback from 4ohm to the 8ohm on a push pull amp and there is a difference when watching on the scope with 10k square waves on the 4ohm tap 8ohm resistor and giving a 1uf cap hit across it to check for stability, and the ringing settling can better if the feedback on one or the other tap, can'nt remember which one you'll have to try for yourself if you have a scope and a signal generator.
Better still if the amp and output tranny is up to it do'nt have any feedback, I prefer the sound this way, but the output transformer has to be good if the bass is to stay tight and the highs can get to 20k no less than -2db, the amp becomes as stable as a rock also without feedback and the gain goes up, this way you can say goodby to your pre amp and go passive.

Cheers George
This is a difficult question to answer, as it depends on the construction of the output transformer. Circuit diagrams, in order to simplify, often show connections for several loudspeaker impedances as "taps". These may be an internal re-arranging of windings or really taps on a larger secondary winding.

I would be very cautious of the latter set-up. The output transformer is an instrinsic part of the feedback stability, and to connect a load to part of a properly designed OPT's secondary will change the influence and balance of leakage reactance (the "degree" of coupling of primary to secondary at high frequencies for the uninitiated). Unfortunately there is little one can do to a commercial transformer's intestines. Members would know that two 4 ohm windings in series give 16 ohm, so there no unbalance would occur. (For 4 ohms one uses both windings in parallel, not just one winding or section.)

With very great caution I would suggest that it is probably better to take the global feedback off the highest (16 ohm?) terminal when using a lower impedance, except as said where 2 sections come out and can be used series/parallel, with different feedback resistor values. The 8 ohm tap is the problem - which will probably be what most people want! When using a commercial OPT it is almost essential to have access to a square wave generator and scope to check the final result as one member suggested. And at the risk of under-estimating members' knowledge, also remember that that stability is far better set with your loudspeaker as load instead of a resistor - the loudspeaker impedance is bound to be non-resistive at high frequencies and higher than nominal. (It only needs a tolerable volume setting.) If you are in the fortunate position to have an OPT wound, then the answer is obvious - go for 8 ohm only and forget the rest, if that is your required load impedance.

Hope this helps,
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.