Basic question? Should you twist the blue and brown plate wiring and also the ultralinear wiring. So brown and blue together and then the brown-blue/whites together?? Not sure see pic
There is no electrical need to twist any wires together. However if you wanted to do so for appearance or neatness, then loosely twist the blue & blue/white. And then the brown & brown/white. Do not combine the opposite phases.
OT primary wires carry several hundred volts of signal voltage at currents of several 100mAs.
This makes them the strongest radiators of electric and magnetic signal fields in a tube amp.
To reduce radiation and avoid unwanted coupling, opposite phase wires should be twisted.
This effectively lowers emission by self-shielding.
Also they should be as short as possible and routed close to the chassis.
This makes them the strongest radiators of electric and magnetic signal fields in a tube amp.
To reduce radiation and avoid unwanted coupling, opposite phase wires should be twisted.
This effectively lowers emission by self-shielding.
Also they should be as short as possible and routed close to the chassis.
Last edited:
I was thinking along the lines of what Rikaro is saying. But I have to say I haven't run across any info pointing this out for sure.
When I first read your post, it sounded to me that you wanted to twist/connect the wires! Wow, I thought, this guy is going to be in trouble.
Twisting the out of phase wires sounds like a good idea that I hadn't considered before.
Twisting the out of phase wires sounds like a good idea that I hadn't considered before.
As Ricaro stated, to reduce radiation and avoid unwanted coupling. I thought that it might apply to reducing common mode noise. though.
You don't want a power tube plate signal to couple into a grid circuit and cause HF instability by positive feedback.
That's one of those parasitic effects you won't see with circuit simulation.
That's one of those parasitic effects you won't see with circuit simulation.
OK so Rikaro your still arguing for it correct? the two out of phase plate wires should be twisted and the two out of phase screen wires would be twisted. how come I haven"t heard this discussed before? Looks like it would be standard on P-P output stages? So what about this scenario. The plate currents would be equal and opposite at least when the output stage is in Class A but once in transitions to Class B the currents in the plates would not be equal during some portion of the waveform. In that situation would you have increased chances of crosstalk between the twisted pairs?
By transformer action the ends of a PP OT primary must have opposite phase signal voltage wrt CT (which is an AC ground) no matter if class A or B.
Twisting primary wires helps to avoid HF oscillation and is good practice with PP and SE amps.
Also twisting OT secondary tap wires with the ground wire can help.
My recommendations are based on physics and professional experience with EMI and amp instability.
Any points against twisting?
Twisting primary wires helps to avoid HF oscillation and is good practice with PP and SE amps.
Also twisting OT secondary tap wires with the ground wire can help.
This question I can't answer.
My recommendations are based on physics and professional experience with EMI and amp instability.
Any points against twisting?
Last edited:
Thank you . I follow your explanation. I will twist unless someone has a better argument. Just for the sake of argument when one transistor is in conduction in B isnt the other in cutoff and current is markedly different in the phases? The transformer is still summing them but one is off??
In class B one tube (or transistor) conducts while the other is in cutoff.
But that's just current, not OT voltage.
I have seen many an amp undersides (50+) from the golden age of tubes and never were Output transformer leads twisted.....
OT voltage is not the point here. When in class A the current flowing in the opposite phases is opposite and equal and field strength around wires cancel. In B one phase is relatively high and the other low. That means field strength in one phase might induce noise into other lower phase and hence crosstalk. I just trying to think through this.
Take my advice or find out yourself.
Not every amp is close to instability (e.g. when open loop gain is low and OT wires are short), but quite a few are.
In any case I recommend to do square wave stability testing with speakers connected.
Not every amp is close to instability (e.g. when open loop gain is low and OT wires are short), but quite a few are.
In any case I recommend to do square wave stability testing with speakers connected.
Voltages on both sides of the primary are always of equal amplitude.
That's how a transformer works.
It's voltage which causes electric field emission and capacitively couples to other circuit points.
Currents cause magnetic fields and are not symmetrical in class B as you said, so twisting only helps at low signal class A.
Last edited:
