Transcendar Output Transformers ST70

[Wharf_Creek]

I have just completed a scratch-build of an ST70 clone using the Transcendar Chassis, the SDS Filter Board, a Dynakit Parts Driver Board, a stock PT, and two new Transcendar OTs...which I believe is the source of my problem. I've been trouble-shooting this amp for nearly a day now, and I still don't have the answer. I'm getting a horrible feedback thru the amp....it occurs to the point where it's almost audible thru my dummy loads. If I pull the driver tubes, the amp bias's up just fine. But, if I install the driver tubes, as they warm up, the bias starts creeping upward, and the squeal begins. I've been told this may be a negative feedback problem, or possible a "Phase" problem with the OTs. I've tried switching the feedback wire to the other tubes on each side, and even reversed the feeds from the driver board. These OTs don't come with leads that are marked the same as Dyna Original. Thus the possibility of having the pairs reversed does exist. Also, the OTs only have 4 and 8 ohm taps, but I've compensated for this on the driver by changing the value of the feedback resistors from 1K to 720 Ohms. All voltages check out properly, and bias range is very good...from under 1 volt on upward. I just can't find a source for this feedback problem. I did have one kind of wierd suggestion....that I consider pulling the feedback wire off the 8 ohm lead and apply it to the Common terminal. Still haven't done that yet.....and I'm kind of afraid to. I don't really understand that one. Anyway.....if anyone can help, please do...........I'm desparate!! Thanks, Tom

[sorenj07]

720 ohms is ludicrously low for a feedback resistor. have you tried leaving it unconnected to try the amp without NFB?

[Ty_Bower]

The original ST70 schematic shows a 1K resistor connected to the 16 ohm output tap. It appears to provide the global NFB back to the 7199 driver stage. Since he has no 16 ohm tap, I think he may be on the right track with the value of the NFB resistor.

Soren's recommendation is a good one. What happens when you leave the NFB completely disconnected from the output transformer? If the amp stabilizes, then you're onto something.

Think of the push-pull pair of output tubes as the "upper" tube and the "lower" tube. If the amp is stable with the NFB disconnected, but oscillates with it connected, then you probably want to try swapping the primary connections of the output transformer. Take the plate connection off the upper tube, and move it to the lower tube. Move the plate connection from the lower tube to the upper. Do the same thing with the UL taps as well (swap the UL taps from the screen of the upper tube with the screen of the lower tube).

While you're at it, you might want to double check and make sure you've got the OPT's plate leads on the plates, the UL taps on the screens, and make sure that each plate connection is matched up with its proper UL tap.

I also noticed on the ST70 schematic there is no resistor between the screen and the UL tap on the OPT. Some designs put one there - I think it helps increase stability. I'll let someone else chime in with their recommendation here.

[kevinkr]

You may need to swap the order of the primary connections, it sounds like you have positive feedback. Alternately you can swap the connections to the grids of your output tubes to accomplish the same thing. One or the other ought to be easier, but don't do both.

720 ohms is the correct feedback resistance value based on the square root of the difference between the 8 and 16 ohm taps. (You use the turns ratio as opposed to impedance transformation ratio.)

I think a resistor in the range of 100 - 1K on each of the screens of the EL34s helps reliability with modern tubes and might slightly improve linearity. (According to data in one of my Mullard handbooks.)

I have used Transcendar output transformers in SE amplifier projects and have been more than pleased with their performance.