Feedback weirdness.

[wobuse]

Hello Tubbies!
I just fired up my ST35 and I got a strange "motorboating" sound,without any input signal. I shut it down and checked the ohm readings around the output trannys. The readings around the feeback return was way off,so I disconected the feeback from the 8ohm tap. Now,it seems to work fine. I am using Uncle Neds TF110-48-UL output transformers. Any one else have a simular problem?

[EWBrown]

The Uncle Ned OPTs, the brown and blue wires/sides are reversed from the Hammond wiring scheme, just try swapping the plate and UL leads to the opposite half of the channel, it should go away. I'll check my built up unit at home and post how it is wired here later.
I've sen this blue/brown swap on other makes of OPTs as well.

A quick easy check is to simply reverse the ground and 8 ohm secondary leads, just to make sure. That involves only 2 wires instead of four, for each channel. If you're not going to use 4 ohms, it can be simply left this way and it will be fine for 8 ohms.

WHile you have the feedback wire lifted, check its resistance to ground, it should be approximately 21.3K , the pot contributes 20K, and add the 1.3K resistor in the voltage amplifier cathode. It doesn't have to be exact, just close (just like in horseshoes, hand grenades and atom bombs, close is good enough)....

/ed B in NH

[Shannon Parks]

Hello Tubbies!
I just fired up my ST35 and I got a strange "motorboating" sound,without any input signal. I shut it down and checked the ohm readings around the output trannys. The readings around the feeback return was way off,so I disconected the feeback from the 8ohm tap. Now,it seems to work fine. I am using Uncle Neds TF110-48-UL output transformers. Any one else have a simular problem?

What Ed said is exactly correct.

Firstly, the blue wires are pairs (blue to plate, blue/white to screen) and the browns are pairs (brown to the other plate, brown/white to the other screen). This is similar to Hammond.

Secondly, if you have that correct, then the phases are swapped and you were applying positive feedback instead of negative feedback. Good thing you had no input, or you would have had a nice howl of oscillation instead. When Ed says that the tranny is reversed from the Hammond I'm assuming the blue and brown sides are swapped. My apologies, as I didn't document this in the Fender ST35 mod doc! Please report back and I'll fix this.

This 'phase swap' was the first trouble shooting I ever encountered in the realm of tube amps. I wish I could say I figured it out, but my dad did. Actually, the amp should sound OK at this point, but you'll notice the high noise floor. That will disappear after the NFB is added.

Shannon

[EWBrown]

The blue and blue/white wires connect to the terminal block closest to the front of the board (the front being where the audio input connections are located). The brown and brown/white go to the terminal block towards the rear of the board.

When I first built up the rev D with TF-110-48-ULs, I had an electronic foghorn, reversing the primary connections turned it into a darn fine sounding amplifier.

I ran into this same situation when UL-modifying the K-502 11BM8 amp, the original cheap OPTs had their brown and blue wires phase reversed from those of teh Hammond 1609s.


/ed b in ice-coated NH

[Shannon Parks]

When I first built up the rev D with TF-110-48-ULs, I had an electronic foghorn, reversing the primary connections turned it into a darn fine sounding amplifier.

I ran into this same situation when UL-modifying the K-502 11BM8 amp, the original cheap OPTs had their brown and blue wires phase reversed from those of teh Hammond 1609s.

We always have a 50/50 chance of getting it right the first time. 8^)

/ed b in ice-coated NH

It's been very wet around here - but I'll take that over snow and ice any day!

Shannon

[EWBrown]

One nice thing about single ended triode (or pentode) amps, with no feedback, is that this phase swapping isn't an issue normally. But if one ties the two channels' ioutputs n parallel, or in series (both methods are valid) to increase the power output, then phase does matter.

Slightly OT, but useful:

FYI, there is an easy cheesy way to add NFB to the output stage of an SET amp, and that is to ground the cathode (and resistor & capacitor) to the "hot" end of the OPT secondary. This does put a very small DC voltage across the secondary and speaker, but it is millivolts. Of course, it is phase sensitive. There are some neat tricks for "parafeed" SE amps, too but I won't go into those here, too far OT.

/ed b in the frozen zone (and with bruises on my bum to prove it)

[wobuse]

Thanks Guys.
I will try and implement those changes,tonite.I will get back to you with the results. I will also try and post some pictures,after I figure out how.
Thanks & Best Regards.
J.Phillips
Check out my website at www.retrosonik.com

[Thermion]

When I rebuilt my Dyna MKIIIs the first time, I had to replace the original output transformers because one had gone bad. I couldn't afford the Magnequest replacements at the time so I bought a pair of Hammond 1650Ns. I was very careful to adjust the wiring color codes when building the amp. Upon starting up, I used a speaker for the load and got the foghorn siren that Ed was talking about. My wife thought that I had electrocuted myself in the basement and came running to investigate. It took me several days to figure out what had occurred since we didn't have internet forums back then.

After long investigation, I discovered that David Hafler used the output transformer to invert phase and maintain the overall phase of the amplifier. It just isn't marked that way on the published schematic. Usually a transformer diagram has dots to show when the winding gets phase inverted. Anyway, since the amp was point-to-point all I had to do was swap the drive lines. The amps have sounded great since. (much better now with diytubes prototype driver board)

Maintaining overall phase was a hot subject in high-end audio several years ago. The good news is that with tube amps it is easy to insure that overall phase is maintained by swapping the leads on the output transformer.

JT