• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Squeals of displeasure from DIY Dynaco ST35 clone

currituckco

currituckco

Super Member
I previously discussed this amp here:

http://www.audiokarma.org/forums/index.php?threads/problem-with-a-diy-dynaco-st-35-clone.820247/

But the problem persists - squeals & motorboating as soon as it warms up. I wondered if it was the driver tubes (new JJ 12DW7) but didn't have any others on hand, so I popped in some 12AX7s just to test and it was actually stable. I assumed that meant it was faulty tubes and ordered some NOS 12DW7s, but when I popped them in the squeals returned, and now even with 12AX7s there are pops and oscillations along with the music.

I took voltage measurements which showed higher than they should be all around - see below

Test point-----------------Reading-------------------Expected
First filter stage----------428--------------------------385
2nd filter (OPT prim.)--418---------------------------380
3rd filter (Driver B+)----379---------------------------320
V1/V4 pin 1---------------264/261.5------------------215
--------------2--------------132.7/135.8----------------112
--------------3--------------142/150.2-------------------116
--------------7--------------132.4/135.4----------------112
--------------9--------------1.0/.98----------------------.85

I may increase the dropping resistor before the second filter stage to try to bring the rest of the voltages down. I'm not sure why the voltages seem high coming off of the transformer (Edcor 120/330-0-330) - I don't think my line voltage is particularly high.

One interesting thing - when I place the DMM lead on V4 pin 2/7 (they're connected) the oscillation stops, or at least mostly does, while the music continues to play. V4 is the right channel 12DW7. The same spot on the left channel did not stop the noise.

Any ideas what might be going on?

The good new is that the Mullard EL84 I thought was fried is still going strong (see previous thread).

Thanks!
 
Last edited:
Register to hide this ad
cademan said:
Sometimes people get the output tubes plate wires reversed for one channel which will also cause squeal. Also the preamp tube sockets and pins may be dirty.
Click to expand...
The parts are all new, so I don't believe dirty sockets are at work, and I'm positive the plate wires are correct, as it was operating stably with the 12AX7s in for a time.
 
Cademan is probably right with the diagnoses. You might also check to make sure you don't have a problem with your fb circuit.
 
Maybe the capacitance that's naturally occurring in the meter is quelling an oscillation? I had the same thing happen when I was testing an amp where one channel appeared dead. Touched the test lead to the tube pin and it came back to life. Put an 10 pF cap on it and has been happily playing since. (Plate to driver tube, SEP amp)
 
I have glanced over the previous thread and this one, and one item of fact has not been clearly stated that I can see: Are you in fact using genuine Z-565 OPTs, either original ones, or the reissue modern ones? If not, that will certainly play havoc with your project as it relates to HF stability!

Dave
 
I've had motorboating with bad coupling caps. Go throught the amp with the schematic. All connections are where there are supposed to be. Make sure you dont have a cold solder joint in all the connections.
 
currituckco said:
But the problem persists - squeals & motorboating as soon as it warms up.
Click to expand...
Just on it's own, without music playing?

If all else fails, try temporarily replacing the feedback resistor (R10) with a two watt 10K pot. Connect one side of the feedback circuit to one end of the pot. Connect the other end of the feedback circuit to the wiper of the pot. Turn the pot up to 10K, then turn on the amp and after it's warm, slowly decrease the resistance of the pot. At some point the amp it will start to thump, squeal, howl, etc. Back off just a smidge until that howling stops. Measure the resistance dialed in on the pot. That will tell you the smallest resistor size your circuit can support without going unstable. If the measured resistance is more than the resistance of R10, then it's a very good sign that R10 is too small which is causing things to go ballistic.
 
dcgillespie said:
I have glanced over the previous thread and this one, and one item of fact has not been clearly stated that I can see: Are you in fact using genuine Z-565 OPTs, either original ones, or the reissue modern ones? If not, that will certainly play havoc with your project as it relates to HF stability!

Dave
Click to expand...
I'm using Edcor transformers that are the same specs... Was that an idea doomed to fail?
 
Yes -- unless you have the specific know how and equipment to adjust the NFB and HF Stability circuits appropriately.

High quality amplifiers -- of which the ST-35 is one -- use high quality OPTs, and a circuit designed with compensation circuits added that specifically adjust the circuit to the characteristics of the OPTs. We're not talking power rating, winding impedance, or even the position of the UL screen taps. Rather, we're talking about the supersonic characteristics of the transformer (that above 20 kHz), where resonances and response characteristics can actually cause NFB to become positive feedback if not corrected. To prevent this from occurring, the circuit is specifically tailored so that the response of the circuit is less than that of the OPT -- preventing the feedback from becoming positive.

As well, high quality OPTs use extensive interleaving of the windings, where a little of one winding is wound, then a little of the other, and so on, interleaving the windings like this a number of times. This is an expensive process but it ensures that the coupling between the windings at high frequencies (like 20 kHz) is as good as it is at 1 kHz or at 30 Hz. If you measure the power output of your amplifier, you will likely find that the power output will drop off notably at 20 kHz, and that distortion will increase notably as well. This is due to the incomplete coupling that lesser transformers have. When there is loss through less than ideal coupling, there is phase shift and loss of response, reducing performance versus that at mid and lower frequencies.

This is hardly to disparage the Edcor transformers you are using. Rather, they simply do not have the characteristics necessary to remain stable in the Dynaco circuit. The Dynaco circuit operates with 20 db of NFB. Your Edcors will likely only handle about 14 db, and require adjustment of the HF stability circuits as well. To make the adjustment properly requires a scope, square wave generator, and the skill to make the necessary adjustments.

I strongly suggest not using the amplifier with any high quality speakers until this is sorted out. The HF (ultrasonic) oscillations can burn out the tweeters quite quickly.

With the proper adjustments, no doubt your project can still be quite successful, although it will not have the performance of the original Dynaco amplifier. It is almost a certainty however that this is the source of your problems. It has always been said that with regards to high fidelity amplifiers, the OPT is everything. Your project is no doubt reinforcing that fact yet again.

I hope this helps!

Dave
 
Thanks so much, Dave - very I informative (if disappointing) post. I have much to learn, and I appreciate the knowledge you're imparting! I'm almost tempted to get a scope and try to learn to do the work of tailoring the circuit just for the learning experience, but for now I guess I'm off to try to source some original (or reproduction) Dyna OPTs. Anyone have experience with any worthwhile modern versions tailored to this circuit?
 
The modern manufactured Z-565 transformers are an electrical drop in fit, and do an excellent job of replicating the original transformer (based on personal experience). They are advertised as being wound on original winding equipment using the exact winding pattern as the original. The only difference between the new and original transformers is that the new transformers have an 8Ω secondary tapped at 4Ω, whereas the original had a 16Ω secondary tapped at 8Ω. If using a modern reproduction Z-565 transformer in an ST-35 (or a diy version of its circuit), the NFB resistor (R10) must be reduced to 18K (from 27K), while the phase advance cap (C7) must be increased to 39 pF (from 27 pF) -- this to account for the fact that the NFB is now being supplied from an 8Ω tap rather than 16Ω tap. All other component values remain the same as originally specified. Other than the different secondary impedance offerings, the performance between the original and re-issue transformer is virtually identical.

Dave
 
Not sure which specific Edcors you are using, but from my own experience running a set on a Magnavox 9300 amp and comparing what I needed to do for HF tuning vs what Dave's amp used with the Z565, I can definitely say my transformers and his are not identical.

This can probably be tuned, it will just take some time and tweaking to make it behave itself. The fact that it stops when you meter the phase inverter tells me quite possibly a step network will make it behave. The meter is basically some extra capacitence, and its causing a bit of roll-off at the very high end.
 
6DZ7 said:
Are those OPT's the CXPP-8 with the 23% UL winding taps? If so, I think the tailoring of the FB would be fairly easy to get adjusted with a little trial and error if you don't have a scope and sig. gen. The other standard OPT's use a generic UL winding tap and so may have other adjustment needs.
Click to expand...
Yeah, those are the ones. How do you recommend I proceed?
 
6DZ7 said:
Let me preface this by saying the problem may be other than FB, but this may still workout to give you something to go on that points to real problem.

So those have the 16 ohm speaker taps and I'm assuming you use that one for the FB line and 27K resistor just as the ST-35 does.

-- If so, then double the FB resistor value, just as a place to begin. You could use a 47K to start with. You are just looking for a stability point by reducing the FB.

-- If you are using the 8 ohm tap for the FB then use Dave's FB resistor value, but double that to 39K.

Post what happens.

With less FB the amp is going to have more gain so the volume control will be more responsive.
Click to expand...

I tried 47k and it still motorboated. I also put 18pf caps from pin 4 of the driver tubes to ground and that killed the oscillation, but at the noticeable expense of high end performance, so I'd rather try to dial the feedback. Thanks for walking through this with me.
 
6DZ7 said:
We want to seperate the squealing from the motorboating if they happen together. So you're using the 16 ohm tap line for the FB and have a 47K in place of the 27K and the squeal is gone now. Just have to work on motorboating, yes?
Click to expand...
Correct!
 
You must log in or register to reply here.
Back
Top Bottom