I definitely removed the damping control components on my W-6s. You might check the actual FB level present. Mine measured at 23 db.
Dave
Dave
Heathkit W6 testing and issues
- Thread starter nerdorama
- Start date
You might also scope the drive to the output tubes to see what’s happening there. There is a little bit of grid current happening at the onset of clipping, so the peak drive signal should slightly exceed the bias voltage.
Dave
Dave
I'm getting the second amp delivered from my friend this afternoon. Thought I would check that one before making any modifications. Hopefully it behaves better but if not I'll begin by removing the damping circuit and see if that helps. It's a little baffling as I have compared the installed circuit to the schematic and I can't find a flaw. Not ready to give in yet. If this second unit shows the same bad behavior it seems like there must be something wrong in the restoration because these amps must have worked ok in days past.
Thanks and I appreciate the advice and comments.
John
Thanks and I appreciate the advice and comments.
John
If all feedback is removed, it shouldn't oscillate. There has to be a signal path of some sort from output to input for oscillation to take place. It might be unintended, such as poor decoupling or unwanted coupling between stages, but there has to be a path. What you're seeing may in fact point to the problem.
The inductance of those pots is insignificant at audio frequencies. If the oscillation is below 1MHz, they're unlikely to be contributing. Spealking of which, do you know the approximate frequency?
Jack
That seems like a lot of GNFB, especially considering the feedback additionally provided by the UL connections. The OPTs must have a very wide bandwidth!
Jack
Well, have some progress to report. I removed the damping circuit from one of the W6's and that seems to have largely resolved the issues. Square waves at 10k are much squarer. The still have a little wobble across the top but it's minor. FR is very smooth with -3dB at just over 100k. I haven't made the other measurements requested but I will still have them for a couple of days. Hope to get the other damping circuit removed this afternoon.
I also have another question. The rebuilder that did these amps sold my friend to bias the 6550's at 56mA. This seems a bit cold to me. The GE data sheet says 75mA at zero signal. Any suggestions for a "good" bias current?
This is with an input equal to what a 1W sine requires.
This is with mid level power
And this is with approx a 1W output
I also have another question. The rebuilder that did these amps sold my friend to bias the 6550's at 56mA. This seems a bit cold to me. The GE data sheet says 75mA at zero signal. Any suggestions for a "good" bias current?
This is with an input equal to what a 1W sine requires.
This is with mid level power
And this is with approx a 1W output
You still have feedback connected somewhere, or has everything been removed?
Jack
Jack
The W6 OPTs tested better than any other, I believe, in the John Atkinson tests with only the Cit II being competitive.
There is still a bit of wobble in the top of the 10k square that doesn't show in the original manual so not sure where that comes from but it's a big step from where I was yesterday. If Dave G hadn't said that he definitely would remove the damping circuit I wouldn't have been positive that I needed to do it.
I'm not convinced this troubleshooting has gone the right direction. The amplifier must have worked at one time with the damping control connected. In this case, it doesn't appear that the amp can even meet spec in the factory configuration. Is that normal for a W6? If not, I think it's still essential to disconnect all the feedback, damping, and compensation, and check for a problem. Square waevs aren't necessarily useful for this, mostly gain and frequency response.
Jack
Jack
trobbins
Super Member
The manual plot relates to a 16 ohm load, and you were using 8 ohm, and feedback is from the 4 ohm. Can you clarify if 'removing the damping circuit' means the 300 ohm pot wiper was at the cathode end?
Yes, wiper was at the cathode for max damping.
Correct. I paralleled a 470 with a 10k. The combo measured 450.2 ohms.
Did a little more testing after removing the damping circuit from the second unit. Did an open loop and closed loop comparison which calculated at about 20.97dB of feedback. Unfortunately I forgot to disconnect the shelf network so the FR I measured open loop isn't accurate. Maybe I'll get motivated again this afternoon to do more.
Did a bit more work this afternoon and found something very interesting. The load resistor I've been using is a big aluminum casing and rated non inductive for 100W. I moved it out of reach so I grabbed a Dale 50W gold casing type non inductive resistor and now the amp seems to work fine. Another lesson learned for me.
Except for the damping control which is already gone so to late to try it with this load but with otherwise all of the stock values for feedback, shelf and input resistor I get these results:
1kHz
10kHz
FR
I can tweak the HF extension by playing with the phase advance cap and the shelf but it seems good as is and very smooth.
Except for the damping control which is already gone so to late to try it with this load but with otherwise all of the stock values for feedback, shelf and input resistor I get these results:
1kHz
10kHz
FR
I can tweak the HF extension by playing with the phase advance cap and the shelf but it seems good as is and very smooth.