More Sunn Sonaro questions

[dwhutchens]

I'm 99% sure this a Sorado circuit in a chassis with Sonaro printed on it. The tubes, voltages and component values are the same as the Sorado schematic. I would assume this thing should be very clean and loud with a guitar. Mine is not very loud and is very distorted with a guitar. With a bass the E string is pretty loud and strong but as the notes get higher the sound gets weaker. See my previous thread [Sunn Sonaro weak output] for some background. Can someone look at the attached waveform and schematic. The peaks on the ends of the wave form start to occur at about 7 on the volume control and are worse if the treble is turned up. This waveform is off pin 2 of the 6an8. The signal on pin 8 looks clean. The input is a 600hz 100 milivolt peak to peak signal from a function generator. I have the amp plugged into a variac set at 117vac and the voltages on all the tube pins are very close to the shcematic. Could this be the coupling caps or should I be looking somewhere else 1st.
Thanks,
David

[Amp Kat]

Wave doesn't look that bad considering it just got boosted hard by the Pentode. Pin 8 is the input to the control grid and directly coming from the treble pot which adds gain and why it gets worst. Scope out pin 5 and see what you get. If you have DC on the AC it would be jumping. If you check one side of the coupling cap with your DC meter and it's say 350 volts the other side should be close to zero if not zero. Try checking pin 2 to pin 9 with your ohm reading and see what you get. This is an Ultr Linear Amp so it will be a little less THD % than normal straight double ended OT.

[dwhutchens]

More measurements

I'm a little confused. I assume you mean pin 5 on the 6550s. Pin 5 on the 6an8a is the heater. I have taken more measurments and discovered some strange things, maybe. I calibrated my sig generator to my scope for a 600 hz, 100mv signal and the trace was centered top to bottom. On pin 8 of the 6an8a with the volume at 10 and bass and treble at 5 I have about .5v peak to peak and the wave is still centered. When I go to pin 2 the wave jumps to the top of the screen and I have to move it down and obviously raise the volt/div setting to see it. I have to back the volume down to 8 to clean up the signal and I have about 12v peak to peak. Now when I go to pin 5 of either power tube the trace drops off the bottom of the screen. When I raise it back up it looks the same as the signal at pin 2 of the PI. Also the wave on pin 5 of either power tube are identical. I was under the impression they should be 180 out of phase. Now something else I noticed. If I drop my sig gen to 40hz and slowly dial it up the amplitude gets smaller to 160 hz and then gets larger the higher the freq gets. I can keep the volume at 10 and get a clean sig at pin 2 of the PI up until just over 400hz. This may not be as odd as it seems to me, I don't have a lot experiance on these things. I did build a 5e3 from scratch and modified it into kind of a cathode biased princeton and I came out pretty good but I'm lost as to why I can't get a decent sound out of this amp.

The other checks I made, ohm reading between pins 2 and 9 on the PI showed an open circuit on the 2meg setting on my meter. As far as the coupling caps I have 315 on the top one and 130 on the bottom one and the other end of both is the grid equal to the grid bias of the power tubes (63v as I have it set now)

Thanks for reading all this mess and any other suggestions are welcome, I am about ready to start replacing all the caps that haven't allready been changed.

David

[Enzo]

You need to brush up on your scope skills, I suspect. Switch the scope over to AC coupling at the input, I bet you are set on DC now.

The pentode and triode sections of the 6AN8 are direct coupled. The grid at pin 8 is essentially at ground potential in terms of DC, so the signal varies around the zero point. But at pin 2, the signal is connceted to the preceding plate, so it is sitting at +75v, and teh signal varies around that level. Then when you go to pin 5, the grid of the output tube, it sits at the -55v of hte bias supply, an the signal varies around that.

If teh scope is on DC, then the trace includes both the DC offset AND the signal. Switching the scope to AC coupling allows it to ignore the DC and focus on the AC part alone - the AC part being your signal.

You want the signals at the two power tube grids to be identical. The phase splitter's job is to provide the same signal to both, but of opposite polarity.

They ARE 180 degrees oput of phase. The scope syncs to the waveform, not to its absolute phase. Normal sync to positive waveform movement wil always put the rising edge of the form on the left of the screen. The fact that the waveform you are watching is 180 degrees out from the other one won't make it look any different on the scope. The only way to see them out of phase is to have a duak trace scope connected one channel to each grid at the same time. Select the scope sync for one channel or the other, and then on your screen one waveform will be opposite in polarity.\

SO explore both input coupling and sync selection on your scope.

As to the variation in freq response, it might be nothing mor than exactly that. Guuitar amps are hardly flat in response. Did you try the same test with different EQ control settings? Also, did you scope your signal generator to make sure the output from it stays at the same amplitude throughout the range?

[dwhutchens]

checked my scope settings

You were exatly right, I checked my scope and I did have is set on DC. I put it on AC and now the trace stays in the center. I hooked up the other channel and put one channel on each power tube grid and sure enough the traces were exactly 180 out of phase. I plugged the function generator directly into the scope and confirmed that the aplitude stays the same from 40hz all the way to 20khz. Now all looks good except I think the frequency responce is all over the place. This a bass amp and that is what I want to use if for so if possible I'd like to even it out a little bit. I measured the voltage at pin 8 of the 6an8a at various frequencys and got these results:

40hz = 2.75v
160hz = .7v
500hz = 5v
2khz = 8v

Playing with the bass and treble and boost switches does changes things but I can't seem to flatten it out much. Does this much variation seem normal or could a flaky coupling cap cause something like this. I wish I knew what this thing was supposed to sound like. The best I can describe it if I plug in a guitar it sounds about like an old Silvertone (2 6l6) I used to have, kinda constipated.

[Enzo]

As I said, guitar amps are by their nature not flat.

[soundmasterg]

I'm friends with Conrad Sundholm, who was a co-founder of the Sunn company and now runs his own Conrad amps company. He's told me in the past that with all the EQ controls at zero, and the boost switches off, the amp was designed to have a flat response. These Sunn amps are very close in design to the Dynaco Mark 3 hi-fi amp for the power amp, and very close to the Dynaco PAS preamp for the preamp, so they have extended hi and low frequency response as compared to most amps made for guitar or bass. The output transformers are a large part of that response.

Personally unless I'm playing clean, I don't like any of the Sunn amps for guitar. For bass they sound pretty good, but you need to have an efficient speaker setup to have them really perform. I would suggest to email Conrad with the questions that you have about the amp and he can probably help you out just fine. Check out his site.

Greg

http://www.conradamps.com/