freestompboxes.org

Hey freestompbox'ers! I'm having issues with a build I started a couple weeks ago. It's a Pheonix amplifier (early version of what is now a Dr. Z Carmen Ghia). The schematic was posted on here a while back, as well as a layout, but I'll attach them again here. I was looking to get some advice here, if that's OK to post here.

This is my first 'on your own' build, without kit etc, so wiring is a little messy AND I got the bottom of the chassis drilling backwards (maybe a detrimental choice to keep going when that happened).

I am passing signal, but I have a significant issue.

My problem is, I'm getting feedback (maybe more high-pitched whine than feedback) that reacts with the volume knob (the feedback level drops when I increase the volume knob, but never goes away completely). In my switched layout, the rectifier tube is some 2 or 3 inches from both the volume and tone knobs. (I may attach a picture if it would help visualize the problem). The input and output jacks are NOT isolated from the chassis (in other words I didn't use plastic/rubber grommets) and I have a star ground for mains and the power transformer about 5 inches away from the input jack

Further symptoms include C18 (which I changed to .01 uF 600V because I didn't have any .0022 uF) and R32 are getting smoking hot..

I've decided I will take a picture of the layout, so be prepared for a mess as far as off-board wiring goes.

Upon screwing with copper tape trying to shield some of this stuff, I've a mind to botch the whole thing. Unsolder all the tube sockets and whatnot and restart with a new chassis, bottom plate and all. I'll be sure to take more pics once I'm finished with that bugger

Still would like to know why those two components are heating like they are. Maybe because I changed that value? I'm not even sure what they're accomplishing there between the plates and the OT.

That configuration of resistor and cap between the plates would just cut treble, rip them out and do the same between the grids if it's too bright. They shouldn't be getting hot at all unless the cap is leaky.

That fixed bias phase inverter is pretty weird too, never seen it before. Usually squealy feedback is easily fixed by switching the primary connections on the output tubes, but if your layout/wiring is screwy it could be anything.

Hmmm, sounds like your hunch is right to me. Layout needs some work. Try using a chopstick to move your output transformer wires around a little, that cured some whining in one of my builds surprisingly. I don't know why R32 is hot, I don't know why it's there either. Haven't seen a resistor and a cap accross power plates before like that.

@Fuzzy: wouldn't that feedback be super loud (the only thing you hear) if the OT wires were reversed? Sounds like he's got some oscillation sitting on top rather.

Usually the ot wire thing does make a super loud squeal, but the feedback resistor is pretty big so it might only cause a bit of instability, it's an easy thing to check just to rule it out. Check all the grid wires from where they leave the coupling cap to where they meet each grid, they're the one's that usually need poking with a chopstick or shortening if they're to blame. Then mess with the grounds, I like to daisy chain them together from the main filter to the input and ground to the chassis as close to the input as possible.

You could try replacing all the control-grid wires with shielded audio wire, just remember to ground the shield at one end of the wire only, the control-grid of a Valve/Tube is a high-impedance input and is very sensitive, and it only takes a signal-level of a few milivolts to produce a fairly large output since Valves/Tubes are Voltage Amplifiers....

Hope that helps fix the squealing issue...

You can also try installing Grid-Stopper resistors on the control-grids of the Power Valves/Tubes, C18 and R32 form a low-pass filter to tame the High-End response of the Output-Transformer, I would say that if R32 and C18 are getting hot then C18 is probably leaky, it's Dielectric Insulation could have suffered a puncture from a back-EMF spike from the Output-Transformer, if you need to have R32 and C18 in place for the Amp to work properly, try and replace C18 with a same-value cap rated at 3KV at least....


As a side-note, I noticed the strange configuration of the phase-splitter (V2), I'm wondering if the amp would work better if the configuration was changed to a Long-Tail-Pair/Differential-Amp Configuration, as it appears to me that looking at the schematic,the control-grids of V2 seem to be biased positively, or am I mistaken?...

Actually...no, it's a form of Back-Biasing....

Thanks for all the suggestions! I'll be sure to screw around with the OT wires before I dismantle the beast. I'll also try replacing the possibly leaky cap and move the resistor/cap combo across the grids. I'll be sure to post my findings!

Also ordered another chassis last night, so maybe I can do this amp justice.

Upon more tinkering with the circuit and things, I found that when the amp is first powered up, I hear the same whine I heard from the speaker in one of the tubes or transformers. I think it's the rectifier. It's like the tube does the whine thing and then all of a sudden it starts coming from the speaker and LOUD

Also, when I go to discharge the caps, I ground the positive side of the cap to the chassis with a screwdriver and nothing pops or anything. Usually in the past, I associate discharging the caps with an audible pop and sometimes even a small spark. Any thoughts?

To check for a leaky cap, I can apply a voltage to one side and check for a voltage drop, right? If I have any voltage drop smaller than the supply, it is leaky, correct?

I don't know man, looks like something's funny, it's so hard to troubleshoot without the amp in front of me. I can tell you the filter caps not sparking is good, this amp has a 150k/27k resistor combo going to ground that will slowly discharge all the filter caps. As far as checking for leaky caps, I've checked leakage before with coupling caps by leaving them attached to the b+ voltage, and lifting the other end, then checking the lifted end to ground with a multimeter. I've read you need to put a 1meg resistor from the lifted end to ground and measure across that if your meter has low internal resistance, but I did it both ways and it worked fine without the resistor using my cheap Canadian Tire meter.

OK, so I've got this guy boxed up in a new home and almost done minus three ground connections (those things you crimp onto, not sure what they're called).

I've got a question about the negative feedback loop there from the output jack. In redoing the amp, I figured I may as well utilize the other two taps on the OT for 16 ohm and 4 ohm as well as the 8 ohm I was already using. If I connect all three jacks as the schematic calls for, won't all three jacks always be connected together, and cause some problems? Should I just go for one jack with a on/on/on switch to choose 16/8/4 ohms?

New and much cleaner gut shots to follow when I'm done

You changed C18 is 10nF when 2n2 was asked for. So it's 5 times to big and shorting the plates,,,that is why it's hot.
It may also be putting serious load on the PSU.
Remove that thing if you don't have the right value. It's a conjunctive filter or something trying to wipe off the excessive hi end hash.

Better ways to do such things,, Normally a a small cap across the plates on the PI is enough.

at 3m3 the FB is of the planet?? In my understanding with Valve FB past 100k you may as well not have FB at all.

Some one mentioned going back to a basic LTP PI and scrap the fancy stuff. GREAT IDEA .
Make R14/15 100k if you do.
Phil.

Haha thanks for the suggestions Phil. I figured the cap value change was the problem. I've since replaced it with the 'correct' value. What do you mean by LTP phase inverter? Maybe once I get this guy working I'll try that. I'll also try wiring the feedback just to the jack I'm using at the time and switching the loop in and out. I'll let you know if there's a big difference; 3.3 M is a big value I may also try subbing 2.2 M for a little more feedback (it should give me less gain, right?)

EDIT: Long tailed pair

So the oscillation wasn't due to my poor first attempt. Something is up with the schematic or layout. I'll be double checking the layout vs schematic.

My PT has 3 wires for 120/110 volts. One grey which is for 120V, one white which is for 110V, and one black which is common to both. I should be wiring the black to the switch and then fuse and then the power jack, and the white directly to the power jack, right?

If any of you guys could look over the schematic for any obvious issues, I'd be grateful. I gave it a once over and didn't see anything funny, but I've certainly not got the best eyes yet for this sort of thing.

jwpartain1 wrote:So the oscillation wasn't due to my poor first attempt. Something is up with the schematic or layout. I'll be double checking the layout vs schematic.

My PT has 3 wires for 120/110 volts. One grey which is for 120V, one white which is for 110V, and one black which is common to both. I should be wiring the black to the switch and then fuse and then the power jack, and the white directly to the power jack, right?

If any of you guys could look over the schematic for any obvious issues, I'd be grateful. I gave it a once over and didn't see anything funny, but I've certainly not got the best eyes yet for this sort of thing.

If it's any help, the normal convention for wiring-up mains power to a power transformer, is to have the fuse in the Active side after the power switch, so you wire from the mains socket Active, to the switch, then from the switch to the fuse, and then from the fuse to the transformer, the whole point of this is that if for some reason the transformer's primary suffers a short-circuit, the fuse blows and cuts the power and prevents the chassis from becoming live.....

Sounds like a positive feedback, try to reverse the wiring of the OT (primary OR secondary)
Beware, C18 MUST be a high voltage cap, 1Kvolts or more.

Yep, I think greenmachine said it too, switch the damn primaries. Oscillation is gone, 60 cycle-ish hum still there. Why does switching the primaries work by the way? I thought they were identical just 180 out of phase?

Just remembered I had put an ohmite resistor I had laying around in the power section, so I replaced that with a metal film, but didn't help the hum. Whenever I touch the chassis the hum quiets a little, but is still there. I've got a bottom plate on the chassis too.

The tone I'm getting is pretty sick though, minus all the hum. Now if I can just get rid of that

jwpartain1 wrote:Why does switching the primaries work by the way? I thought they were identical just 180 out of phase?

Because of the 180° !
180° means that your negative feedback becomes a positive one, hence an instant purely electronic larsen.
First cause of hum : the heaters wiring, you have to twist them TIGHT and put them as close to the chassis as it's possible.
Then, you should provide them a virtual ground if not done -not on the schematic, so i don't know- with two 100 ohms resistors (2 watts) and even better, as the amp is cathode-biased, connect the 2 resistors not to the ground, but to the power tubes shared resistor, to make the AC 6V3 floating on a DC 10 volts.

The whole point of the Negative Feedback system is basically to do two things, to flatten the frequency response of the amplifier, and to reduce Harmonic Distortion, to do this, the output signal has to be fed-back into the input in such a way that it opposes any changes, in other words 180 degrees out-of phase, if the output signal was fed-back In Phase, the result wouldn't be an Amplifier, but a very powerful Oscillator oscillating at some frequency.....

Exactly, happened to me when i tried a NFB in my Special 6.
A nfb also lowers the output impedance, and therefore rises the damping factor of the power stage.
Not always a good thing

I've also experienced it too, deffinitely makes your ears ring after hastily turning off the power....