Selmer Concord 10 Rebuild/Conversion Issues

[BungleSim]

I'm rebuilding a Selmer Concord 10 for an employee of the shop I work at. I'm also converting it from 220V to 110V. Attached is the schematic. Please excuse the poor quality, it's the best I could find online. Also, the amp is using EL84s so ignore the pin numbers on the power tubes.

I got a PT from Mercury Magnetics with a HT winding of 290V-0V-290V and the rectifier I'm using is an EZ80. I wired up the primary of the PT as follows...

HOT --> power switch --> fuse --> PT
NEUTRAL --> PT

My power supply voltages are looking solid from what I can tell. The problem is when I turn the volume up, the volume starts to increase briefly but then actually decreases and it then this slow, volume swelling starts to happen. It feels like it takes a great effort to make sound come out, so there's this real "sucking" feeling as you play.

Pin 4 on the secondary of the OT is connected to chassis ground and the negative terminal of the speaker, so I have pin 5 of the OT going to the positive terminal of the speaker. When I first got the amp it appeared that pins 4 and 6 of the secondary were shorted together. I tried this but it only made matters worse and almost no sound comes out.

Any suggestions as to where I should start to try to troubleshoot this problem? I should also mention that it does not show it in the schematic but the center taps of both the HT and the 6.3VAC heater voltage are grounded on the PCB. I did this as it followed the previous wiring scheme, though I only figured this out from deduction (looking at cut wires and matching them up) so I am not 100% positive on the accuracy of this wiring scheme.

Thanks in advance for any help.

[Stevem]

So what plate and screen voltages are you seeing on the 84s?

[BungleSim]

So what plate and screen voltages are you seeing on the 84s?

282V on the plates, 285V on the screen grids.

[BungleSim]

I'm seeing about 24V of ripple on the B+. I'm thinking I need to add a 22uF/450V cap right after the rectifier to try to kill that.

Attached are two screenshots of the oscilloscope. I was feeding a 220Hz signal into the amp. One shot is with the volume at about noon, the other is with the volume maxed. You can see every other peak being slightly lower than the others, demonstrating the volume swell.

[Phil_S]

I wired up the primary of the PT as follows...
HOT --> power switch --> fuse --> PT
NEUTRAL --> PT

Place the fuse in line before the switch so that the hot line feed sees the fuse when you insert the plug into the wall outlet:
Hot>fuse>switch>PT
If a fault condition is present, the fuse will blow immediately, protecting both the house (most important) and then the amp. If there is a fault on the other side of the switch, the fuse will blow when you close the switch.

What follows is amateur level observation....

Did you retain the oil filled cap? It must be old and past its useful life -- an accident waiting to happen. If so, I'd replace it with an ordinary dry electrolytic caps. I'd use 3 individual caps if room permits, or you can buy a 3 or 4 section can cap, but at a much higher price. I am thinking that will address the ripple very nicely.

Again, if it was me, I'd rearrange the B+ ladder. The amp presently has 3 nodes:
B+1 (marked LI?) feeds the OT CT and nothing else. That looks OK.
B+2 (marked OR) feeds the screens and 3 triode sections with no decoupling between the screens and the triodes -- 5 takeoff points. I don't like this at all. I think I'd feed B+2 to the screens only or maybe the screens and the tremolo section.
B+3 (marked RT) feeds only the 1st triode. I think I'd use this one to feed both sections of that tube unless there is a meaningful voltage differential to be maintained for the sake of keeping the design intact.

I'd be inclined to add a cap and decoupling resistor (maybe 1K) between B+2 and B+3 to feed the cathodyne inverter. I might use this node to feed both sections of that tube -- the inverter and tremolo.

I'd ground both center taps to the same place. I'd run a bolt through the chassis and do it there, near the transformer, but not on a transformer bolt. Secure it well -- star washer, ring lug(s) on CT(s), star, nut.

On the speaker, what happens if you connect 4 and 5 to the speaker and ground 6 to the chassis?

[Stevem]

You have three nodes of filtering, but if you are still trying to use that original multi section can cap then forget it , as that is likely the source of your 24 volts of ripple!

[BungleSim]

I forgot to mention the original multi-section can cap is no longer present. A previous tech removed it and replaced it with three electrolytics.

Any insight on the "sucking" sound as I turn the volume up? Or do you think that's related to the ripple? When I really dig in, I see the B+ sag about 15V.

[BungleSim]

I wired up the primary of the PT as follows...
HOT --> power switch --> fuse --> PT
NEUTRAL --> PT

Place the fuse in line before the switch so that the hot line feed sees the fuse when you insert the plug into the wall outlet:
Hot>fuse>switch>PT
If a fault condition is present, the fuse will blow immediately, protecting both the house (most important) and then the amp. If there is a fault on the other side of the switch, the fuse will blow when you close the switch.

What follows is amateur level observation....

Did you retain the oil filled cap? It must be old and past its useful life -- an accident waiting to happen. If so, I'd replace it with an ordinary dry electrolytic caps. I'd use 3 individual caps if room permits, or you can buy a 3 or 4 section can cap, but at a much higher price. I am thinking that will address the ripple very nicely.

Again, if it was me, I'd rearrange the B+ ladder. The amp presently has 3 nodes:
B+1 (marked LI?) feeds the OT CT and nothing else. That looks OK.
B+2 (marked OR) feeds the screens and 3 triode sections with no decoupling between the screens and the triodes -- 5 takeoff points. I don't like this at all. I think I'd feed B+2 to the screens only or maybe the screens and the tremolo section.
B+3 (marked RT) feeds only the 1st triode. I think I'd use this one to feed both sections of that tube unless there is a meaningful voltage differential to be maintained for the sake of keeping the design intact.

I'd be inclined to add a cap and decoupling resistor (maybe 1K) between B+2 and B+3 to feed the cathodyne inverter. I might use this node to feed both sections of that tube -- the inverter and tremolo.

I'd ground both center taps to the same place. I'd run a bolt through the chassis and do it there, near the transformer, but not on a transformer bolt. Secure it well -- star washer, ring lug(s) on CT(s), star, nut.

On the speaker, what happens if you connect 4 and 5 to the speaker and ground 6 to the chassis?

Excellent advice. Thank you. I will get back to you on your questions when I'm back in the shop tonight.

[BungleSim]

Phil_S, is this what you are talking about with the power mods?

[Stevem]

What is happening to your plate and screen voltages as the amp gets to the sucking / swelling point as you state?

[BungleSim]

What is happening to your plate and screen voltages as the amp gets to the sucking / swelling point as you state?

At max volume (and max "suck"), I'm seeing the B+ sag about 15V when I strum as hard as I can on the guitar.

On the speaker, what happens if you connect 4 and 5 to the speaker and ground 6 to the chassis?

This did not improve things. Everything sounded the same. Since the owner wants to use a 16-ohm speaker, and I measured about 1.1 ohms between pins 4 and 6 on the OT, I have the speaker wired up with the positive going to 6 and the negative going to 4, which is also connected to chassis ground.

Edit: Last night I put a 22uF cap right on the output of the rectifier, so that brought the B+ ripple down from 24V to 4V and killed almost all of the hum. Increasing the cap to 47uF didn't have any further effect, so I'll likely just keep it at 22uF.

[Stevem]

So the hum is gone but the suck issue is still there?

[BungleSim]

So the hum is gone but the suck issue is still there?

Yes.

[Phil_S]

I know that I am an amateur, so please keep that in mind. I've got a dozen or so builds behind me. This is a hobby, not work. My chops may not be where some of the other guys are.

I am having trouble reading the schematic. I don't believe you have drawn what I'm saying but I suppose that will work, too. I'm posting my take on this. I'm guessing the screens need a separate node on the B+ ladder. That will stabilize the voltage on the power tubes and maybe prevent the voltage sucking. I also think the voltage sucking may be related to the tremolo and that is another reason for separating them.

On the OT, I'm thinking the schematic says it is a multi-tapped output for different impedance. If 4 is grounded attach the + side of the speaker to 5 or 6 and ground the - side. I generally ground the OT secondary and the speaker at the same point as the ground for the 1st filter cap. I would not leave the speaker floating.

[Stevem]

He has a separate filter node for the screens.
The tube spec chart for 6aq5 in class AB1 calls for a max of 250 volt on the plate, it also calls for a max of 30 volts of audio drive signal across the grids, is this the drive signal level you have?