Harmony H-306A loss of volume

[cjonesplay]

Hello,

I have an H306A by Harmony, and the following happened:

I replaced most all of the resistors and caps in it, and it still had a waterfall sound when I turned the first/second channel volume up. I lived with it until one day the volume suddenly dropped to almost nothing in both channels. Letting it rest and turning it back on would initially allow it to work again, but as soon as I turned up that first channel it would have the waterfall sound and then cut volume again. Now it just cuts out as soon as I hit a note.

I have switched all the tubes except the rectifier and tested voltages on the filter caps and power tubes. Dc Heater voltages are both 274, and the voltage coming to the 12ax7’s is around 120. The filter caps all test ok and put out 270-324 volts. I tested ohms on the power transformer and they all are within reason and it seems to work ok. No weird smells or burns, no caps or resistors testing as bad. Tried different speakers and I’ve poked with the chopstick until the cows came home. Now I have like 30 cows outside my door and nothing seems to give a clue as to what the problem is.

I’m out of solutions at this point. Anyone got any idea what could be causing this issue? I’m open to any/all ideas, any help would be appreciated!

[cjonesplay]

It’s the top Schematic:

[pdf64]

Dc Heater voltages are both 274

Can the above be clarified, it seems weird?
A full voltage survey may be beneficial, VDC of every electrode of every valve.
For both the normal and the fault (cut out) modes

[acheld]

I looked at the schematic for the H306A ( the bottom schematic), and do not understand why the heater circuit shares its artificial ground with the cathode of the power tube.  Couldn't that introduce 60Hz hum into the bias of the power tube?   

Was that done simply to save the cost of a 100R resistor?  Or is it an elegant solution to a problem that I don't understand?

[sluckey]

Harmony = CHEAP!

http://www.valvewizard.co.uk/heater.html

Read the section on heater elevation.

[cjonesplay]

I can’t get voltages in normal mode yet. It only works if I leave it overnight, and even then only until a hard note is hit. Here are the broken amp voltages:

V1 (12ax7)
1. 126
2. -
3. .73
4. 15
5. 15
6. 137
7. -.33
8. .5
9. 15

V2 (12ax7)
1. 164
2. -
3. 1.4
4. 15
5. 15
6. 237
7. 6
8. 15
9. 15

V3 & V4 (6v6)
1. -
2. 15
3. 360
4. 246
5. -
6. 15
7. 15
8. 15

Rectifier
1. 1
2&8: 360
4&6: 330 AC

[pdf64]

I think you’re mistaking terminal 9 (heaters) for terminal 8 (cathode), as there’s no realistic readings seemingly for the 12AX7 pin8 cathodes.

V2pin8 is particularly significant.

VDC across HT caps C12, 13 & 14 would also be helpful.

[cjonesplay]

Sorry, v1 & v2 both have 15v on pin 9. V1 has .5v on pin 8 and v2 has 15v on pin 8.

C12:  167
C13:  250
C14:  355

Thanks for following up on this!

[pdf64]

There seems to be a problem in the cathodyne circuit, pins6-8 section of V2.
I suggest to check the resistors there, retension the socket contacts, try a different valve.

[cjonesplay]

Thanks for the info. There is an old ceramic .003mf cap in that part of the circuit that might have died and I’m waiting for a replacement, but I’ll try the other solutions first

[pdf64]

So R19 measures in spec for 2M7?
The problem seems to be very low anode and cathode current. If the cap C7 was leaking DC, the issue would probably be anode current would be too high, not too low.

[shooter]

V3 & V4 (6v6)

15(volts?) on 4 pins probably isn't good

[cjonesplay]

I had a 3.3M for R19 so I switched it out for a 2.7M. No difference

[sluckey]

V3 & V4 (6v6)

15(volts?) on 4 pins probably isn't good

Well, 15v on cathode (pin 8) seems reasonable. And since the heaters are elevated to the cathode, that explains 15v on pins 2 and 7. Pin 6 is not connected inside the tube so who cares. Probably just being used as a tie point. That should cover all 4 pins.

[shooter]

heaters are elevated

 

[cjonesplay]

So I finally got this amp to fire up healthy to make measurements the power tubes and rectifiers measure the same, but as soon a I touched Pin 1 of V2 with my multimeter the sound cut out and it went back to super quiet. No Amount of poking and wiggling brings it back.

Any ideas what would cause this? It’s connected to the .003mf cap which I am hoping to replace soon.

[pdf64]

VDC readings affected?

[cjonesplay]

I tried to take measurements of the v1 and v2 DCV’s but every time I touch a pin on either tube it immediately shuts the volume off and it goes back to broken and won’t work until the following day. So I got nothin’:(

[shooter]

with power OFF, unplugged, and left for a few hours
ohm from V 2 plates to chassis, let meter stabilize, what ohms do you get?

[pdf64]

Voltages in the fault mode may well be useful 

[cjonesplay]

I posted the voltages in fault mode above. Thanks y’all for the help, I replaced one of the last original parts today with a new .0033mf cap, as well as some several resistors. Not a huge difference, but now it’s a tiny bit louder with the volume all the way up. It sounds the same as if it was all the way up normally, at about 1/100th the volume.

I measured plate resistance to chassis on v1 & v2:

V1, pin 1:  655k
V1, pin 6:  568k
V2, pin 1:  770k
V2, pin 6:  767k

[cjonesplay]

For the sake of closure, the problem I had was a failed OT. Luckily, I found a great amp tech who replaced it for $50.

This amp is very quiet (4-5 watts)and the tremolo doesn’t come through strong. We couldn’t figure that part out, but we did find that the Lectrolab 600b is the same and I just wanted to post that schematic with voltages on it for those who also have this amp.

Maybe one day I’ll make this thing loud, but for now it sounds great and I’m happy. The end?

[sluckey]

Nice postage stamp schematic.   

[cjonesplay]

I have a better image of it but the forum says the file is too big, so here is the link:
https://lectrolab.wordpress.com/lectrolab-models/r600/

[pdf64]

R20 and 21 are the anode and cathode resistors of the cathodyne, they should be the same value, pass the same current, and so have the same voltage drop them. Yet R20 looks to have 30V across it, R21 has 220V.

[PRR]

File too big.... 90% of the bytes are in the tweed's shades of beige. Crop, hack the contrast, reduce the colors, it gets smaller.

[PRR]

> the cathodyne

What a nasty drawing. Took me many long seconds to figure that out.

And yes, the voltages are wrong, like a real-cheap amplifier 60 years after.

[shooter]

Took me many long seconds to figure that out.

I gave up and went to work!