New DR504 Build (Ceriatone Hey What 504) - No Sound

[MattG]

I just finished assembling a Hiwatt DR504. This is a kit build, using the Ceriatone Hey What 504 and transformers from Mojotone. Before the first power-on, I went through and double-checked every component and connection against the layout diagram, highlighting traces as I confirmed them. (That's not to say there isn't something missing or an error, I've had mistakes in other projects even after multiple checks!)

So, the fundamental problem is, it doesn't appear to pass a guitar signal. What I've tested/observed so far:

• I do see the faint red glow of the heaters on all tubes when the amp is on or in standby. None of the tubes red-plate when switched from standby to run. The B+ voltage (measured where the rectifying diodes meet) is 430v.
• When I switch from standby to run, there is a slight hum coming out of the speaker, which is pretty consistent with other tube amps I've used. I used my DMM set to DC volt mode, grounded one probe to the chassis, and used the other probe on the power tube grids - and that gets me audible crackling. So this says to me the power tubes, output wiring (including transformer), speaker cable and speaker are all good.
• With an instrument cable plugged in, I measured DC resistance between tip and sleeve: it's either 1M or 136k depending on whether I'm on the high or low input (this goes for both the bright and normal "channels"). I also measure nearly zero ohms between sleeve and chassis; and the measurements between tip and chassis are the same as between tip and sleeve. So I think the input jack wiring is likely right.
• I pulled the tubes from a working amp, so they should all be good (unless the circuit is so broken it's killing tubes immediately).
• On pin1 (annode) of V4 (phase inverter) the voltage is over 400 volts, and it should be in the lowish 300v range. I measured resistance from the power diodes to that pin (with the amp off and caps drained of course), and it's the series sum resistance of all the resistors in the path. So to see so little voltage drop at this pin suggests that the tube is pulling virtually no current. I feel that's a clue, but not sure what it hints at.
• The pots don't appear to have any effect currently - turning any of them has no effect on the small baseline hum, nor do I hear any scratching, crackling, pops, etc.
• Doing the DC meter probe on the preamp grid pins is... inconsistent. Sometimes I get very faint crackling (much quieter than the same thing done on the power tubes), sometimes I get nothing at all.

My next step is to put the tubes back in the working amp, to confirm that they aren't being destroyed immediately. Assuming the tubes still check out, my guess is that I have a short or open circuit in the preamp signal path. I'm thinking if I simply had a wrong value component somewhere, it would pass sound, but sound "wrong". No sound at all suggests the signal doesn't have continuity (or is being shorted to ground). So my next planned step is to use my meter to trace the signal path, looking for shorts or continuity breaks (with the amp unplugged and the caps discharged of course). Before I go down that road I thought I'd check here to see if anyone had any suggestions that might be quicker.

Thanks!

Edit: corrected B+ voltage; also corrected V3/V4 mixup.

[martin manning]

[*]On pin1 (annode) of V3 (where V1 is the first tube seen by the guitar signal) the voltage is over 400 volts, and it should be in the lowish 300v range. I measured resistance from the power diodes to that pin (with the amp off and caps drained of course), and it's the series sum resistance of all the resistors in the path. So to see so little voltage drop at this pin suggests that the tube is pulling virtually no current. I feel that's a clue, but not sure what it hints at.

Missing cathode ground?

[Dr Tony Balls]

[*]On pin1 (annode) of V3 (where V1 is the first tube seen by the guitar signal) the voltage is over 400 volts, and it should be in the lowish 300v range. I measured resistance from the power diodes to that pin (with the amp off and caps drained of course), and it's the series sum resistance of all the resistors in the path. So to see so little voltage drop at this pin suggests that the tube is pulling virtually no current. I feel that's a clue, but not sure what it hints at.

Missing cathode ground?

Yeah this sounds plausible. It might be a good idea to post all of your voltages, especially in the preamp.

[MattG]

Note I corrected some mistakes in my original post, specifically, I mistakenly called the phase inverter V3, when conventionally it's referred to as V4. The physical preamp tube layout goes: V1, V2, V4/PI, V3.

Anyway - thanks for the suggestions! I did verify with my ohm meter that all preamp tubes' cathodes have a path to ground. I took voltage readings last night. B+ voltage (measured at the rectifier diode junction) is 430v (also corrected in my initial post).

In the following, my measured voltages are shown first, and expected voltage is (in parenthesis). This is just for the preamp tubes:

• V1 (input):
  • pin1: 161 (170)
  • pin3: 1.2 (1.4)
  • pin6: 164 (175)
  • pin8: 1.2 (1.4)
• V2:
  • pin1: unused
  • pin2: unused
  • pin3: unused
  • pin6: 196 (203)
  • pin8: 1.2 (1.4)
• V4 (phase inverter):
  • pin1: 420 (308)
  • pin2: 59 (78)
  • pin3: 66 (81)
  • pin6: 420 (304)
  • pin7: 54 (70)
  • pin8: 66 (81)
• V3:
  • pin1: 289 (321)
  • pin3: 59 (78)
  • pin6: 149 (213)
  • pin8: 3 (2)

Given that my B+ is a little lower than what's typically called out on DR504 schematics/voltage charts (>450v seems the norm), it's not surprising that all the other voltages are a bit lower than expected. (I didn't measure mains voltage, but it's really hot and humid where I'm at, the whole neighborhood likely has their AC blasting, so mains voltage is likely sagging a bit.)

But the exception is V4/phase inverter - that's the one place where the anode voltages are substantially higher than expected.

[sluckey]

Measure the resistance between V4 pins 3/8 and chassis. Should be about 24KΩ. What have you?

[MattG]

Measure the resistance between V4 pins 3/8 and chassis. Should be about 24KΩ. What have you?

Confirmed - 23.5k.

[MattG]

Well, now it's working - but I don't know why.

Prior to my previous post, where I confirmed the 23k5 resistance, I had pulled V4, the phase inverter tube. Later, when I replaced that tube, I realized I had a 12AT7 in the V3 position. That's because I initially overlooked the fact that the tubes are physically in a counter-intuitive order: V1, V2, V4, V3. So I mistakenly treated V3 as the phase inverter, dropping the 12AT7 in there. So I replaced V3 with a 12AX7, and kept a 12AT7 in the V4/phase inverter position. I turned it on, and immediately measured the plate voltages on V4 - they were right around 300, where they should be. And all preamp tube grids made a popping/crackling sound when touched with my meter probe. And the V3/V4 grid pops were noticeably louder than the power amp grids, when touched with the probe. So I plugged in a guitar and... sweet glorious sound! All four inputs work, both speaker output jacks work, and all the pots sound like they do what they're supposed to.

So then I went through and tested all the 12A*7 tubes I had swapped in and out of this amp during testing. I exclusively used the V4/phase inverter spot. All tubes work. I still haven't put a 12AT7 back in that V3 spot, but I really don't think that would cause it to not pass signal at all?

So unless that V3 position is particularly sensitive to the different characteristics of 12AX7 vs 12AT7, my hypothesis is that maybe the tube sockets themselves had a bit of corrosion or maybe some protective coating, and the act of re-inserting tubes naturally cleaned that away. I guess I could try reading anode voltages with the tubes pulled, and if they are similarly high as before, then that would be more evidence of my "tube pins not making good contact" notion.

I'm glad that it's working, but randomly working without a clear fix is extremely unsatisfying. This is to be my gigging amp!

[MattG]

Updates:

I did one quick test: I swapped V3 and V4, i.e. put the AT7 in the V3 spot and the AX7 in the V4 (phase inverter) spot - worked fine.

Then I moved the AT7 back to the V4/phase inverter spot. I pulled V3 completely. I powered up and measured the plate voltages on V4: >400V. So I take this as some evidence that one or more of the tubes (likely V3) wasn't making good contact initially when the amp wasn't working.

I replaced V3 and powered up. Then I set about strumming the guitar while at the same time using a chopstick to poke/prod/tap/push all the solder joints, wires and components. I found a couple sketchy solder joints that would cause the sound to cut out if the wire was pushed just right.

I touched up the solder work on those sketchy joints. What's also interesting: I found one connection that was completely unsoldered - it had good enough mechanical connection that even hitting it with the chopstick didn't break the electrical connection. (Very humbling, I missed it on countless visual passes I did before first power-up.)

What I also found interesting: I did all this touch-up soldering with the amp rotated 180 degrees from how it had been oriented when I did all the initial assembly. I saw a few other joints that survived the chopstick and looked ok from the other side, but looked like they needed touching-up from my new perspective.

So, still no slam-dunk explanation for the initial problems. But bad tube connection and/or those couple lousy solder joints are certainly likely culprits.

Lessons learned: even if the amp is working, the chopstick test is worth the time it takes! Also, rotating the work 90 and/or 180 degrees for a new perspective can be revealing.

[martin manning]

I found a couple sketchy solder joints that would cause the sound to cut out if the wire was pushed just right.

I'm guessing one of these was the reason you had no sound initially. It would be unusual for a tube socket-to-pin connection to be open circuit with new parts, but it could happen. A 12AX7-12AT7 swap should allow some signal to pass, so that's not likely to have been the cause.

[MattG]

I found a couple sketchy solder joints that would cause the sound to cut out if the wire was pushed just right.

I'm guessing one of these was the reason you had no sound initially. It would be unusual for a tube socket-to-pin connection to be open circuit with new parts, but it could happen. A 12AX7-12AT7 swap should allow some signal to pass, so that's not likely to have been the cause.

It's kind of a coin-toss to me... the 12AT7 tube actually isn't "new", it's NOS - an old JAN Phillips tube. So the pins weren't shiny-clean like new current-production tubes have. Or it could have even been something really embarrassingly stupid, such as me not pushing the tube all the way in the socket.

Those sketchy solder joints weren't that bad - I missed them on the first pass with the chopstick. They really had to be pushed quite a bit to mute the amp.

Either way - I felt confident enough to load it up in a cab today, and played through it for actual enjoyment. It's great!

Anyway - thanks everyone for the help!

I posted a build report with pics here if anyone wants to see the finished amp.