JCM 800 clone being mean.

[Supersonic Amplification]

Hey guys, need help on a build for a friend.

This one has been elusive and tricky. It worked okay on the first power up and 2hr test. Sent it down to him and he says it doesn't work. I get it back and check it out, turns out the plate voltage is way low- and when the standby is toggled off, i get a rise in volume and voltage over about three seconds and fadeout. The problem seems to come and go, I've replaced diodes, tubes, checked everything up and down, even reheated every last solder joint in the amp, and the problem persists. I had typed up a much lengthier description but my browser was kind enough to refresh and erase it all for me. :) I will elaborate on anything necessary. Any help from fellow builders out there? Seen this problem before?

Schematic and layout below. Pics upon request. Note that layout is closest I could find but not exact- bridge rectifier is not used, but 4 (very beefy) diodes in center tap full wave format. Also bias pot is full size chassis mount trim pot, not board mount. Mojotone kit, basically.

[HotBluePlates]

.... Note that layout is closest I could find but not exact- bridge rectifier is not used, but 4 (very beefy) diodes in center tap full wave format. ... Mojotone kit, basically.

Is the amp a Mojotone kit? If so, we oughta be looking at their schematic & layout, and comparing it with photos of your build.

You'll want to verify the power supply wiring first. Do you get stable B+ voltage with the power tubes removed from the amp?

[Supersonic Amplification]

Hi

Yes, it is a Mojotone kit- 2 signal caps are .002 uF different, but other than that it is a mojotone british 800 through and through. And yes, the links you provided are much more accurate for this build. (Thank you, I had trouble finding the right ones.) Currently running new Czech Mullard tubes. I will perform the "output tubes out" test (which I neglected to think about, thank you!) and report back to this forum- thanks! I'll be checking anode, g1 and g2 voltages.

[HotBluePlates]

... I will perform the "output tubes out" test (which I neglected to think about, thank you!) and report back to this forum- thanks! I'll be checking anode, g1 and g2 voltages.

No, we will need d.c. volts at all pins of all tubes during this test. And linger a bit on pin 5 of the output tubes to see if the negative voltage here stays constant or drifts.

The idea behind this is verifying the power supply B+ wiring & parts. At the same time, we're checking the preamp & phase inverter for obvious errors. Also we're leaving out the output tubes in case there is a bias supply error which might cause damage. We would expect B+ voltage to be somewhat high without the output tubes drawing current from the power supply.

You may also wish to measure a.c. volts at each tube's heater pins. If you use 1 meter lead to heater and the other to ground, you should expect to see ~1/2 the total heater voltage.

[Supersonic Amplification]

Okay, so here are the results of the power tubes out test:

V5 socket: Va= 18v with slow rise to 45 or so
      Vg3= 0.5v
      Vg2= 18v with slow rise to 45
      Vg1= -44v
      Vk= 0.4v

V4 socket had identical readings except grid bias varied by 2 volts. (-46v)

V3: Va 1&2= 10v and rising, Vg1&2 =0.8v, Vk1&2=0v

V2 and V1 look about the same, +\- 5v variances.

Bias node (after rectification and filtration) measures right at 71v.

All filament voltages look excellent- solid 6.4 under load across the filament winding. All seems well other than my b+ supply-

I'm definitely stumped now!

[Supersonic Amplification]

Also I did double and triple check the grid bias voltage- stayed very constant at -44 or -46 volts.

[sluckey]

Is the standby switch in operate position? Is the HT fuse good?

[Supersonic Amplification]

Yes, and yes- I even replaced the b+ fuse on the slight chance that it was operating intermittently.

Here is something else that may be helpful to us here: I have observed that when I "play" with the standby switch (e.g., cycle the switch with a few seconds in between toggling) the amp will sometimes crackle to life- but the volume is about half what it should be. I have two ammeters hooked in series with the OT and the plates of both power tubes, respectively, and they both read 36 mA @ idle (when the amp is working) again, very intermittent. I have tried bypasing the switch altogether and to no aval- the problem persists.

[Supersonic Amplification]

Sure seems like I have a screwy connection somewhere- but where? I've gone through the whole circuit twice and reheated everything- I have to question whether or not it's a component failure.

[sluckey]

Ain't much left. Check the B+ directly at the diodes.

[Supersonic Amplification]

Same deal, Sluckey- about 18 volts and rising. When the amp decides to work properly, a healthy 470 volts.

I wonder if this isn't a grounding issue i've overlooked-  if for some reason the HT center tap was intermittently or improperly grounded, then wouldn't the old Fluke measure a weird floating voltage? Hmmm...

[sluckey]

Measure the AC voltage at the input to the diodes.

[Supersonic Amplification]

795v RMS. 2,248 vP-P

[sluckey]

Measure the voltages in reference to chassis.

[Supersonic Amplification]

Bingo.

Sluckey's suggestion to check HT leads with respect to ground made me realize the problem: something I must have overlooked 100 times... The cheap, crap fuse holder used in the Mojotone kits has an adapter for the miniature fuse types, which is what was used in the build... The problem was that even with the fuse in its locked position, the rear terminal was noticeably loose and was not making contact except under certain conditions. This explains in part why the amp would occasionally work upside down but not in the cabinet right side up.

New fuse and holder, problem solved. Yay!!

Except...

During my testing, I managed to pull a really dumb move- I went to discharge the filter caps from the 200v charge still present,  but, unable to find a suitable resistor, I decided it would just be okay to short the caps to ground with no resistance... So, upon putting tubes in and trying to play guitar the amp has a raunchy and loud 120Hz buzz with the sound... No smoke, but I only left it on for a second or two. I'm pretty sure I messed up one or a few of the caps. Darn shucks. 

[HotBluePlates]

... I went to discharge the filter caps from the 200v charge still present ... I decided it would just be okay to short the caps to ground with no resistance... the amp has a raunchy and loud 120Hz buzz with the sound... pretty sure I messed up one or a few of the caps. ...

Not necessarily. Before I knew any better, I routinely discharged filter caps with a screwdriver to the chassis; I never had a failed cap (though the spark would sometimes startle me).

Now it's time to keep troubleshooting, but now looking at a different symptom...

[Supersonic Amplification]

Hmmm... Okay, so is there a way you guys know of to test a cap without a z meter or a capacitance meter? My scope goes to a max of 400V Pk so I am hesitant to look at the waveform on the scope... Theoretically if it can handle 400 pk, it should handle 400 pk-pk, right?

[sluckey]

Just turn the amp on. Measure voltage on cap. If the voltage looks good and the cap does not get warm, it's probably still good.

I would be surprised if you damaged the cap by discharging it as you did.