Marshall JCM800 Lead (2205) VERY HI HUM---HELP!

[nevetslab]

Correct me if I'm wrong, but with the scope set at 5ms, isn't that 60Hz hum we're seeing in those pictures? I would mess with the heater wires in that amp and see if the issue changes.

Earlier, I had replaced the heater supply all together with an external 6.3V CT source from my HP712C Power Supply (0-500VDC@ 0-200mA, 0 to -150V bias supply, 6.3VAC @ 10A heater supply).....that didn't change anything. I had lifted the heater wires and the heater C/T from Ground at that time.

Yeah, both 60Hz & 120Hz are present....the largest spikes are 60Hz...every 16.67mS.

Yesterday, I had also lifted out the IC, since it is run off of the rectified 6.3V CT winding. That didn't make any difference. I wanted to look again at that 3.7VDC power supply. I was very concerned it was the source... may still be. I haven't replaced that 3300uF buss cap yet.....only bypassed it with another cap...2200uF. The ripple on it was roughly the same as on a good working JCM800.

[nevetslab]

I was also wondering about that diode. I'm just getting back to the amp now, having been dealing with moving issues all day. I was also wondering about the low current bridge and the second diode around it....that's between VR4b Gain control and VR5 Volume control...ahead of the grid of V3A. But, with VR3 turned down, I don't get any of the radical hum/supply ripple coupling thru with turning up the Volume pot VR5.

About to go digging

[nevetslab]

While running with IC 1 removed, it sort of takes C11 out of the equation.....but, I'll yank it out too, just to be sure.

Something is sure coupling 60Hz field into the grid of V1B. Though I didn't see it on the grid..... Time to go yanking and cranking.

I'm still under the impression from the client that whatever went wrong with this amp, it happened at a gig.

[nevetslab]

With IC 1 out of circuit, I shorted Pin 5 of the IC socket to ground, and the hum virtually disappeared....takes C11 to ground. I'd expect that, as it's doing the same thing as turning the VR4a gain control down. There has to be high ripple at that local ground...which makes me suspect that 3300uF cap's ground.....nothing like having charging current on your presumed low-noise ground reference line. Gotta pull the board up to see, and will yank the 3300uF 3.7V buss cap out and replace it with something else....ain't got a direct replacement at the moment.

[g1]

As you have said, this problem just cropped up "overnight". olddawg wondered about generator usage, which I think is a good possibility. Maybe that cap you mentioned, or a compromised/slightly-toasted ground trace?

[nevetslab]

V1B cathode diode, D1, replace it.
Not that? Try bypassing the whole gain/ volume control.

I just tried a number of things.....including disconnecting V1B's cathode diode D1. With it removed, the 60Hz hum level dropped 11dB. Now, what I had also done was take out the 3.7VDC supply by lifting the two rectifier diodes D2 & D3, removed the bypass caps on the back side of the board C41 & C42, along with removing the 3300uF buss cap C4...that takes the charging current out. Since I already have IC1 out of circuit, I didn't see why I couldn't shut that supply down.

I also lifted the IC1 -pin 5 side of C11, so it's effectively out of circuit, though is hanging in the breeze while connected to V1B's grid.

Result is, the hum is still present, only 11dB down, and the waveshape is different, due to removal of V1B's D1 cathode resistor. Looking at the plate of V1B, with the cathode resistor lifted, there's not much shape change in the noise pattern as I turn the Gain pot up and down. And, again, with the dual gain pot VR4A & VR4B turned down, turning up Volume pot VR5 doesn't let anything in.. So, still have the hum/ripple source at V1B's circuit.

When I looked again at the ground traces, I see the ground path of C4 taking it's own 'dirty ground' path back to the common ground line at the end of the board, which collects all the other grounds and takes them to Chassis ground at the Third/Fourth filter stage caps C44A & B.

Back to the hunt.

[nevetslab]

does it go away if you ground the grid of the first stage? Sorry if I missed something but you did check the input jack tip is grounding?

I've been using a Shorting Plug inserted into the Input jack. But, with it removed, it doesn't change the hum level. I had de-soldered & re-soldered that input jack, after confirming the contacts were ok.

[nevetslab]

Looking closer at V1A at the plate, there is hum present....even though it looks more like power supply ripple, it is a 16.67mS repeating waveform....that's 60Hz. 120Hz is also present. Just for giggles, I replaced D1 cathode diode across R3 at V1B with a faster 1N4148 signal diode. Just the thing to do if you want even higher supply peaks to couple in. I didn't solder it in all the way...left one end open to see what it did. Won't do that again. What purpose does D1 have in this amp, anyhow? It's NOT helpful in this POS amp the way it's presently behaving.

I also removed the dual gang 1M Gain Pot VR4A/B, just to make sure both sections were ok, and not one end open. All was ok, though I'd like to thank the assembly folks at Marshall for their wonderful installation technique of totally twisting the dickens out of every lead, to insure foil damage and pot lead damage can be eminent! Only lost a little foil in the process, but it did feel good venting some choice verbage at the bloody thing in the process! Anyhow, the pot was not an issue. Found C39 to be a tiny axial ceramic cap...47nF, and just to be safe, I replaced it with a 47nF film, and did the same with C7...a 2nF ceramic that couples the plate of V1A to R11. Shorting either side of R11 didn't change anything.

I also removed (for the time being) the diode bridge and D6, which sits between ground and the junction of R28 and the top of the Volume pot ahead of V3A. Wasn't an issue....since the hum problem is already disasterously high once ya turn the Gain Pot up, and see it at the plate of V1B.

Calling it quits for the evening.........and have to resume this long tedious process of packing up lab and music studio assets at home to move out into storage, and perhaps become homeless in the process. Endless task. Back at this one on Monday morning.....

So, to recap.....I've still got IC 1 removed, have removed the 3.7VDC supply that runs it (D2, D3 lifted, C4 3300uF cap removed, C41 & C42 removed, D1, D6, diode bridge removed, main buss caps C43 & C44 replaced, V2 removed, C7 & C39 replaced, and it still hums like the dickens. Replaced V1 again, with no change.

Cheers, Steven

[nevetslab]

More of what the problem ISN'T

Back for more abuse by this JCM 800. I removed the pair of 6.3VAC leads off the heater string that go up to D2 & D3...which are still lifted, out of circuit, as is the 3300uF buss cap. IC 1 & V2 are still out of circuit, Cathode diode D1, bridge and D6 are also still out of circuit on V1B and ahead of V3A's grid.

This morning, I decided to once again try an external 6.3VAC heater supply. This time, I disconnected V1-V2-V3 from the rest of the JCM800's heater supply, and used the external heater from my HP 712C supply. First tried it with the CT tied into the JCM 800's ground where it's CT's are tied in. Didn't change a thing. Lifting and re-attaching the external heater's CT didn't change anything.

I then clipped in a 100 ohm 2W linear hum balance pot, with the wiper returned to that same Ground, while the Ext heater supply's CT wasn't grounded (floated the chassis on that move, as there was a path thru the HP 712C's heater CT to chassis). This did change things. Made it lots worse, where I could balance 60Hz & 120Hz hum content! Didn't see anything to that magnitude with just grounding or floating the external heater supply for V1-V3. So, that was amusing. I had forgotten to bring in a small heater transformer to do the same thing, so used the lab supply.

I had also checked continuity of all the circuit grounds of V1A & V1B's components, just to make sure none were open. I had virtually done that before with comparing the DC levels to that of a working JCM 800, but worth a second look

Improved the heater lead dress and twists for V1, V2 & V3, and played with the lead wires of V1, watching the plate of V1B in the process. I can't alter the hum pickup that way either.

Now about to dive under the PCB to lift out C6 & C3, disconnecting the tone circuit of the Normal Ch, just so that path isn't contributing to this hum source for V1a & V1B...and removing C11. With C6 & C7 out of circuit, there's only the one path to the grid side of V1B's Gain pot VR4A.

[nevetslab]

Lifting out C3, C6 & C11 didn't change anything. I didn't think it would.

I was reading an earlier thread on a JCM 800, concerning the voltage on V1B's plate being low, and the discussion regarding why D1 is included in the cathode circuit of V1B. With D1 removed, plate resistor being 470k and cathode resistor being 10k, I get 214V at the plate and 2.5V at the cathode. With D1 restored to a 1N4007, I get 70V at the plate and 0.54V at the cathode. And, the resultant waveform at the plate of V1B looks just as bad as it did before I removed that diode...which did drop the offending hum waveform 11dB. Looking for more like 30-40dB reduction of this offensive hum so I can kick this amp back out of the shop. But I would sure like to know the source of this hum, and how it changed overnight (according to the client). It was a rental. I routinely service their gear, and don't normally run into this sort of thing....Mesa Dual Rectifier 3-Ch Solo amp not withstanding, if anyone read any of that nightmare's posts.

On the working JCM 800 in our inventory, I measured 85VDC on the plate and 0.54V on the cathode.

I too have no idea why the diode is there. Gotta be a tonality thing...same with the diode bridge and D6 around it. I haven't put them back in, but removing them didn't change anything regarding this hum problem. I'm not convinced I should raise the cathode voltage to 2 diode drops and have a higher plate voltage, unless there's some magic potential the plate and cathode need to be for this hum to go away.

[nevetslab]

Magnetic Field measurements (relative) of this JCM 800

Still being out-foxed by this amp....including a hopeful moment of finding at two seperate times, a failed solder joint on one of the HT leads to the rectifiers, and the other, a bad solder joint on the front end grounding wire off of the Effects Loop jack's shielded cable...which looks to be how the front end circuits ground gets to chassis.....at the ground lug on the Reverb Drive Xfmr......neither, after repairing the joints had anything to do with it.

So, knowing this hum has to be induced into V1A, and amplified more by V1B.....it's gotta be coming from......what. The most likely has been the heater wiring, and next to that....magnetic field from either the power, output transformer(s) or the choke, which is the closest to V1 & it's circuit. The 120hz family of ripple is seems well filtered, with fresh caps installed.

To get a basic feel of what magnitude of stray 60Hz field is present, I reached for a flux loop I made for inducing current into guitar and bass pickups....a borrowed technique from using a flux loop to calibrate the Playback tape system of pro audio tape machines from years ago. None of this method is calibrated, and, with pickup coils and magnetic fields, proximity (distance) is your basic sensitivity adjustment, besides how much gain you have following this pickup coil. In my case, 80dB of gain, feeding the scope channel.

The highest field produced is, of course from the power transformer. I started by placing the flux loop a few inches from the side of the lamination stack, as seen in the first image. Then, I moved it close (maybe 1/2") to the output transformer (which is rotated 90 deg), and then close to the choke (also about 1/2"). The waveshape seen on the choke looks the closest to the shape seen coming out of the amp with the master full CW, Gain full CW and Volume of the Boost Ch full CW)



Now, inside the steel chassis, the magnetic field is substantially reduced, as can be seen by placing the flux probe right above the power transformer in the image below.



Then, checking to see what sort of field is in the vicinity of V1 at the opposite end of the chassis. In all these images, I've left 80dB of gain on the probe, and magnitude differences seen is by probe placement...here, it's sitting right above the tube socket. Scope image is distant, but it's close to a flat line with wideband noise.



Further probing inside the chassis, I found some field pickup along the power supply decoupling network, and also from the Reverb Transformer, who's primary is the load.,as seen in the next two images.



Above V3, there's just a tiny bit more field, presumably from the choke. V2 is still unplugged, and the low voltage DC supply that runs IC1 is still disconnected.



So, after probing within the chassis using this flux loop, I'm not seeing evidence it's magnetic field pickup. I haven't yet gone in with a current probe to see what's on the ground lines. My fear is that either the power transformer, output transformer or choke overheated, developed a non-catastrophic leakage path to the chassis. But, I don't see evidence of that in the current consumption/wattage of the amp under idle conditions. No easy way to clamp a probe around a portion of the chassis, short of removing the IEC power connector, or the AC Mains or Output Impedance switches.

So far, no luck in finding what seems like it should be simple to identify. I don't see the hum source in the power supply line, and haven't yet looked with the current probe on the front end ground return line (to chassis....connects at the Reverb Xfmr mtg screw. I did lift the jumper that has a black wire soldered to the rear pot of V4's Gain control, as the other end is from the ground return of the low voltage supply....not yet restored.

Any suggestions?

[nevetslab]

Eliminated O/T & Choke as hum source

After having pulled up the PCB again to review the preamp grounding, and having disconnected the ground lead from the preamp ground, which gets to chassis by way of the shield of the Effects Send/Return cable to the ground lug / mounting screw for the Reverb tank drive Xfmr, I tried different chassis connections, and pick-off points of that long ground trace that heads towards the rectifier end of the PCB before reversing it's course and meandering back towards V4, V3 & V2. Didn't matter where I connected it, it was equally bad. It was worse when connected to the 'dirty ground'...having seen that near the beginning of this entire trouble-shooting exercise.

So....it still left me with the nagging question of how is the hum getting coupled into V1A, and subsequently V1B and on. I decided to next eliminate the Output Xfmr. Started with finding and removing the Primary C/T lead from the HT fuse, and re-connecting the red lead of the choke. That took off the Primary of the O/T as a hum leakage path, but it still left the screens connected...I hadn't yet removed the power tubes. NO CHANGE IN HUM LEVEL...STILL AWFUL

Pulled the power tubes, looked again. NO CHANGE

Then, removed the Choke...found the other end tying into Pin 6 of V6 power tube socket...and lifted it off, then jumpered the HT fuse output back over to feed the rest of the power supply decoupling string. NO CHANGE

When I looked to see how much higher the power supply went with the O/T & power tubes out.....about 490VDC, touching the RED lead/terminal attached to the 1N4007 rectifiers common point with my Fluke 8060A...1kV DC range (Blk lead tied to chassis at the main buss cap), there was a loud bang with the AC Mains fuse blown. Huh??? One side of the rectifiers shorted, both of them. Checked the Secondary winding.....82 ohms, with near-equal resistance between the C/T and HT leads. I replaced the diodes, checked the main buss cap sections...all ok. Primary was fine. I restored order, this time with a 1A Fast Blow in place of the 4A Mains fuse. The HT fuse didn't blow. It powered up fine, walking it up slowly with the variac, watching the power analyzer. No surprises.

Still have no idea why it just up and blew...unless those diodes had somehow become weakened. It was the same side where I had discovered earlier the HT lead had come out of the solder joint yesterday...presumably from constant removal/restoration of the PCB assy into the chassis.

With the repairs done, I looked again at V1A and V1B's plate on the scope....hum level is worse than it was with the Choke in the system. That doesn't surprise me, though the ripple still looks very low.

Since the Marshall P/T still works, and I still have this horrible hum that is still mysteriously being coupled into the input stage, I don't know if the source IS from the P/T. If it is, I've no idea why I never saw the problem disappear way earlier when I powered it up from the HP 712C lab supply completely. I also thought I could disconnect all grounds from the chassis, period, and float the circuit, all connected, so there's no path to the chassis. Leave the Safety Ground in place....just isolate the amp circuit from chassis. If it's from a leakage path on the P/T, and comes up via chassis, that should eliminate it. Should.

Have I un-soldered and re-soldered all of the preamp tube wire connections to the PCB? No., but have re-soldered all tube connections on V1, V2 & V3. Played with lead dress, added twists to the heater wiring from stage to stage.

It all happened at a gig on a rental. What? What happened? It reliably has extremely high hum. Solid and dependably bad. And, I just can not find the cause. What is staring me in the face that I'm not seeing?

So, I guess the next thing to try is the external supply again, or a different one....I've a Kikisui 0-300VDC @ 150mA. leave it in the state it's in, disconnect the Heater supply from the Marshall P/T, and maybe a different Heater source. That's easier than lifting all the grounds off chassis.

Anyone have any ideas?

[Jazz P Bass]

My thoughts where the same.
The PT.

Oh, and Bob Pease, bless his soul, had a sure cure for problematic equipment.

[John G]

Hi,
Since the bottom of the barrel is in sight, I thought I'd throw in my 2Cs worth.I remember some 15years ago I had similar hum problems which turned out to be the earthing points (structure) around the reverb transformer ???. All this was a long time ago(memery fade) so a long shot concidering the problem appears to be at the front end. Also is the heater line still balanced ??
Cheers,John

[nevetslab]

Eliminated Power Transformer

Towards the end of yesterday's punishment session with this JCM 800, I dug out another All-Tube regulated power supply built by Trans Electronics, Inc, which provides a 225V-375V Regulated DC Supply voltage at up to 50mA, with a separate 6.3VAC C/T heater supply. It hadn't been powered up in decades, power cord was badly frayed, but apart from that, it looked ok. Brought it up slowly so see if it would work or not, and it did

This morning I replaced the power cord, and then prepared my extension leads, after first searching high and low for some # 5-40 Binder Head screws to thread into its' barrier strip...it only having one left! Why did it HAVE to be $ 5-40?!! But, found some on the back of an old UREI dual graphic, so I was in business1 Supply looked clean, in spite of its' age. I once again disconnected the heater supply from V4 preamp tube and beyond, and had already disconnected the HT supply from the Marshall-based supply. Unplugged V4 & V5, so only V1 & V3 tubes were present. Marshall Power Transformer now fully disconnected for this experiment.

Powered it up to have a look.



No Standby with this, so I engaged HT by clipping onto the power supply decoupling string between R59 & R58, applying 375VDC there, and ended up with 338VDC to feed V1 from....about what I normally see with the Marshall supply. I was expecting to finally see absence of hum....though having been down this same road with using the HP 712C power supply, I would have been surprised.



Same nasty high-hum waveform I've been seeing all along! Even disconnected the C/T lead, and clipped in the 100 ohm 2W hum balance pot. It just adjusted balanced the same 60Hz /120Hz components. Just as bad. The 50mV/Div scale deflection should read 500mV/Div....sometimes the Tek readout probe coding doesn't respond. Anyway, NO CHANGE......same as before. Even though all the rest of the amp is disconnected, this is what I've been seeing at the plate of V1B with the Gain control turned up. With the rest of the amp all together and speaker attached, it's just totally obnoxiously loud hum. Good for nothing.

So, that eliminates all of the Marshall transformers as source of the hum. Turned off the adjacent florescent light, no change.

I think this is just how this amp wants behave after that one rental for the client. It LIKES having that 60Hz tune and will sing it forever. I will no longer try to stop it!

Yes, I re-mounted the grounding lug at the Reverb Drive Xfmr mounting screw, and yes, the balance of the heater, both with the Marshall and the external supplies, be they the HP 712C or this latest TEI RS 503 supply....is good.

I hate admitting defeat, but there isn't any more time available for this one. I've lost my shirt and all patience with it, and DO wish I could hurl it, just to have SOME satisfaction of reprisal for the abuse I've received from it!

Replaced the input cap C1, grid leak resistor R1,, cathode and plate resistors R2 & R4 on V1A, coupling cap C7 & C39 ahead of V1B, cathode & plate resistors R3 & R7 on V1B, diode D1 across R3 at V1B. Moved the preamp ground connection point around to different chassis locations, as well as where Ground on that long winding path thru the PCB was picked off and taken to ground. Replaced the pair of buss caps C43 & C44, tubes V1, V2, V3. Re-soldered all the tube socket wire connections for V1 thru V3. And, all those moves made to find where the source is as posted prior to this one.

I'm baffled. By now, the source of the hum would have revealed itself. It starts with V1A stage and just gets intense after that, and the other JCM800 in our inventory doesn't. Two different animals. I like ours. It doesn't sing out of key, which 60Hz IS.

So, now putting the rest of it back together as I found it, while it has some fresh new parts, for all the good they did!