Problem with the high hum in a tube amplifier for a guitar

Hi,

I am building a 3ch 100w high gain guitar amp and I'm almost there..I think

All three channels has great sound and all the tone stacks works well.

Channel 1 is working perfectly..no hum at all!
Channel 2/3 got a 100hz hum..

The hum increase a bit when i turn the bass past 0.5 and change shape with the rest of the EQ..the hum also rise with the master volume..If I ground the grid of V2b (see schematic) the hum disappers and the amp is totally quiet.. If I ground the grid of V2a the hum increase and becomes "more square" sounding.. so my guess was that it could have something to do with the 15uf cathode cap at V2a which is switched in/out by RY1'a..I've tried to replace the 15uf cap and changing it to a lower value, but the hum still remain the same...it the same issue with both channels 2/3..

I don't think it's the input grid at v2a since it's shared with channel 1..which is quiet..

The hum is not affected by the gain pots at all..

The humming decrease a bit when the negative feedback is connected..but not much

Any suggestions how to solve this issue?

btw..All my voltage are close to the shown in the schematic..My PT has center tapped fillament (grounded at 220uf cap)..the heaters are tight twisted and runs underneath the pots and along the chassis' sides. I've tried swapping tubes..

Thanks

Best regards

Thomas

tube amplifier hum

Hi Thomas,
You seem to have what is called a hum loop. Have you taken all your ground wires back to a common ground point, this is usually referred to as the star ground point because all the ground wires return to the same point in a rough star formation., this is VERY important. They can have a small number of common points but make sure that the grounds that are carrying any significant current have there own path back to the star point which should be at or as close to the centre-tap of the transformer that you refer to as possible. It may look like a birds-nest but I trhink you will find that it will fix your problem. Your heater returns especially should not share an return path with anything to do with the signal paths, they should all go back to their own ground points that are all linked together as close as possible to the power supply.

Route any leads carrying low level audio well away from the mains transformer, valves as they are known in the uk or vacuum tubes elsewhere are high impedance devices, which means that they can very easily pick up extraneous noise. You can use screened cable for the signal leads if you wish, try to only ground it at one end. Not everyone will agree with the use of screened leads, you can try using twisted pairs for your signal leads if you prefer, I normally use screened cable for low level audio, the capacitance effects should not be enough to affect the audio response within the hearing range of most humans unless there is something really wierd with the load it feeds into, but not everyone agrees with me on that one.

Sorry if all this is all second nature to you but it is very easy to get it wrong and experience the effect that you describe, most of us have been there snf done that.


Bob.

tube amp humming

Hi Bob

I'm sure you're right about that this is a "classic" ground loop failure..

I'll check all my ground connections and see if I can find anything wrong..

The amp is picking up hum between V2a and V2b..so I guess thats where I'll begin..

I don't think that the heaters is interfering with the signals..my scope shows sawtooth signal..it's DC, right???

But thanks for your help..I'll report my findings..

Best regards

Thomas

tube amp ground loops

Hi Thomas

I don't think that the heaters is interfering with the signals..my scope shows sawtooth signal..it's DC, right???

Click to expand...

DC is only on the heater of V1 if it is wired up as per the circuit diagram you put for download, the rest of the heaters are on the AC side of the bridge rectifier! That could be part of your problem. You did notice the values of the 4 low voltage caps on the +/- 3V supplies these are 10KµF 10V (an unusual unit today but common enough in the past) that is 10 000µF 10V working. You really should not see much saw tooth with those values fitted unless you are seeing the amplified hum signal with your scope - I dont know where you are probing. There is one in the bias supply with a K in the range too - just in case you missed it! If you have got the heaters all on the DC side, the load will be too much for the value of smoothing caps fitted to handle but this would not explain how channel 1 only is OK. (The bridge rectifier would be overloaded in this case too and would have almost certainly burned out, so I do not think you have done this!)

You also said that the hum is 100Hz, that is what you get from a bridge rectifier! on 50Hz mains.

Thanks for publishing the circuit incidentally, it looks like a nice one.

Best regards
Bob.

tube amp dc filament bsupply

Thanks Bob! ..and sorry for giving wrong informations!:|

I am aware of the kv caps and so on ..but here is something new:

It's not the 15uf at V2a cathode that is used on channel 2/3 but the 1uf! anyway..I've tried to elevate the heater supply via a voltage divider to give the heaters dc offset..but it didnt helped at all! It's not the heaters and not AC hum. I scope at the speaker out (see attached drawing)..

My next guess would be that it could be a leaking coupling 0.022uf cap between RY1'b (no) and the 475k V2b grid resistor. As i wrote before..if I ground the V2b grid (pin7) the hum totally disapper..which eliminate the EQ 'ground loop' issue, right?? If I ground the V2a grid, the hum is still there.

Don't want to just change the 0.022 coupling cap to see if it is that before I've got your oppinion, - it's a large PCB and I am affraid of ruin it.

Thanks very much for your time and help Bob

Thomas

hum tube amp

Another simple test would be in removing V2a to check, if the hum is amplified by V2a or coupled to the anode circuit and RY1b. Without knowing he circuit topology, it's impossible to see favorite interference locations. RY1b would be a candidate, also the D voltage supply could inject the hum. It's present at V2b anode also, but with reduced sensitivity. Can you tell what's the hum fundamental frequency respectively add a time scale to the waveform?

Where is RY1b mounted? I also see, that the +/- 3V supply has capacitive ground connection, so no worth mentioning AC voltage referred to ground should be present at the relay coil? Depending on the relay type, the coil may have a coupling to signal contacts.

tube amp hum problem

Hi Fvm!

Pulling V2? wouldn't it be the same as grounding the grids? If you look at the schem Channel 1/2/3 shares the first half of the V2 tube and channel 2/3 shares the second half of the V2 tube. If I ground V2b grid (shared by ch2/3) I've got no hum. If I ground the V2a grid (Shared by all channels) I've got hum! but only at channel 2 or 3, not channel 1 :s

Thanks

Thomas

dual rectifier 3ch (thomasdj)

You're right, I didn't consider that V2a and V2b are a double tube. The idea was to switch off V2a operation (Ia = 0), this coulc be done by a negative grid voltage also. But I rather expect the coupling to the anode net.

tube amp power supply hum

Yes..I will try to replace the 0.022cap and see if it helps..

Thanks

Thomas

Added after 51 minutes:

Ok..replaced the cap..didn't helped at all ..running out of ideas

Help please..

Thomas

2 amp ground hum

I wouldn't particularly expect defective parts. But you should systematicly check for possible coupling of interfering signals to the anode V2a - grid V2b nets. This may be solder or PCB shorts, as well as capacitive coupling of nearby AC voltage and inductive coupling of AC currents.

I'm not sure, if the previously discussed ground-loop topic has been finally cleared. Assuming the hum wouldn't be sourced by V2a (in case of any doubt, it could be disabled as suggested), the next possible source is D anode voltage. V2b input is effectively sensing the "D" AC voltage component. Assuming, D filtering is sufficient, then the ground terminal of the resepctive bypass capacitors is sourcing the voltage. If it has a AC voltage difference to the ground of V2b cathode network, then this voltage would be directly amplified by V2b.

To my opinion, the possible interference sources can be checked with linited effort, mostly without desoldering parts, rather by measuring differential AC voltages, applying shorts to different ground potentials and similar. Also RY2 can be utilized to cut nets.l

hum in tube amp

I'm not sure if I understand it right..

'I've shown my test points at the attached schem.. if i ground at the 'red' points I've got no hum! This eliminate that the V2b is the problem, right

if I ground the 'green' point I get hum..but not on channel 1!

I've tried to change the 0.022uf coupling cap and the 1uf cathode cap..

Could it be that I am so unlucky that the 1uf V2a cathode cap, I've replaced, also is defect?? (And again, the 1uf V2a cathode cap is also used in channel 1..) Both the 15uf and the 1uf cathode are physical near the D voltage, but again, no hum at channel 1..

Thanks

Thomas

tube amp grounding

Basically, a defective V2a cathode capacitor can't cause hum, to my opinion.

As you reported "no hum on channel 1", V2a can be possibly excluded, if both channels have almost similar sensitivity, which I don't know.

If channel 1 is selected, is the hum still present at V3b cathode (when measured with an oscilloscope)? When shorting V2b grid to the ground at HV supply, particularly "D" bypass capacitors (by a wire), does the hum still disappear?

tube amp ground loop

I've got no hum at ch1 (neither with the scope)..

I've soldered a wire to the HV 10uf filter cap "D" and shorted to V2b grid, no hum!

Ch1 is the "Clean" sounding channel and channel 2/3 is distorded channels..

I am not educated in ellectronics or have been teached..totally learning by doing..so sorry if I'm a bit slow

I've attached a part of the layout around V2..see the yellow traces and the pink components..

Thanks for your help

Thomas

tube amp hum ground

During highschool, I've been repairing guitar amps and TVs without having learned electronics, that's possible indeed.

I expect a higher gain for the distorted channel, so there could be misinterpretation of results.

Another test could be by operating channel 2 or 3 and keeping RY1 demagnetized (contacts in N.C. position), so V2b input, starting at RY1 N.O. contact would be open circuit. What about the hum then?

P.S.: I see from the PCB, that there are limited interference possibilities in the circuit. I would like to suggest additional tests.

- disabling V2a. I suggested this before, but applying a negative grid voltage may be possibly dangerous to the FET switch at the grid, so alternatively, the cathode can be pulled up to 10 or 15 volts. The cathode resistor can bear the current.
If this action stops the hum, it would be at least partially related to V2a operation.

- a somewhat exotic cause could be from high frequency oscillations of V2/V3 (most likely in the SW region). But it should be detectable with an oscilloscope and also indirectly by a V2v/V3 anode voltage sensitive to touching it's circuit with a probe or screwdriver.

ac hum in tube amp

hehe ok!

I decoupled the +-3volt supply to RY1' (guess that is the relay you ment and not RY1) so the input is right at "normaly open" at RY1'. Dead silence!

That exclude indeed ground loops and so on..glad for that

What do you think? - the only parts left is the 100ohm res and 1uf cap at V2a, right?

Thanks

Thomas

Added after 2 minutes:

Hehe..sorry...but the last part you wrote (PS) I partially don't understand..simply not good enough, I guess

rpoblem with guitar amp hum

The hum must be expected to come either through V2a (respectively be amplified by V2a) or from the anode resistor.

The disable V2a check suggests to connect V2a cathode (V2 pin3) to the +15V supply or to use an external power supply with 10 to 15 V. This effectively shuts off V2a anode current.

valve amp hum clean channel

OK..and this won't harm the fet?? because I can't replace the fet..:|

And can i connect the+15v without disconnecting the 1uf/15uf caps..?

Thanks

low frequency hum tube amp

It only increases a positive cathode voltage present in normal operation and leaves the gate voltage at zero.
-Yes, provided they have 15 V voltage rating.
An additional test is by shorting the 100k V2a anode resistor.

removing tube amp hum

ok

15v on pin 3 killed the hum totally!

what do I have to do next??

Thanks very much!

Thomas

wrong capacitor hum tube amp

I'm still somewhat surprized. You're sure, that the 15V wasn't shorted by the test, as this would possibly disable other circuit parts?

From the PCB topology, I don't see other possible hum sources than the filament supply. Would it be possible to move V2 to a DC filament supply either parallel to V1 or from a laboratory supply? I assume, that you have the filament supply through additional wires, not shown in the PCB layout.

If the filament supply would be the hum source, I wonder how the original amp avoids it?