Mesa Stiletto 50Hz rumble

[Lockwood]

I have a problem with a Mesa Stiletto amp that I want to try to diagnose myself as I have been trying to educate myself on tube amp repairs. I know the safety issues involved and necessary precautions. I have a mechanical engineering background so limited electronics experience beyond a few fx pedal builds and amp mods but building up an understanding as I go.

Problem: Low rumble that can be heard and felt (speaker cab vibrates, things placed on the amp rattle. Its audible too but not very loud. The amp all works, although holding a note I can hear the 50Hz, like a fast tremolo. The amp seems to run very hot and the sound is rather compressed, also seems low on power for 100W. Checked with my simple hand held scope and saw 7.5Vac on the speaker output (speaker plugged in, no input signal) and a very clean sinusoidal output!! Used scope to check frequency and got 43.1Hz, did same on the mains power switch and also 43.1Hz, so its mains frequency hum. Changing modes eg bold/spongy, tube/Si rectifier. 50/100W changes amplitude but doesn’t eliminate it. Worst is Si, 100W, bold - as probably expected as that’s max. power setting. Its worse in 4ohm socket. Did the same check on my other tube amp and got a few mV noise - a factor of several hundred less. I guess all this lost power wound explain the hot running.

Checks so far:- Rumble is loudest with master vol at 0 and reduces as master is > ½ way. Connected from fx send to another amp – noise doesn’t transfer and preamp sounds good. Plugged cable into fx retn to disconnect preamp from power section - no change in noise. Guess that all means the problem is in the power amp? Checked ripple at the standby switch, after main filter caps - trace looks normal and 1.15Vac, didn't seem bad on 470V dc. Anyway I think noise would be 100Hz if it was power supply. Pulling phase inverter tube stops the noise dead. PI tube is new and trying another has no effect. If it’s all quiet with no PI, my interpretation is that power supply/EL34s (not replaced)/bias supply/tube sockets are not the cause as noise caused there would still appear at output transformer. So I think maybe I have narrowed the problem to somewhere around the effects return ½ tube, PI or negative feedback circuit. I wondered whether there could be some positive feedback going on? As I bought the amp used (I don’t think its ever been right) I can’t be certain that it’s never been modified or messed about with but all seems straight inside. All earths from board to chassis seem secure. Using stick to shift wires etc had no effect.

Next job to try to identify the components/signal path for this part of the circuit on the board (I have a schematic) but it will be tricky as the traces are small and disappear under components. If I need to I will take it to a tech but I would really like to figure this out myself and learn in the process so any ideas would be appreciated. Thank you.

[Jazz P Bass]

Try running the amp (idle) with only one EL34 installed.
Try all four tubes that way.
Maybe this will help identify a bad tube.

[stokes]

The fact that pulling the PI stops the noise indicates its coming from the preamp.You mention 50hz and 100hz,can I assume you are in Europe?Assuming that is the case 50 cycle hum would be associated with the heater supply.You describe it as being a rumble which,to me,is not normal heater supply hum.Perhaps it is causing an oscillation?Try pulling your preamp tubes one by one to see if you can narrow down the exact stage it is being introduced into the signal.

[Steve Conner]

Sounds like a bad preamp tube with a lot of heater to cathode leakage. Try pulling the preamp tubes as suggested.

[stokes]

Exactly where I was going,Steve.

[Lockwood]

Thank you for all the replies. I have spent some time on the amp tonight following your suggestions and I think moved things forward a bit.

1st new thing - on the 'clean' channel 1, I found that on fat clean and tite clean modes the hum is gone, nothing. Checked with the scope - in crunch mode on the same channel the sine wave is back at about 7Vac, but on cleans it drops instantly to around 10mV (similar to my other amp)!

Then followed the advice pulling preamp tubes starting at V1. Got all the preamp tubes V1 to V4 out leaving just the phase inverter and the hum is exactly the same in sound and voltage. Took the 2 tube rectifiers out - no difference. Took a pair of EL34s out - hum is reduced down to about 5V but still there. So its sitting there humming away with just the PI an 2 EL34s in. Tried swapping the pair of EL34s - no difference.

Took out the PI and checked voltages. 420Vdc on both plates, no ripple on the scope. No ac on the 2 grids. So whatever the cause of the hum is, it goes with the PI out.

The only thing that changes with the clean channel mode switch in the part of the circuit that is still working with just the PI in seems to be the presence circuit which has several relay contacts in it. One relay puts all the negative feedback through another 82K resistor which would reduce it a lot. I find it really hard to understand the switching circuit to work out what state the relays are in - there are over 20 relays in this thing - but looks like I need to try and also to identify the components of the presence circuit on the board????

Again feedback based on this new info would be gratefully received.

[stokes]

"Took out the PI and checked voltages. 420Vdc on both plates, no ripple on the scope. No ac on the 2 grids. So whatever the cause of the hum is, it goes with the PI out".
Check voltages with the tube in place.Did you try a known to be good tube in the PI position?Getting back to Steves reply about heater/cathode leakage.

[Dr Kaos]

A lot of Mesa amplifiers have tubes in the preamp operating as cathode followers. This puts their cathodes at a high voltage (150v approx) with respect to their heaters which can cause some valves to break down causing hums and other noises. If I remember correctly Mesa suggests putting Russian tubes in these positions rather than Chinese tubes (it may be the other way around).
Pulling the tubes out one at a time should reveal this.
Rereading the OPs reply I see he has already pulled all the preamp tubes before the PI and the hum continues. If it is not the PI tube itself then it must be something associated with its grids.

Dr Kaos

[g1]

Is this the correct version schematic for your amp? http://music-electronics-forum.com/a...e-stiletto.pdf

[Lockwood]

Yes, thats the same schematic that I have. I found it on the web but it seems to be a copy of the official Mesa document.

To reply to an earlier post the phase inverter tube is a balanced one selected for that position and only has a few hours on it (as do all the preamp tubes). The hum was there before I changed the tubes too. I have just put in another 12AX7 in that PI position (a new one that I have not used before) and the hum is not changed. So I am pretty certain that the PI tube itself is not at fault. I have not changed the EL34s but having run then as pairs and swapped the pairs and moved from inner to outter sockets with no effect again I think they must be ok. I did check idle current some time ago, cant remember the values but they were all very consistent.

Thanks again for all replies!

[Lockwood]

Well after a lot of headscratching I think I fixed it but would be interested in any opinions on what I found.If anyone is interested the schematic that g-one kindly posted above shows the components below. After measuring voltages all over the phase inverter and finding everything as expected other than the overpowering 50Hz hum all over the circuit.

I found that the negative feedback comes from the speaker jack through a 0.1uf cap then a 47K resistor, but after that a relay switches in or out another 82K resistor in series before the presence controls. The 82K is in series in the two clean modes - they didn't hum. So in clean total feedback resistor is 129K & no hum, in channel 1 crunch or channel 2 its 47K - hum. What I found after a lot of time identifying components on the board was that if I shorted out the 82K with a jumper, increasing the negative feedback, the hum appeared on the clean channels too. With that info I pulled one end of the 47K resistor and soldered in a 100K pot in series. In the modes that hum I could very consistently control the level of the hum with the pot. The hum stopped and disappeared on the scope when total resistance increased to 65K. Looking at a Marshall JCM 800 schematic this is very similar in the power amp but this amp has a 100K feedback resistor. So my theory then was - too much negative feedback in the Ch1 crunch and Ch2 modes.

Reading up on feedback I thought that the clean channel would benefit from a bit more of it to increase lows and highs and tighten up the clean a bit. So I replaced the 47K resistor with an 82K, close to the Marshall value and I replaced the 82K switched resistor with a 10K. So now there is 92K in clean (was 129K) and 82K in drive modes (was 47K), more or less the same. End result is really good - this amp has never sounded so good imho. The clean is more sparkling with a really nice shimmer to it and a more powerful bass whereas before it was a rather flubby clean sound. The drive modes seem fuller bodied and the horrible modulation that I could hear on the notes from the 50Hz is gone.

So all seems good - the only question in my mind is that when posting this problem on a Boogie site no other Stiletto users reported similar low hum/oscillation on the Stiletto, so it seems odd that the standard resistors gave too much negative feedback in my amp but not in others. Anyway the lower neg feedback in drive modes seems to have improved some of the over bright drive sounds that many Stiletto users complain about.

Thank you very much to those who replied on this problem.