Tube supply question

[Phoenix]

Ok, should definitely not have 50Hz ripple on that if all diodes in the bridge are functional. Have you double checked it's wired exactly as drawn there?

[rankot]

I did, but I will check once again in the morning.

[PRR]

[rankot]

I actually have those filter resistors and capacitors for preamp tube stage, but I didn't show them here, because I have a problem with rectification

[Rob Strand]

I have a problem with rectification

Try deliberately disconnecting one side of the transformer from the rectifier then the other.  Measure the DC voltage accurately in each case.

This will show any imbalance in the winding + rectifier voltages.   If you have a bad connection, one of these cases might not even work.

[rankot]

When I disconnect one side of transformer, I get +53.9V and -54.8V.

When I connect that side back and disconnect another one, I get +54.1V and - 55.1V.

Also, both coils have almost the same resistance (measured when disconnected) of abou 1.6-1.7 ohm.

I will try to replace bridge with a new one.


I have another problem that I didn't previously notice too much because of hum - there is some HF noise getting into the preamp through the guitar cable. When I touch the cable (even outside isolation) it gets stronger, and when I touch strings, it is even stronger. Turning treble pot makes it stronger or weaker. I tried to parallel 1M input pulldown resistor with a small capacitor (100p) but nothing changed.

[Rob Strand]

When I touch the cable (even outside isolation) it gets stronger, and when I touch strings, it is even stronger. Turning treble pot makes it stronger or weaker. I tried to parallel 1M input pulldown resistor with a small capacitor (100p) but nothing changed.

Try a 10nF to 100nF connected from the input socket ground to the chassis at the input socket.

Do you have your ground network connected up yet? (the one with the cap + 10ohm + diodes)

[rankot]

Ok, should definitely not have 50Hz ripple on that if all diodes in the bridge are functional. Have you double checked it's wired exactly as drawn there?

Is it possible that something else is making those ripples? Dead output transistor, or something like that?

[Rob Strand]

Is it possible that something else is making those ripples? Dead output transistor, or something like that?

It's unlikely for an amplifier circuit to do that.

You could disconnect the amplifier and preamplifer from the power supply, place a dummy load on the power rails then measure the ripple.  Ideally you would put loads equal the amplifier's current draw on each rail but any reasonable load should be OK.   Drawing more current might help also, because ...

From your measurements the rails differ by 0.2V to 0.3V, so it is possible if the load doesn't draw enough current the slight imbalance in the voltages will cause the ripple to be 50Hz.  When the load is increased to the point where the ripple exceeds say 0.6V you should see the other winding suppyling power and the ripple might return to 100Hz.    The rectifier voltage drops might adding the problem but the measurements you gave a while ago looked OK - when you have weird problems it doesn't hurt to experiment.

[rankot]

When I touch the cable (even outside isolation) it gets stronger, and when I touch strings, it is even stronger. Turning treble pot makes it stronger or weaker. I tried to parallel 1M input pulldown resistor with a small capacitor (100p) but nothing changed.

Try a 10nF to 100nF connected from the input socket ground to the chassis at the input socket.

Do you have your ground network connected up yet? (the one with the cap + 10ohm + diodes)

Yes, I have built and installed ground network (just like in Merlin's script), but it seems to do nothing. Noise's the same regardless I connect chassis through it or directly

And to clarify this with cable, I didn't wrote well - noise is bigger when I touch jack (or guitar strings), it doesn't change when I touch cable isolation.

[Rob Strand]

And to clarify this with cable, I didn't wrote well - noise is bigger when I touch jack (or guitar strings), it doesn't change when I touch cable isolation.

If it was only when touching the strings I'd suspect the grounding of the bridge in the guitar.  However, since it also occurs when touching the jack it's not that at all.  In fact it's quite a tricky problem.

At this point I'd probably take a few steps back in order to work out where the problem lies. See what happens with these, do them one at a time maybe not together.   Don't worry about hum, focus on the buzz.
- Replacing the ground network with a short.
- Replace the input jack with a non-isolating type.

Another thing to try is to solder a ground wire to each pot then try connect that wire to the chassis, then try connecting it to the circuit's 0V.    Ungrounded  pot casings can causes weird noise issues.

One thing I"m not 100% clear on:  What other jacks are there?   Are they all isolated types?   (don't forget to include speaker jacks)

[thermionix]

Quite possible the amp's input jack is wired incorrectly.

[rankot]

- Replacing the ground network with a short.
- Replace the input jack with a non-isolating type.

Another thing to try is to solder a ground wire to each pot then try connect that wire to the chassis, then try connecting it to the circuit's 0V.    Ungrounded  pot casings can causes weird noise issues.

One thing I"m not 100% clear on:  What other jacks are there?   Are they all isolated types?   (don't forget to include speaker jacks)

I have already tried to short ground network and it didn't help.

Since all jacks are isolated (there's no speaker jack, it's cable is soldered to PCB), I will try to restore non-isolated one for input and see what happens. I will also try with grounding pot casings.

[rankot]

OK, here is what I've done:
- added 56n cap between jack ground and amp PCB ground (can't solder it to front panel, it's made of aluminium and solder doesn't stick to it);
- added 100p cap between input and ground, just before first stage's grid;
- tried with and without hum breaker.

After that:
- Hum breaker works OK, I have double checked and it definitely kills most of the hum.
- There is still hum that I produce when I touch strings or jack, but this hum is very low when hum breaker is connected; what is important with this hum is a fact that when I touch stings for the first time, there's a loud pop, and when I play it doesn't pop again, but if I leave a guitar for a few seconds and touch strings again, it pops again. The trouble is that I want to rock, not to pop
- No more radio.
- I have killed too much of treble with that 100p cap, I will try without it or with smaller value.

I have also noticed that amplification is not as strong as before and the sound is distorted. Probably that 100p cap too.

This is schematic of input section as it is right now:

[Rob Strand]

added 56n cap between jack ground and amp PCB ground (can't solder it to front panel, it's made of aluminium and solder doesn't stick to it);

Normally you would use a large ground lug.  The input socket thread passes through the hole.  Not these exact ones, but ones that look like this,
http://www.guitarpartsresource.com/images/large/EP-4968-000.jpg

- No more radio.
- I have killed too much of treble with that 100p cap, I will try without it or with smaller value.
I have also noticed that amplification is not as strong as before and the sound is distorted. Probably that 100p cap too.

Agreed.  Maybe 47pF.   Try to work out the largest value before you start losing treble.  Hopefully that's big enough to remove the radio.
The tube has some input capacitance (not sure what it is for that tube) but it is probably around 100pF.
With 2x68k's on the input that already has some HF roll-off due to the tube capacitance.

what is important with this hum is a fact that when I touch stings for the first time, there's a loud pop, and when I play it doesn't pop again, but if I leave a guitar for a few seconds and touch strings again, it pops again.

Yes, that is probably the most important thing to fix.   It is not normal and it points to a problem.
Sounds like a capacitance is charging up.  Exactly what capacitance and how, I'm not sure at the moment.

A broken ground in your guitar cable or somewhere in the amp might be the cause.   With the amplifier turned off, use your multimeter and check continuity from the socket on the guitar to various ground points in the amplifier.

How confident are you the ground on your mains socket is OK?

I have also noticed that amplification is not as strong as before and the sound is distorted. Probably that 100p cap too.

That's a bit strange too.
If you have strong RF or something is oscillating you might see this type of thing.  In fact oscillation could cause the weird pop behaviour as well.  Can you monitor the output of your power amp and various points in the preamp with an oscilloscope?  Use a x10 probe.   Unfortunately when you connect the CRO ground, or the CRO probe, it changes the circuit sometimes even stopping the oscillation!

[amptramp]

The ECH83 characteristics are here:

http://www.mif.pg.gda.pl/homepages/frank/sheets/010/e/ECH83.pdf

The input capacitance of the hexode section from Grid 3 to everything else is 6.0 pF and from Grid 1 to everything else is 4.8 pF.  This is not going to have much effect on the high frequency rolloff.

[thermionix]

With any guitar/amp that I'm aware of, there should be less hum when you touch the strings or other grounded metal part.  Getting more hum suggests somewhere the "hot" and "shield" are reversed.  The pop IMHO is from static on your body.  It is winter after all.  Something like that will be exaggerated with the "hot ground" issue.

[Rob Strand]

The input capacitance of the hexode section from Grid 3 to everything else is 6.0 pF and from Grid 1 to everything else is 4.8 pF.  This is not going to have much effect on the high frequency rolloff.

Thanks.  The Miller effect is very low then, so we only have to deal with the 4.8pF.
That's means adding 100pF is about right.  Right in the sense of emulating the input capacitance of a 12AX7.
The HF roll-off should be noticeable but not extreme.

[rankot]

I was too busy with regular job those days, now I build separate PS for this preamp. I decided to debug it outside amp first, and when it's completely fine on its own, to try to put it inside.

[rankot]

I'm back on this baby! I've built a separate power supply, and it is working almost fine. Almost means that there is hum sometimes, sometimes not! PS is supplying 9V DC to my preamp, and that is later used for heater supply (regulated with 7806) and for ±15V and 50V generated with LTC1054 based boost DC/DC converter. The most interesting thing is the fact that if I plug my preamp together with amp, and at the same time, hum is annoying. If I unplug preamp, then plug in again, hum gets reduced. HOW TO HELL IS THIS POSSIBLE?

When I put my oscilloscope to 9V DC, it shows that there is a small (±20mV) sine wave still present after regulator, but I presume it is acceptable value.

Besides that, I have tried to add bypass capacitor parallel to screen resistor and it just produced distortion, so I removed it. I have also put two 100p bypass capacitors to ground behind pentode and triode and that completely killed all the radio noise I had.