Transformer hum?

[Tom Phillips]

I'm off to work now but it looks to me that you have close to what is known as a "random" grounding scheme.
Could just be one or two connection points causing most of your problem.
Use your tuned guitar to evaluate the hum.
Open low E is ~80 Hz. Move up 2 or 3 frets and if a slight re-tune matches your hum then it is 100 Hz.

[txstrat]

Tom,

thank you so much for your patience.
The hum IS 100 Hz (matches the 3rd fret on spot).
What makes me wonder is the fact that I've built exactly 12 amps (from scratch) which ALL have the same grounding and are dead quiet (all of them being Fender and Marshall copies except one). It took me several years to find out the hard way which grounding scheme to use but once I found out which worked best all of the amps are dead quiet. I'm using the buss wire as a substitute for the brass plate. Main and screen filters go to a PT bolt as well as the bias ground. Only odd ground would be the OT ground at the speaker jacks IMHO.
Any ideas?

[Tom Phillips]

Frustrating - yes. But you have eliminated some things. I think we can safely say that the main problem is not "Transformer hum" and its not heater hum. The grounding scheme you have described should work adequately. However, there may be an unintended connection. I don't think the OT ground location is the problem. Although these grounding schemes are not ideal, they have historically worked well enough for low gain vintage designs. If you pull the pre-amp tube does the hum go away?

I like to use multiple star grounds myself that are daisy chained together in series from the pre-amp to the power amp. I'll keep thinking about it but I don't have a specific recommendation to give you right now. I searched but could not find the extensive grounding discussion that I read about a year ago on Ampage.

[tubeswell]

Most hum when the vol pot is all the way down and when it's over 6. Between 4 and 5 the hum is least.

Aha! Check your input jack(s) tip-to-ground switches are functioning properly. They should (both) make solid contact without plugs in there.

[txstrat]

If you pull the pre-amp tube does the hum go away?

When I pull the first tube (preamp) the hum is reduced to about 50%. When I pull the PI tube the hum's gone.
The multiple star ground is what I haven't tried yet. Sounds interesting.

Aha! Check your input jack(s) tip-to-ground switches are functioning properly. They should (both) make solid contact without plugs in there.

They do, thanks TW
Funny thing is, the schematic shows only one of them switches, on the bright input. (have tried both ways)

[txstrat]

This is preying on my mind.
I just had a look into KOC's TUT3. There is a scheme for a B15 (not the fixed bias version but anyway) and I found the ground has only one spot to enter the chassis, close to the IEC socket. I've tried the one spot theory but I think I put the ground to the input jacks (can't remember exactly since I've done so many things to the amp already - maybe I should start making notes). I'll keep that in mind and might try it later.

[txstrat]

What I haven't tried yet is separating the preamp ground from the PI ground.
I've read some articles about star grounding and with them came some information about the currents in the PI and OT section (including NFB).
I've made a new version of my layout, which I'll try next. Well, ...after christmas (not to annoy the wife). Maybe I find some time anyway

[Tom Phillips]

Oh. You have TUT3. The grounding scheme described as "Galactic ground" is the same thing as multiple star ground.
The wiring you show in your post 52 is closer to that but it would be best if you do NOT connect the multiple stars back to the ground by the PT.
When you get a chance to study TUT3 further you will see that the various star points are daisy chained from lowest level stage to highest.

As a sanity check it would be good to verify that the B+ is clean to the pre-amp and PI. The ripple should be less than .01% maybe as low as .001%.


Regards,
Tom

[txstrat]

I finally found the time to tinker around a little with what I intended.
Took the DC heaters out again and put the PI to the star ground. The hum is less then 30% as it was before now and it doesn't increase with the volume pot anymore.
Think I'm gonna study TUT3 now for the daisy chained solution. BTW my input jacks are still not isolated from the chassis.

[Tom Phillips]

Glad to hear that you are making progress. It appears that your last changes corrected at least one problem ground loop.

Keep in mind that there are many amps, such as the vintage Fenders, that use less than perfect grounding but still gave adequate performance. They just didn't have any ground loops that caused major problems. It was good enough for the time and they went to production that way. However, it's nice when you can improve on the design when you are doing your own build.

I have worked on a 1960's Ampeg Atlas head that basically used the "galactic" ground scheme as described in TUT 3. Positive proof that some designers of yesteryear had the knowledge and went the extra mile.

You don't absolutely need to isolate the input jack ground from the chassis unless you are going to install a ground lift switch. In my last build I daisy chained multiple star ground points from the power amp back to the pre-amp with the only circuit chassis ground at the input jack. It worked very well and was not much extra work. I added a few extra terminal strips so that I could avoid soldering any parts to the back of the pots.

There is also good information in Chapter 12 of Designing Tube Preamps for Guitar and Bass by Merlin Blencowe. As in TUT3 it is not new information but, IMO, the figures are better and the presentation is clearer in some respects than in TUT3. Both books are excellent. I find that it's sometimes helpful to read the same information from the perspective of a different writer.

Regards,
Tom

[txstrat]

I was thinking more than once to order Merlins book, although I own a lot of tube amp related work.
This might be the right moment to order my copy. Thanks for the hint.

[Tom Phillips]

I have many technical books and I continue to add to the library. Sometimes a light bulb goes on when I read the same information from the perspective of a different writer. Often I get a better grasp of a subject when I review a book I have owned for years. Just missed it the first time around or didn't have the background information to help me understand the new concept. I figure that it's worth it if I only get one good new idea from a book. The whole learning process is very synergistic for me. I'm glad we can buy these new books. They are not collectors items (yet) and I can feel free to mark them up and add my own notes as I read.

Tom

[txstrat]

Tom
You express exactly what I think. That's why I've ordered Merlins book yesterday.
I think the reduced hum can be traced to the filter caps grounds of the preamp and PI which I had tied together.
Now I'm thinking of pulling the buss wire (from the back of the pots) and wire the ground connections of the pots (via 22k resistor and .0047 cap) to the preamp filter ground.

[txstrat]

Tom
Got the book yesterday. Great stuff. Very understandibly written. Can't wait to get through it completely.
I rewired the amp to galaxy grounds, where the preamp, the PI and the bias/power stage each share a star. The hum is less than before, even with the old transformer. What I haven't done yet is to put a wire from the OT ground to the buss wire (at the PI star). Oh and I grounded the buss wire at the input jacks as you proposed.
Thanks for your encouragement.

Matt

[Tom Phillips]

Now I'm thinking of pulling the buss wire (from the back of the pots) and wire the ground connections of the pots (via 22k resistor and .0047 cap) to the preamp filter ground.

Hi Matt,
Glad the performance is improving. What did you end up doing with the buss wire?
Tom