Transformer hum?

[txstrat]

I wired it alongside the tag board from input up to the rectifier ground (gonna post some pics later).
I think there's still some room for improvement. Seems I haven't found the right spot for the bias, treble and bass pots and power tubes grounds yet - gonna put the bias ground to the OT ground and try to find the other spots by try and error.
You think the tone pots grounds would be better at the "PI star" or with the volume ground?

Matt

[Tom Phillips]

First review my post #55 above.
I’d then recommend that you connect the tone stack ground to the star defined by the pre-amp filter/decoupling cap. That would be closer to the “volume ground” rather than the PI ground. I experimented with bias ground point in my last build and did not find a dramatic change when I moved it around, I ended up connecting it with the power tube cathode ground.
The first filter cap ground should be connected to its own star with the first stage filter cap ground. Nothing else connects to that star (or sub-star since we are doing multi star) except the wire connecting to the next sub-star which is the screen supply.

The only signal ground to the chassis will be at the input jack in this arrangement. Therefore, it is OK that the input jack is not isolated. I did not isolate the speaker output ground in my last build but I will if I work on that amp again. If the output jack is isolated then the OT secondary + and - are still connected to the jack. That’s the power feed to the speaker. However, with the output isolated a ground lead must be run to the PI sub-star. That is needed to complete the feedback circuit.

Reference fig 12.10 in Merlin’s book and the galactic grounding discussion in TUT3.

Since you are using the big bus wire, make sure that the sub-star points are right next to each other and that the bus wire is isolated from the chassis. The last connection is to the input jack ground.

Hope that makes sense. Kinda hard to get it all into words but you have the figures in the books.

Cheers,
Tom

[txstrat]

Here's some pics and a new wiring. The broad blue dotted line is the bus wire. All the ground wires are dotted pink lines.
Only connection to the chassis is at (both) input jacks. No other chassis ground except the safety ground of the IEC socket, which is at a PT bolt still (will make up a separate chassis connection for this later).
I'd seen Merlins fig. 12.10. The difference in my build is that I connected several wires before they go to the respective star point (e.g. the cathode resistor grounds of the preamp).
I even doubt it is a star (or galaxy grounding) I have right now, since the bus wire runs alongside the terminal board and everything is connected to it one after the other. Even the first filter cap and the bridge rectifier ground, which is on the end of the bus wire.
Although the build develops way less hum (I'd better say buzz from now on) it still is dependent on the volume pot setting (least at about 4-5).
I'm gonna try to put separate wires to each star ground as good as I can, cause it's not easy to connect more than 4 wires to one terminal.
Also I'm poking in the dark, where to ground those 1ohms cathode resistors of the power tubes.

cheers
Matt

[Tom Phillips]

Matt,
Sounds like you have made good progress. Is the buzz with the volume control at 4-5 more than you would hear on old production amps? After you optimize the grounding you may want to experiment with lead dress to minimize noise pickup. Especially the low level signal lines connecting to the pots. Have you tried a shield plate cover on the chassis yet?

...I'd seen Merlins fig. 12.10. The difference in my build is that I connected several wires before they go to the respective star point (e.g. the cathode resistor grounds of the preamp). I even doubt it is a star (or galaxy grounding) I have right now, since the bus wire runs alongside the terminal board and everything is connected to it one after the other....

Good point. There are subtle things that can blur an intended star point. Merlin’s figure 12.9 and the associated text is good information that may help you locate a few things to improve.

Regards,
Tom

[txstrat]

Tom
With the volume on 4-5 I'd say it is absolutely normal if not better than normal. As soon as I turn the volume up or down it starts to buzz more than I would think is acceptable.
Haven't covered the chassis with a shield plate, yet but playing it in different rooms (different lights) made no difference.
Now that I have assistance from your posts I understand Merlins figure 12.9 better (I think point number 4 is exactly what is wrong with my build).
Next step would be to get all those rats nest wires out and configure a "real" star for every filter cap ground (aside from the looks it might be sounding better ). From my understanding I tend to think the different cathode resistors might need different ground wires to the star BECAUSE of the volume pot in between (and that's why the vol pot has a hum cancelling position).

Matt

[Tom Phillips]

...With the volume on 4-5 I'd say it is absolutely normal if not better than normal. As soon as I turn the volume up or down it starts to buzz more than I would think is acceptable....

Ahh ... Up or DOWN is worse. That very significant. It means that you have hum components both before and after the volume pot that are out of phase. 4-5 is the sweet spot of best cancellation.
You are on the right track with your other comments.

[txstrat]

It means that you have hum components both before and after the volume pot that are out of phase.

And that could be the cathodes resistors of triode 1 and 2.
See pic3.jpg in post63 Under the left (blue) filter cap is a yellow/green wire that leads to one resistor and from ther to the other... might be the culprit. I'll give each resistor it's own wire and report after that.

[txstrat]

Moved the grounding of the 5.6M resistor from input jack to the preamp cap ground. Hum seemed to be slightly less but I can't tell for sure.
Moved the tone pots grounds (2.2K resistor and .0047 cap) to the PI star ground to no avail.
I'm planning to separate the groundings for the preamp and power amp again to see what happens.

[tubeswell]

With the volume on 4-5 I'd say it is absolutely normal if not better than normal. As soon as I turn the volume up or down it starts to buzz more than I would think is acceptable.

And you are positive that you have thoroughly checked the switching on the input jacks and it is 100% okay? (Its just that I had exactly those symptoms once, and the input jack tip switches weren't connecting properly).

[txstrat]

Thanks tw.
I'll go into that again, very thoroughly. Does your amp have two inputs like mine?

[txstrat]

The input switches work good as far as I can see from moving the "switching tap" (the jacks are brand new).
Separated the preamp and PI grounds from the power amp. Preamp and PI grounds go to the input jacks. Power tubes cathode resistors (1ohm), bias supply ground and main/grids filter caps as well as the bridge recto ground go to a PT bolt. No change whatsoever.
I'm gonna buy some probes on monday and will boldly go and fire up that old oscilloscope to actually see where the hum comes up. I think that might shorten the troubleshooting way of "try and error".

[Tom Phillips]

Matt,

...I'm gonna buy some probes on monday and will boldly go and fire up that old oscilloscope to actually see where the hum comes up. I think that might shorten the troubleshooting way of "try and error".

OK but that is a very difficult method to see a hum component unless it is really large. It will be a learning experience though. Listen to the output while you are probing to make sure that connecting the probe does not introduce noise that swamps out what you are looking for.

...Separated the preamp and PI grounds from the power amp. Preamp and PI grounds go to the input jacks. Power tubes cathode resistors (1ohm), bias supply ground and main/grids filter caps as well as the bridge recto ground go to a PT bolt.

Did you fully implement the grounding set up we talked about in post #64 & #65? Note that the only intended connection to the chassis would have been at the input jack and the ONLY connections at the first power amp star would be the HT negative and the first filter cap neg. Then it’s a matter of daisy chaining the stars in the correct order as you work towards the input jack. The goal is to make sure that signals from different parts of the circuit do not share the same parallel ground path.

Keep in mind that hum signals can be coupled from one part of the circuit to another due to the layout. This could be between wiring, tubes, components or transformers. That has already been mentioned but I thought it was worth bringing up again. If that is going on in your amp then you could still have hum problems after the ground scheme is optimized. It’s also worth checking if the hum you have now is still 100Hz or if it is now 50Hz.

Hang in there.
Tom

[tubeswell]

Thanks tw.
I'll go into that again, very thoroughly. Does your amp have two inputs like mine?

Yep it had a 'Hi' and 'Lo' gain input. The tip swicth was just brushing the tip contact on one of the jacks, and when I brushed by finger against it (with the amp off), I could hear the vibration from the insecure contact. Shimmying the switch contact a bit with a screw driver/pliers fixed it. But it had the same symptoms as yours (i.e. the volt pot acted as a humdinger at around mid position)

[txstrat]

I don't know if it's worth a try but meanwhile I wonder if it would be a good idea to rebuild the amp with different preamp and PI tubes. Those 6SL seem to be VEEEEERY sensitive.
The Ampeg SVT's second channel has a structure similar to this built from a 12AX7 and a 6C4 (which is more or less like half a 12AU7). The arrangement of volume and tone control circuits are the same as in the B-15.
Would the internal structure of those tubes tend to less hum/buzz?

[txstrat]

...due to the layout. This could be between wiring, ... It’s also worth checking if the hum you have now is still 100Hz or if it is now 50Hz.
Hang in there.
Tom

Tom,
Im gonna try to move some wires to be sure the layout's OK. Will also check if it's still 100Hz buzz.
I'll come up with some more pics of the current grounding.

Matt