Twisted PT pairs

[pjd3]

Hey there, today (after a caffiene OD at Nero's here) I'll be beginning the wiring on Rob Robinettes BlackVibe 6L6 today. PT, powerswitch, standby, fuse holder, Rectifier and power tube socket (actually missing one socket, shit) and wanted to find out if there is any special considerations in making the twists in the pair, like how tight do they need to be, should they all be twisted in the same direction, are there any wire pairs that best be kept far away from each other, things like that.

One other thing, The OT transformer position track with the headphones doesn't sound like a bad idea, seeing as that this is not a completely classic layout i'm doing. Once the output transformer arrives can I still perform that "trick" with the PT mostly hooked up? I mean, with no tubes or rectifier tubes would that be like the PT output wires are floating?

Thanks everyone, I look forward to your comments and guidance. (and smart ass cracks, I know Ive earned them).

Philly

[R.G.]

Twisting wires is a way to reduce the area of the current loop between the wires, and to make the external field mostly cancel because the field is reversed relative to the axis of the twisted pair once each turn.

How well this works to reduce and cancel depends on frequency. It works really well at near-DC, which includes power line frequencies. At RF and especially high speed logic frequencies, it morphs into a twisted transmission line. The twist rate is critical for, say, ethernet cables, where you can muck with the transmission properties by untwisting enough to stitch it into a connector.

It's probably good enough to twist power transformer leads, period, without sweating the exact twist rate. It's probably most critical for the PT's primary and secondaries that get rectified. The secondaries that get rectified have large current pulses with lots of high frequency content running through them on alternate AC half cycles. Getting these twisted near one another will provide the most benefit from twisting.

Some people will quote cases where they didn't twist and it was just fine. Yep, sometimes the benefit is so slight that it's not noticeable; also, sometimes people are satisfied with whatever they get without twisting. This issue is heavily dependent on the construction details of every part of the chassis.

[pjd3]

Thank you for the generous responce RG.

I won't sweat bullets about it but will simply emulate many of the many pics and images of well done chassis wiring. Before I even turn it on I will most likely hand it over to Stan Day for a wellness check. If I've done anything out of sorts he would probably notice it and provide a remedy. I've build many things with a nearly 100% success on fire up but, this is my first tube amp and I take the stance that no matter how much I read articles, threads and Youtube vids my inexperience could or would lead to missing some things that experience would otherwise teach.

Thanks so much RG, feel much more confident now!

Phil

[xtian]

RG is smart on steroids.

In practical matters, we keep the PT and rectification way over on one side of the amp, and the guitar input and first amplification stages way over on the other side of the amp to avoid any noisy interference. Bottom line, unless you've got a very small chassis, or things are packed tightly together for some reason, there's no need to twist PT wires.

[pjd3]

Thanks for the response,

I would like to hear a word on whats happening with the 5 Volt filament and the B+ out from the GZ34. Am I seeing that correctly that the B+ is coming from pin 8 of the GZ34 rec tube, the same pin as one of the 5Volt filaments is connected too? Of course, I'm no expert in the inner functioning of a rectifier tube but, Its almost unnerving to see that. One reason is because there are the two 355vac lines coming in and the rectified voltage is coming from one pin that is coupled with one of the 5Volt filament wires, and there is no observance that I've seen anywhere of making sure that positive voltage going wires of the pairs are matched with the positive going wire of another PT pair. I have to take it that because it doesn't seem to have been mentioned anywhere, that it doesn't matter and we don't need to be concerned with that. Which I find amazing it so.

Yeah, if anyone could comment on that I'd be able to relax a tiny bit more.

Thank you!

Philly

[xtian]

Yes, is correct. Remember, 5v AC goes in, rectified DC comes out—they cohabitate happily on the same wire!

[pjd3]

Thanks xtian, I'm actually at the kitchen table moving through the beginnings of the wiring trying to make sure I don't create any "fatal situations". Thank you for your guidance, its very helpful for moving along. I'll have to find some time to research the workings of these mysifying Rectifier tubes!

danka1

Philly

[tubeswell]

^what xtian said^

The 5VAC coming off the rectifier winding would otherwise present as 'varying DC' riding on top of the rectified B+, but for the fact that it is smoothed out by the filter cap (along with the much bigger rectified 'pulses of DC' from the High Tension supply).

[pjd3]

Yup, since thats what filtering is pretty much about. And 5Volts is relatively small compared to the V that it's riding on.

Well all wired up like that now with the faith that thats exactly how its going to be!

Thanks xtian

Philly

[gingertube]

Read in one of the tube era text books - 3 to 5 twists per inch.
Less is not enough, more is a waste of time.
Cheers,
Ian

[xtian]

...and twisting heater wires is another common area of discussion. In my experience, heater noise (60Hz) bleeding into your audio path is only troublesome near the first couple of amplification stages, and there only if you're really sloppy and run signal wires parallel to heater wires. Don't obsess about it. OTOH, grounding scheme is critical.

[xtian]

...and orienting PT vs OT to minimize hum is also critical to low noise operation--and this has to happen early in the build, especially if you're cutting your own chassis. See "headphone trick".

That is all.

[martin manning]

Read in one of the tube era text books - 3 to 5 twists per inch.
Less is not enough, more is a waste of time.

3-5/in seems excessive. 1/in is fine. See: https://www.edn.com/design/components-a ... -crosstalk

[pompeiisneaks]

...and twisting heater wires is another common area of discussion. In my experience, heater noise (60Hz) bleeding into your audio path is only troublesome near the first couple of amplification stages, and there only if you're really sloppy and run signal wires parallel to heater wires. Don't obsess about it. OTOH, grounding scheme is critical.

I just finished that marshall build and did the heater wires as straight w/ no twists and it has no heater hum that I can tell. There's a touch of noise coming from the JCM channel but I can isolate that to an area I've not yet fully figured out, but it's not 'horribly' loud. The plexi channel is super clean comparatively, with the same heater setup.

~Phil

[pjd3]

I'm getting that if I get a least a few twists per inch, as straight
down into the tube socket as possible, plus keep the filament wires up higher and non-parallel to any signal wires I
stand to get better hum results.
Also, where the heater wires are higher up and coming straight down into the tube socket, the signal wires would more so be low and come off the side of the tube sockets. Thats what I see happening in many amp builds.

Phil