B+ cap location and wire length

[pjd3]

Hello,

The B+ (1st filter cap to OT primary CT) in my current build runs a tad longer than typical. I was curious if this was relatively OK. From the diode pair (rectifier) its about 4-5" to the 1st filter cap then another 5-6" to the OT primary CT.

Ya think we are looking OK here? Or is there something I could improve?

Thank you,

Phil D

[pdf64]

On one hand, the wires between rectifier and reservoir cap tend to be the noisiest in the amp.
On the other hand, classic Fenders use long wire runs to the doghouse, without it being an issue.
Whereas the wire run between reservoir and OT CT, whilst carrying high signal current when the amp is putting out high power, shouldn't cause a problem if you use good lead dress. Where feasible, twist all 3 primary wires to reduce the magnetic field radiated.

[R.G.]

pdf64 is right.

If the wire from the transformer CT goes directly to the first filter cap negative, the length and resistance of the wire doesn't much matter as long as the wire is routed in a way that doesn't radiate magnetic interference. Twisting the three transformer secondary wires together for as much of their length as possible is the easiest way to reduce M-field emission.

Since it's the currents that matter for magnetic field emission, twisting the CT wire and both end wires from the transformer to the rectifier(s), then twisting the output wire from the rectifier with the CT wire on its way to the first filter cap reduces emissions in that path. In both cases, the current pulses are equal-and-opposite in the twisted wires and the current loop has the minimum area.

[pjd3]

Thank you for your input here.

This amp has a row of a 3" terminal strip that happened to be same lenght as the PT mounting length. With that, I can wire the 2 diodes to the HT coil right at the PT so those HT coil wires don't need to be any longer than a couple of inches long, which I thought would be great. Would it still be helpful to twist the PT CT with the diode rectified 120Hz pulsed AC wire to the first filter cap? That would be easy to do. I realize that there is no longer an "equal and opposite current going on but, it might at least be physically tidy as long as its electronically acceptable.

Oh just to mention - you both has contributed to a thread I began in this stereo amp and to your recommendation Its beginning with fully isolated input and speaker jacks plus I just ordered a bag full of solderable rings for soldering the 100pf caps from input ground to chassis for RF protection - something I wouldn't have ever thought of !

Thank you RG and pdf64!

Best,

Phil

[R.G.]

This amp has a row of a 3" terminal strip that happened to be same lenght as the PT mounting length. With that, I can wire the 2 diodes to the HT coil right at the PT so those HT coil wires don't need to be any longer than a couple of inches long, which I thought would be great. Would it still be helpful to twist the PT CT with the diode rectified 120Hz pulsed AC wire to the first filter cap? That would be easy to do. I realize that there is no longer an "equal and opposite current going on but, it might at least be physically tidy as long as its electronically acceptable.

Glad I could help. Yes, twist/pair the wire from the joined diode cathodes and the transformer CT going to the first filter cap. The essence of avoiding magnetic field interference is knowing where the currents flow. For a full wave center tap, each half of the primary conducts for one of the AC half cycles, alternately as forced by the diodes. The CT wire conducts each half cycle, returning the current to the transformer windings. The reason for twisting both halves of the primary winding with the CT is that since the primary end wires conduct alternately, together they make for one pulse for each half cycle and the CT wire then offsets them alternately. The three wires make for one effective send-receive group.
After the diodes, there is a single send wire which conducts at the same time as the CT return, so they complement each other in a single pair.

It >>IS<< still an equal and opposite pair. It's the half-primaries that are funny, needing both sides of the primary to complement the CT wire.

It's hard to get used to thinking of where the current is traveling. Someday I hope to get there.

[trobbins]

Imho it's worthwhile to also appreciate the current loop that forms from the main reservoir capacitor terminals, out to the output transformer primary winding B+ terminal, and back through the output stage valves to a cathode end junction that returns to the reservoir capacitor negative. That 'loop' is often not optimised to reduce loop area, because there are many segments to the loop that can be spread out for various reasons. As indicated, that loop does circulate relatively large current levels, and with a wide range of frequencies, so could easily couple into sensitive circuitry if nearby. The peak of that loop signal current is certainly not as high as the current of a rectifier loop which can be very peaky, and also only gets large with signal level so is somewhat masked from being obtrusive if it does affect other circuitry. It's likely impractical to closely couple/twist any output stage signal loop wiring (ie. outgoing wire with its related return wire), but it is likely that the 'out and back' wiring can be located close to each other with a bit of care and attenuation to layout.

[pjd3]

Ah OK RG, thank you.

For some erroneous reason I was imagining the HT coil CT as a neutral of sorts but sure, it has current on each half cycle. In that case it makes good sense to twist/pair it with the diode joined coil endings.

Thank you trobbins. I will keep a good eye on the where abouts of the B+ going to the OT CT and the plate wires, and the whole of that loop. I'm pretty confident I'm able to keep them at a fair distance from any of the signal input area, at least say, 4-5" of distance. I made it a point to keep the sensitive input area front and center, with as much of the high current/voltage circuits pushed to the far ends of the rear corners. It was tough but, hopefully I managed to make sufficient proximity.

Best, and thanks,
Phil D.

[R.G.]

You two are messing about with an idea I was trying to get out in the "Guitar Wiring Notes" article. I put it as the idea of domains - a certain volume for power supply wiring, another for input and signal amplifying wiring an another for the output stage. It impractical to do this strictly for guitar amps, but it's a good goal - just keep the noisy stuff in its own volume and don't intermix stuff any more than you absolutely have to.

[R.G.]

Hmmm. I probably should put the three-wires idea into the article as an illustration of it being the currents and conductor arrangements that matters.

This may have exceeded my drafting skills...

[R.G.]

Edit: Please stand by. The forum seems to not like png images. mmmm
apparently it doesn't like jpg either. working...
Can't get the forum to display it. Click the link.

Let me know how this works:

https://www.dropbox.com/scl/fi/s5hg6hso ... 4rs84&dl=0

The little colored footballs (also know as charge pulses) get allowed through the diodes on alternate half-cycles. They combine in the wires to the first cap. Since the diodes force them to happen only on alternating half cycles, twisting the two transformer primary wires and the CT, all three, together means that the CT wire is always right next to a wire that has equal-and-opposite pulses headed the opposite way, so the magnetic fields cancel, as in the dotted ellipses.

Whew.

[martin manning]

Edit: Please stand by. The forum seems to not like png images. mmmm
apparently it doesn't like jpg either. working...
Can't get the forum to display it. Click the link.

Either of those types have worked fine for me.

[sluckey]

I just posted a png file at 9:00am.

[pjd3]

Yup,

Good illustrative image - and shows how and why that twisting works for post diode pair as well. Thank you.

Phil D

[GAStan]

GREAT illustration RG, it made the light bulb in my head turn on brightly.

Thank you for this and all else you share so freely.

[didit]

Hello R.J. --

Edit: Please stand by. The forum seems to not like png images. mmmm
apparently it doesn't like jpg either. working...
Can't get the forum to display it. Click the link.

Let me know how this works:

https://www.dropbox.com/scl/fi/s5hg6hso ... 4rs84&dl=0

The little colored footballs (also know as charge pulses) get allowed through the diodes on alternate half-cycles. They combine in the wires to the first cap. Since the diodes force them to happen only on alternating half cycles, twisting the two transformer primary wires and the CT, all three, together means that the CT wire is always right next to a wire that has equal-and-opposite pulses headed the opposite way, so the magnetic fields cancel, as in the dotted ellipses.

Whew.

This seems to work here at my screen. Needed to download to my filesystem and then upload as an attachment, and then select "place inline". The page builder somehow used embedded metadata to insert and display correctly. Not the most intuitive UI by far.

XfdHskbi.jpeg

Excellent illustration BTW.


Best ..