how close can a choke be to the OT?

[elipsey]

I'm building a princton reverb with deluxe reverb transformers. I'd like to mount a choke but it's a little tight. Here's a picture:


Is this OK?

[Tom Phillips]

I think it will be OK.
Tom

[tubeswell]

It can be right next door, and if the EMF flux axes of the two cores are perpendicular to each other, that is probably even better.

[raiken]

On a related note, beware that the choke wires can radiate 120Hz hum into input preamp circuitry if the choke is located over the top of the preamp section of the amp and the wires enter the chassis and run underneath the preamp board components.

Best to keep the choke wires away from the low-level preamp sections, they carry bursts of high current noise as the input AC signal "tops off" the charge on the filter capacitors twice on every AC cycle, and these high-current bursts create electromagnetic fields that can radiate quite a distance. I had to fight this problem on one of my own designs.

[R.G.]

+1 It's widely UNDER appreciated what rectifiers and filters do, and what high pulse currents and discontinuous voltages flow in there.

Most of the high pulse currents are in the wires from transformer through rectifiers to the first filter cap, then out of the first filter cap back to the CT. The choke *can't* change current fast, that's what it's there for. But ripple voltages can get steeper with a choke load driving them if the choke and cap are not operating at resonance. And the stock filter cap -> choke can't operate at resonance and do a good job of filtering. So you can get fast transients at a 120Hz rate.

Good, careful design of a power supply pays close attention to currents and wire routing for everything up to the second filter cap - and beyond if you're really trying to make a quiet amp.

Did you try shielding the choke leads? That had to be voltage noise, cause chokes don't do fast current pulses.

[raiken]

The choke *can't* change current fast, that's what it's there for. But ripple voltages can get sr with a choke load driving them if the choke and cap are not operating at resonance. And the stock filter cap -> choke can't operate at resonance and do a good job of filtering. So you can get fast transients at a 120Hz rate.

Did you try shielding the choke leads? That had to be voltage noise, cause chokes don't do fast current pulses.

Yes, of course, you're absolutely correct, current can't instantaneously change through an inductor. However, as you mentioned, the charging current pulses do create fast transients at the 120Hz rate. I may be wrong, but I assumed the choke wires were radiating these high current noise bursts, rather than just the few volts of 120Hz sawtooth ripple voltage on the wires, although, if I understand you correctly, are you saying it could be the faster risetime of the edges? I was able to fix the problem by rerouting the choke wires out and away from under the board that had the preamp components and wires.

[R.G.]

I believe it was the voltage change. It would be interesting to try moving the wires back so they interfered, then subbing in shielded ones.

I normally go for twisted pairs for reducing interference, but in this case, I don't think twisting the choke leads would help.

Edit.

You know, the choke leads *would* be a good place for launching bursts of RFI from rectifier slam-off. Was this a tube rectifier or solid state? And if solid state, un-snubbed rectifiers or snubbed, or FREDs?

[raiken]

The choke wires were tightly twisted. Rectifiers were garden-variety 1N4007, unsnubbed. I've since switched exclusively to fast-recovery diodes in all my designs.

[daz]

I don't know if this is good practice, but it works for me. Maybe you guys can tell me if this is not good for some reason. But i run all my choke and OT wires outside the chassis so theres a grounded shield (the chassis itself) between the wires and the other circuitry. The wires then enter the chassis at their destination point via small hole with a rubber grommet and the wires are bunched up inside hear shrink till they enter the chassis. As for the placement of the choke, ot and pt, i used a marshall as a model since it's a known good design.

[g1]

Maybe you guys can tell me if this is not good for some reason. But i run all my choke and OT wires outside the chassis so theres a grounded shield (the chassis itself) between the wires and the other circuitry. The wires then enter the chassis at their destination point via small hole with a rubber grommet and the wires are bunched up inside hear shrink till they enter the chassis.

Not sure about the deterioration of heatshrink. It would mainly be a safety issue. For commercial applications I doubt you would get away with heatshrink around high voltage wires as the user can make contact with them without chassis removal. They would probably require something like the Fender "dog-house" they use over the filter caps.

[big_teee]

I've not had issue with the OT, and PT leads.
All of my past hum and noise issues, have either been Ground or Heater wiring issues.
Maybe I got lucky.
On my last build I did, PT--Choke--OT.
Reasonable separation between each.
The OT wire routing worked out better that way.
It turned out to be the quietest Cascade Preamp Amplifier, I've heard to date.
T

[Joe L]

I fought 120 Hz pulse noise on an amp I shipped about 3 weeks ago. The choke was inside the chassis ala SLO style and the wires were short and never came close to the preamp board. I tried FREDs, put the filaments on battery along with the switching circuitry, checked out different grounding schemes and nothing phased it. I finally ran out of time and had to ship it.

It wasn't very loud and the customer laughed at me when he got it, saying it was the quietest amp he ever owned, including his dad's SLO. But you know how it is, they don't have to make noise if they are designed right.