Location of choke (electrically)

[EFK]

I picked up an old basket case Selmer T'n'B 50 not too long ago that I've been using as a mod platform to make it more Marshally (they're usually very bright, cleaner amps - think more Hiwatt than Marshall). All of these Selmers - and I have a few of them - use a pretty unique (for a guitar amp) choke arrangement: power from rectifier hits the first filter cap, then the choke, THEN the OT and onwards. So, the plate supply is being run through the choke as well, obviously different from a Marshall where the plates are fed before the choke. All of these chokes (at least all of mine) measure about 2.5 to 3H and around 55-65 ohms resistance, and apparently are designed to handle the current draw because I have not come across one that is burned out yet. What I'm wondering is, *theoretically*, what effect is this choke location having on the characteristics of the amp when it is completely cranked to 10? I know someone is going to tell me to just try it and find out, and I likely will, but that portion of the wiring is still quite neat and tidy and I don;t want to go digging into it if the difference is negligible. the circuit is quite different from a Marshall, actually more Fender than anything, so I can;t really make a direct comparison. My suspicions are that the choke location is tending to try to stabilize the voltages under transient loads (like if you really whack a chord hard) so the audible effect *might* be a stiffer and "harder" sounding amp, all other things being equal. IF this is the case, I'm torn, because I like the really tight bass but would also like some sag on the higher frequencies as when soloing etc. Just curious to hear some thoughts and opinions on this. Anyone?

[EFK]

I should also add, that while one might think that because the choke has full plate current going through it, it should already be saggy under full load, in these amps it is not. At least not that I've measured. I measure less plate voltage sag in these particular Selmer amps than my metal panel 50W Marshalls w/ the 1202 164 PT (which are lower voltage anyway, but do get quite saggy). Apparently the Selmer chokes are low enough resistance that they really don't effect that much sag.

[Chuck H]

60 ohms of resistance can create some sag, but... Not so much as the choke is stabilizing voltage. Those amps are typically biased hot IIRC. So the current draw at idle may actually be similar to what it is under load. In this case the resistance means little. In a cooler biased amp that has more dynamic current draw between idle and conducting the resistance of the choke would be more significant. This is one of those amps that just works as it was designed. Genius engineering or happy accident, it doesn't matter. I wouldn't fuss too much about it. Cooling the bias and moving the choke between the plate and screen may be something to try though if your after a different "feel".

[EFK]

Thanks for your thoughts Chuck. I guess about 60 or so ohms is usually what people recommend for a sag resistor to mimic the 5AR4 feel, isn't it? Although frankly I don;t find those rectifiers to be saggy other than the initial voltage drop. Now that I think about, much of the 'hard' feel might be the fact that - aside from the choke location itself - the TnB50 Mkiii uses a dual 50 uf can as the first cap, so the plates alone are technically seeing 50 uf, an inductor, and then another 50 uf. That's a lot of capacitance, at least in my book.

[loudthud]

Much of the sag you see in a Marshall or Fender amp is due to the increased ripple on the reservoir cap when high current is demanded by the amp. Your meter just reads the average voltage which is midway between the peaks and valleys of the ripple. When high current is demanded the peaks of the ripple will drop some depending on how stiff the transformer is and the type of rectifier. You can easily observe this with a scope or in a simulator like PSUD-II.

When LC filtering is used for the screen node, the LC filter does a good job of taking the average voltage of the ripple and filtering out the 2x line frequency sawtooth wave.

A CLC filter turns the 10 or 20 volts peak to peak ripple into a distorted looking sine wave of 1 or 2 V pk-pk. Much of the sag is gone, the bass is firmer and cleaner. I like the change in a bass amp but a guitar amp seems to lose something.

[EFK]

"I like the change in a bass amp but a guitar amp seems to lose something."

Well, these were marketed as a one size fits all bass and guitar amp, after all! I agree with you. It seems no matter what changes I make to the circuit, the overall impression is one of a very hard, very glassy amp with a super-tight bass.

[stevehoover]

choking the plates

Hi guys,

One thing I like about a CLC before the plates in my experimental builds, they seem a lot quieter as far as hum, when I set them up that way.
I've always tried to A / B between the "right" way and CLC before the plates and can't detect much of a sonic difference except less hum.
Maybe it's the Sag part of it that that I'm missing....not sure...but they are quieter.

I believe the old Gibson GA77RV used a CLC before the plates as well.

Cheers,

Steve