5E3 for 6L6 or 6V6 Filter Choke/resistor sw mod

[tubeswell]

Hi there fellow 5E3 junkies.

Modding my 5E3 clone for switchable CLC-CRC pii filter.

Assuming a current of 5mA per screen;

1) do you think that 270R 5W will be enough of a resistor for getting the screen voltage under the plate voltage when the choke is engaged? Or should I go with 470R 10W (which would be clunkier)?

2) Or would a better way to do it be with separate 5W 470R screen grid resistors and a lesser value (say 2k7) Pii filter resistor switchable with the choke?

3) Or should I just go with a simple choke/resistor sw in the Pii filter and forget about the screen voltage?

Please see schematic attached (refers to Qn 1).

[txstrat]

In my last thread I've learned from Bruce, that the difference between plates and screens shouldn't be seen as dramatic.
I would start with the easiest mod and check the voltages. If they're ok - there you go.
If not you could go on with the next mod and so on until you got what you are looking for.
You might wanna check the thread I was talking about:
http://music-electronics-forum.com/s...ad.php?t=10636

[tubeswell]

Thanks for reminding me about that thread txstrat. After re-reading it, I decided in the interim I would try it with a choke-only, and I put a 500R 10W in series with the screen supply. This brought the Plate up to 370 and the screens were 367. But it still sounds great. I also re-did a bit of the grounding (moved the screen node filter cap ground from the pre-amp ground to the output stage grounds - and shifted a few wires here and there). There is almost zero hum now and it sounds absolutely fab with my 6G15 clone on the front end.

I might add a switchable resistor back into the Pii filter later. For now I am over-the-moon happy with it. 8-)

[txstrat]

Are you planning to post some sounds?
I'm very curious.

[tubeswell]

Yeah I'll do some recordings tomorrow

[tubeswell]

TW 5E3 clone with and without reverb unit

This track is the 5E3 with the CLC filter running jj6V6s and a Soldano (Emi OEM) speaker, and has the 6G15 reverb unit lightly on the backing track and the impro' track is dry. ('xcuse the disorganised impro)

SM57 through my desk into the mac line-in (with good ole' garageband) me twidding on my strat

[txstrat]

Sounds great. Hats off.
Does it sound as great when cranked? When does the crunch start?

[tubeswell]

Re-done voltages

I tried a few more tweeks today

with a 1k resistor at the screen - not much difference on plate-to-screen voltage, although the plate voltage dropped to 365 (screens at 361) the cathode (255R) was at 24V (and the tubes were idling at 15.95W each!). I hadn't measured the cathode voltage the other day when I took the voltages, so I had had no idea how the tubes were really doing and I thought I better do a few more tweeks.

A single 5W 4k7 screen resistor brought the screens down to 19V below the plates, (but the tubes were still idling at 15.3W each!)

So I zenered down the B+ a bit with a 15V and a 12V (5W) zener in series and the plates were 353, screens 334 and cathode 22V (13.6W each plate - after subtracting 2mA screen current on each tube - these are JJ 6V6s). So I'll settle with that for now.

Incidentally, I forgot to plug one of the 6V6s back in when I test drove it the first time and thought "Hmmm I'm getting more distortion than I was expecting" (starting at 1 or 2) and played it dimed for 5 or 10 mins without realising (until I saw the other tube lying on the bench) - ah well - no apparent harm done - plugged the escapee back in and it works fine :-0 !

[tubeswell]

A further wee followup

Today I had some spare time so I tested a GZ34 and some JJ6L6GCs in the amp. I had just one 15V 5W zener dragging down the HT winding to about 310VAC-0-310VAC (simply because I was lazy and left if there so it'd keep the voltages lower when using 6V6s - but it could easily be made switchable to enable the B+ to be boosted a bit. Anyway the amp sounds shit hot with either 6V6s or 6L6GCs. I added a switchable choke-resistor filter last week, and that adds even more variety.

The voltages with the GZ34 and 6L6CGs running into an 8K load resistance are: ('twould no doubt be a bit higher with the load resistance at 4k, but I didn't bother measuring that)

PT Spec
325-0-325VAC 180mA primary (with a 15V reverse biased zener making ~310VAC?)
6.3V 4A heater CT
5V 3A recto

B+ = 412VDC

With CLC filter (150R 4H 50mA choke and 20uF caps)
Plate = 404VDC
Screen = 379VDC
Cathode (255R 10W) = 30VDC
~22W per tube

When the Choke Filter is switched in there is a 4k7 5W screen grid supply (measuring 4640R) which drops 25VDC, which divided by 4640 = .00538A = 5.4mA in total for both screens, x 374VDC screen voltage at that point = 2.02W dissipation on that resistor


With CRC filter (4k7 5W resistor and 20uF caps)
Plate = 408VDC
Screen = 364
Cathode (255R 10W) = 29VDC
21.55W per tube

That's about it FWIW

[Five_E35]

Is there much difference in sound between choke vs. no-choke? My limited understanding of the whole screens lower than the plates issue is the plate dives way under the screens when the amp is actually doing something anyway. In these cathode biased Class A's(or Hot-biased Class AB's) the power supply is pretty much pre-sagged and the choke's benefit as a better regulator wouldn't seem to be as great as in a fixed bias AB1 type amp??

In any case, I got similar numbers running mine with a GZ34 and 5881's. Kind of interesting dissipating twice as much power as the thing produces(at 8 ohms) After reading something on another forum, I tried something you might find interesting - run it into a pair of 10's wired for 4 ohms using the normal 8 ohm output with a GZ34 & 5881's. I tried this using the diagonal pair of P10R's in my Bassman. If it's a normal 5E3 OT, it will max out around 25 watts, but the sound almost has me wanting to get a baffle made for a pair of 10's.

[tubeswell]

The choke is a bit less-squishy/more-defined than the resistor, but there's not much in it. There doesn't seem to be any difference in noise floor because there's no hum either way.

I would've run an 8R speaker into the 16R tap (to get 4k) if I had a powerful enough 8R lying around at the moment. The OT is quite big and is probably rated for at least 30W, and running the 6L6GCs at 8K for a wee while didn't seem to bother anything. The 6L6s made the tone a lot more dynamic (probably because the OT was working harder).