6s4a Clementine.....Sick Thought

[EWBrown]

I've built a PSE 6S4A SET amp, with each channel using using two paralleled 6S4As (each with its own 390 ohm cathode resistor and bypass cap), driven with an SRPP driver stage, I used 5963s, the 12AT7s I started out with, had way too much gain. I used RK and RA of 1K ohm,with the RK bypassed with 100 uF / 16V cap.THis runs theSRPP stage around 4.8 mA,the 12AT7 with the same values, ran about 2.1 mA.

I used some cheap 6.6K PP OPTs as SE OPTs, they are not airgapped, but they work and sound pretty good.

If you go with 3 64As pr channel, a 4 to 5K SE OPT should work well, the three paralleled tubes will consume a totalof approx 75 mA plate current, so make sure that the OPT primary can handle the current, look for at least 90 mA primary current capability, the GXSE15-8-5K should work well in this application.

/edB

[zippy1540]

I'm glad you mentioned current capacity on the output transformers, i'd forgotten about that part. i have two 3.2k output transformers which is why i was thinking of the PSE setup but i'm betting these things will only take about 25ma. no good for el-34's either. i may just bite the bullet and build up the full 6l6gc clementine if i can sell one of the amps i have or trade for parts.

[zippy1540]

Update on my little project, I drilled in two extra 6S4A tubes for a PSE setup and changed out the power transformer. Also went a different route with the cathode bypass setup. i used an lm317 for each tube with a 47ohm resistor to give me 26.5ma constant current and bypassed the whole thing with 100uf 35V capacitors. My B+ right now is 366VDC. Other than some 120hz hummmmmm it sounds great. the hum almost goes away when i short the inputs. i think the volume pot is part of the hum problem since it varies with the volume setting. It's a cheap Rat-Shack 100k unit. My source right now is a little portable cd player unit. The rest of the circuit is based on the clementine front end. the output transformers are still the cheap 40-1 scrap box organ transformers giving about 12k with 8ohm output. Anyone see any issues i could improve?

[zippy1540]

Humm fixed, Grounding issue. now it is dead silent. time to hook it to the main system and see what it can do. i know the outputs are going to limit the freq range but i should be able to tell if it's worth buying some edcors for it.

[EWBrown]

With three 6S4As in parallel, the B+ current consumption per channel would be about 75-80 mA (25 to 27 mA per 6S4A) , use a 15 W, 5K OPT, like the GXSE15--5K.
B+ 325VDC, and use a 390-400 ohm, 2W resistor, bypassed with 100-330 uF / 25V (or greater) cap. If the 6S4As are fairly closely matched, then a shared 130 ohm,
5 or 10W resistor would work, but with the separate RK's the tube-matching isn't quite so critical.

I'd use n SRPP driverstage, perhaps 6CG7 / 6FQ7 or 5963 with RK=RA=1K ohm, this would run around 4-5 mA at 300VDC B+, use a 150-200K grrid resistors for the 6S4As and individual1K grid stoppers for each one, Coupling cap should be 0.22 to 0.33 uF / 400V or greater.

I built up a PSE 6S4A with 2X 6S4A and 6CG7 SRPP per channel, and some cheapo 6.6K 10W OPTs. It sounds pretty decent, probably 2,5 to 3 WPC.

In some past posting, someone had mentioned that they would have liked to see the EL84, but built without the G2 and G3 grids, the 6S4A comes pretty close to that
"fantasy tube", just a lower PD rating of 8.5Watts. However, it has beenmy observation that these vertical deflection tubes, when used in a Class A audiopower pushed qabout 50-60% greater than that, the 6AV5GA with its published 11 W PD, can easily handle 18W and even 20 Watts. The 6V6GT in deflection mode has a plate dissipation of 10 watts, in Class A power amp operation, 14W, plus 2W for the G2. And, I've pushed them well beyond that, in a Fender "Tweed" clone, without blowing them up.

I'd wager that the 6S4As could be pushed up to 12W PD (V P-K X IP) without blowing it up 10W PD should be "a walk in the park" for these tubes.

/ed B

[zippy1540]

Ed, What do you think of using the LM317 set up with a 47ohm resistor in constant current mode for the cathode all bypassed with a 100uf 35v cap? Am i asking for any issues doing it this way? Current is set for 26.5ma with 360V on the plates. I'm still learning by trying things out and seeing if i let the smoke out, so far no smoke and it sounds incredible. No red plating and with inputs shorted a silent background. The only shortcoming right now is the slightly limited freq response because of the cheap output transformers. I may go for some 10W Edcors 10k:8ohm.

[EWBrown]

The LM317 , 47 ohms & 100 uF 26.5 mA constant current cathode sink should work very nicely with the 6S4As, I've experimented with a similar approach,
with good results, this is best applied to Class A operation, (as with all SE amps). This will effectively "match up" random 6S4As.
The LM317's maximum input voltage is well above that encountered with 6S4As, which is typically about 10-12 VDC maximum.

In PP designs, the LM317s are OK with Class A operation only, then they start to "go bad" with Class AB1 operation, and then it will introduce distortion.

I haven't yet ventured beyond 325V B+ on the 6S4As, I'd suspect that they would operate well up to about 400V, before serious issues started to erupt
Factoring in the combined voltage drops through the OPT primary, and across the LM317, that would work out to about 9 to 9.5 W maximum plate dissipation,
which should be perfectly fine with these tubes, I'd not exceed 12W PD, except in a "brute force survival" experiment...

/ed B

[zippy1540]

Another Shot

100_2603.JPG (585.73 KiB) Viewed 3836 times

Glamor Shots

100_2602.JPG (343.42 KiB) Viewed 3836 times

Under the hood shots to follow when i get home from work
Paul

[zippy1540]

WOW Those came out too big for comfort!

[EWBrown]

I have found that the maximum ratings on deflection amplifier tubes are extremely conservative (as this kind of operation is very hard work on the tubes).

For instance, a 6V6 in audio use has a plate dissipation rating of 14 Watts, but in vertical deflection mode, it is downrated to 10 Watts.

I've taken numerous 6AV5GAs with a published PD rating of 11 Watts, and have run them very successfully at 18 to 20 Watts PD.

The 6S4As have a published max PD of 8.5 Watts, though I'm sure that they could be safely pushed to 12 Watts PD, in Class A audio service, without going into meltdown / red plate mode.

Another "torture test" for me to try sometime, I have a quantity of used / pulled 6S4As which can be used as the "sacrificial lambs".
No sense killing off perfectly good NOS tubes for this...

/ed B