el84 Bias 9V at grid enough?

[Guitarist]

All there is available upstream is 12Volts for all four tubes. I bypassed a series resistor and raised the bias a couple volts but 9 volts seems dangerously low for fixed bias (at grid pins); no?

The Peacey C30 has 14 Volts upstream and this Duncan has 12V—but the grid blockers are bigger here at 100K vs 47K in Peavey.

Thanks guys.

N.B. Good contact cleaner went a long way cleaning up the row of neglected pots in this amp. It seemed DOA prior.

[Chuck H]

Hmmm.?. 550V on the plates of EL84 tubes. Could that be a typo???

What do you get with your DMM?

Chuck

[Robert M. Martinelli]

All there is available upstream is 12Volts for all four tubes. I bypassed a series resistor and raised the bias a couple volts but 9 volts seems dangerously low for fixed bias (at grid pins); no?

The Peacey C30 has 14 Volts upstream and this Duncan has 12V—but the grid blockers are bigger here at 100K vs 47K in Peavey.

Thanks guys.

The grid blockers have nothing to do with the biasing function, the ( control ) grids don't draw any current, so the bias voltage stays the same, and -9V would be too "positive" even at 320 VDC, let alone at these plate voltages (550VDC !! I hope, together with Chuck, that's a typo... ).

According to my old ( 1967 ) Philips tube data book, 550 VDC are the maximum plate voltage ( design center values ) for EL84s, but, even if willing to operate at these extreme voltages ( which I dislike ) in order to gain some headroom, to stay within the tube's max plate dissipation ( 12W ) the biasing current should be set very cold, at some 18 mAmps....I don't think -12V is enough negative a voltage to achieve this goal.

At 550 VDC ( +B ) and -12V ( bias ) this amp would look like a "perfect tube torturer/terminator", taking this title away from the former champion, the good ol' AC30, biased at 125%, operating in AB1-O.M.G.!* class ( thanks Steve ).

Cheers

Bob

* Oh, My God!

Edit: I took another look to the schematic, and the plate voltage indicated in the output area seem definitely wrong, the +B voltages out of the PS are 340 VDC, 307 VDC and 275 VDC....that seems far more reasonable...

[Guitarist]

Well, those high voltages are correct. I've a strange meter here but it reads this (unless my understanding of it is wrong).
565 V on pin 7
290 V on pin 9

I instantly blew up a new EH and another phillips. I have a new quad of JJs here that have these matching numbers on the boxes:
20.6
4.91

All I need is the bias volts/parts values as I don't really fully understand the theory involved, sorry. Is -20 Volts a good guess?

I know you're busy, but if I decided to go cathode biased would one BIG 35Ω resistor work for a common R.? A two valve Vox has a 50Ω.

OR would an easy fix be to add this resistor to help/augment the -9V fixed bias to work better? The current limit is 12Volts so... This seems to be a "wide-lens" solution, yes?

[Robert M. Martinelli]

Well, those high voltages are correct. I've a strange meter here but it reads this (unless my understanding of it is wrong).
565 V on pin 7
290 V on pin 9

I instantly blew up a new EH and another phillips. I have a new quad of JJs here that have these matching numbers on the boxes:
20.6
4.91

All I need is the bias volts/parts values as I don't really fully understand the theory involved, sorry. Is -20 Volts a good guess?

I know you're busy, but if I decided to go cathode biased would one BIG 35Ω resistor work for a common R.? A two valve Vox has a 50Ω.

OR would an easy fix be to add this resistor to help/augment the -9V fixed bias to work better? The current limit is 12Volts so... This seems to be a "wide-lens" solution, yes?

Wow!!!
So the term "tube terminator" is well deserved....

1 - Should you decide to "cathode bias" the thing, you' d need to raise the cathode from GND about 14 VDC. Biasing the tubes at about 80% ( some 18 mAmps @(545-14)=531 VDC ) and allowing for some 5-6 mAmps of screen current, each tube would need a ( 14VDC / 0.023 mAmps ) = 608 Ohms cathode resistor, with a dissipated power of ( 14VDC*0,023 mAmps ) = 0,322 W. This means you'll need a 152 Ohms ( 2W is enough, 5 W is safe, 7W is safer ) cathode resistor for all the four tubes IMHO. The closest standard value is 150 Ohms.

2 - 50 Ohms is the cathode resistor for a FOUR ( not two ) EL84 Vox ( AC30 ). Interestingly, this makes for a VERY hot bias, which gave AC30s the ( well deserved ) reputation to be pretty rough on output tubes.

3 - Be aware that, even if I believe the above calculations ( point 1 ) of mine to be correct, I don't think running modern EL84s at these voltages is a good practice.

To play it safe, I'd definitely lower that 545 VDC +B to more "human" levels and redo some math, rather than trying to keep the tubes alive but always teetering on the brink of an abyss....modern tubes are not as rugged as the ones produced in the old times, so chances are you'll find it's very difficult to get modern tubes capable of withstanding these voltages.

JM2CW

Hope this helps

Bob

[tedmich]

+1 on the lower B+; I prefer ~400VDC, and another "sweet spot" is at ~320-340VDC.

Maybe try the simple and neat CCS used on the "Oddwatt":

a pot, resistor and LM317 regulator

DIY ECC802S SRPP / EL84 (6BQ5) Push-Pull Tube Amplifier

[Guitarist]

That's a ton. That does help. What value of resistor should I add prior to the B+/center tap of the OT? 47K 10 Watt? Perhaps 350V would be a good goal.

I'd rather just stay fixed bias if possible. Is the JJ spec sheet is suggesting 300V max "limiting value" or not? More likely 450 I'd guess—but

'also found this from a prior thread:

"We want to get the idle dissipation down to about 8-9 watts. At 545 volts on the plate, that means about 15ma per tube. You have four tubes, so about 60ma through that 10 ohm cathode resistor. That means about .6 volts or 600mv.

Try that. Monitor the voltage across that cathode resistor and set the bias for .6 volts there." source, Enzo

[Guitarist]

I talked to a tech here in Vancouver—he knew this design well. And suggested adding a transformer to lower the AC, which I'd rather not do. We talked about adding a small resistor, to the OT center of around 600Ω 5 Watt but he said it will likely sound like crap. Perhaps not that bad, I say! One would have to add another filter cap here too. This make that section a high impedance power supply but not too high, I reckon.

But looking at the schematic, why not move the lower diode in series with the transformer line and connect it to the send-point of the top diode; conventional? Also disconnect it from the junction of bleeder resistors R59 and R58. Lose the voltage doubler.

thanks.

[tedmich]

yeah lose the voltage doubler, there is also the zener trick to drop (typically ~50V) off the B+, which shouldn't gum up the sound like big resistors


(or build up Gary Pimm's Swenson regulator (parts $15 at Mouser) and dial in the perfect voltage; it supposedly works best when it drops >50VDC)

[Guitarist]

That to Valve Wizards new book and great site info, can I insert a bit of cathode bias then? I just gotta get those tube biased safe, foremost.

"Furthermore, there is no reason why we cannot use a little fixed bias and cathode bias simultaneously to achieve the desired mix of compression and 'bark'." Source, The Valve Wizard -Push-Pull

So an idea has formed—because the fixed bias is maxed at 10 or so Volts can I insert a 30 to 75Ω resistor on the cathodes to compound/round out the biasing a notch? This might be amazing tone-wise and be the perfect solution.


"If you've got a 100 Ohm cathode resistor and 31mA of bias current, it will shift your bias voltage 3.1V positive compared to a grounded cathode at the same current." BackwardsBob from http://music-electronics-forum.com/t7531/

A relevant schematic is a Krager designed Sundown Rebel with combo-biasing.

re-560V B+ Having done further reading, though upper end extreme, the plate voltages can work. The tubes may only last a six months but the tone may be worth it. I heard Sovteks are really tough but this quad is JJs.

thanks