Bias voltage for EL84?

[titser_marco]

Hi guys! I got an old Crate Vintage Club 30 and after doing inspection, only the chassis and transformer would be useful. Figured I'd make a new amp altogether after I junked the PCB.

Been searching for schematics and for the most part, all the projects that I came across are cathode biased designs. While I'm OK to go down this route since the VC30 is also a cathode biased design, I'm curious if I can make the new amp a fixed bias type since the transformer had a secondary that used to power the transistor driven fx loop and reverb on the VC 30. That has about 10vAC so I can wire up a bias circuit feeding about 14vDC tops).

BeforeI proceed I was wondering if you had an idea what -vDC should be going into the grid of the EL84s for it to work without redplating etc. Based on the old VC30 schematic I can expect a B+ of about 371vDC so I thought I'd also mention that to give a bit of context as regards my question on bias voltage. Thanks!

[pompeiisneaks]

Looking at this datasheet, at 300VDC you'd want just under -15V

https://frank.pocnet.net/sheets/084/e/EL84.pdf

if you go higher, you'd need even more negative voltage, the sheet shows lower voltages and they go down to lower voltages, also this data is for a specific manufacturer, but most stick to the same general ball park.

The JJ datasheet shows only at 250VDC and says -7.3V https://drtube.com/datasheets/el84-jj2003.pdf

Most of the el84 amps I've seen have been cathode biased. Not specifically sure why you want to do the grid bias, but it surely can be done. You'll need to get higher voltage than the max by a decent amount and then get it lowered with resistors etc. Looking at the numbers I'd say shoot for -25V at the output of the rectification and then setup a dropper resistor into a pot so you can dial in the exact voltage to get you the exact plate current you want.

~Phil

[martin manning]

You could use a voltage doubler supply from the 10VAC tap and have all the negative voltage you need.

[Stevem]

The VC30 schematic specs out 317 volts for plate voltage!
With that much plate voltage on hand I would cathode bias it like a AC30.

[sluckey]

Here's my favorite fixed bias EL84 amp...

http://sluckeyamps.com/november/november.pdf

[joeboo88]

Here's my favorite fixed bias EL84 amp...

http://sluckeyamps.com/november/november.pdf

Hi Steve: on your schematic you have the last e-cap as 16uf, on the layout you have 47uf, which one is correct?
unless i'm reading wrong... I ask because I like using your schematics to build stuff, also what does this November sound like - an el84 JCM? or Plexi type?

[sluckey]

47µF is what I used. I have no idea where the 16 on my schematic came from. Sorry 'bout that. I may just correct that error. FWIW, the original November Rev. 11 used all 32µF caps.

The amp is not high gain. It's meant to mimic a Plexi JMP 1987 amp. It does a pretty good job with the Plexi sound, maybe not as strong on the bottom end.

[joeboo88]

Thanks for the info Steve.... hope I didn’t derail too much

[dbeasley]

You can also derive a bias supply from tapping off one of the HV secondaries and rectifying like Marshall does. Depending on if you're going Full wave bridge you might need a few extra parts (.047uf class X cap and 56K resistor) I believe the Marshall artist series uses this supply


Sent from my Moto G (4) using Tapatalk

[tubeswell]

Somewhere in the -9V to -14V range would be about right.

A Rule of Thumb guide for most types of output tubes is to aim to allow for 10% of the plate voltage - but 30-40V is much more than you need for EL84s.

A more technical approach (from Merlin Blencowe's website) is: "take the screen-to-cathode voltage and divide this by the triode mu of the valve, i.e. the mu quoted on the datasheet when operating in triode mode."

EL84 mu = 20.

VC30 Schematic shows a B+ of 317 (not 371), and a g2 voltage of 307. So the screen-to-cathode voltage is somewhere around 300ish. 300/20 = 15. (So -15V tops)

[martin manning]

Looking at the VC-3112 schematic,
- It's 317 volts on the plates (as Stevem and Pete pointed out)
- The power supply for the solid state stuff is bipolar, with +/-15V raw DC. Using that winding as a negative bias supply you will have more than enough negative voltage, about -30VDC.

[titser_marco]

Totally misread that, yes it is 317vDC. Thanks for all your inputs!

Sent from my SM-G610Y using Tapatalk

[romberg]

Merlin has a nice page which describes how one can calculate the required bias voltage from the screen to plate voltage and the mu of the tube in question:

http://www.valvewizard.co.uk/bias.html

For back of the napkin calculations this is mostly 10% or so of the screen voltage (with the exception of el84s). It explains why many el34 amps often melt down a set of kt66s because kt66s need more negative bias voltage than the el34 based amp was setup for .

Mike