• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Cathode resistor for EL84 P-P Triode wired

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2015
Paid Member
Hi,

I've searched extensively for a concrete answer here and couldn't find one.
I've built an EL84 P-P in UL (7.6K OPT)
It is cathode biased and has a 270ohm resistor per tube, B+ around 310-315V, giving 35-39mA bias per tube.

For converting this amp to Triode (without replacing the OPT), do I need to increase the cathode bias resistor from 270ohm to 560ohm (per tube)?

A high-level explanation on the reasoning would be helpful too :)

Going off of the attached Mullard EL84 datasheet.

P.S Amp is a Tubelab SPP but question is generic.
 

Attachments

  • EL84 Mullard.PNG
    EL84 Mullard.PNG
    61 KB · Views: 389
Last edited:
A high-level explanation on the reasoning would be helpful too :)
The datasheet gives you Rk values for ultralinear ("distributed load") mode in class A and for triode mode in class AB. Hence the difference in Rk (and in quiescent current) for the same plate and second grid voltages.
If you're OK with like 3-4W output in the class A triode mode, you can keep Rk intact.
If you want more watts, you'll have to bias the amp for class AB, increasing the Rk and decreasing the quiescent current.
 
Itsik,
Be careful running upwards of 35mA per EL84. You are getting close to their dissipation limit. Not so bad with the Russian tubes, generally speaking, but the JJ EL84 has a bit lower limit. If you are thinking of running triode it might be safe to increase Rk a bit even if you don't go as far as 560 ohm. Morgan Jones ran 560 ohms on his EL84 in triode for about 5 watts output.

S.
 
DIY TUBE BIAS SCHEME

The DIY Tube Dynaclone (no longer available) used an adjustable cathode bias scheme to optimally balance the current flow through the output tube pairs. I've used this on a few EL84 amps and it could be helpful for cranking back current flow in triode mode too.

S.
 

Attachments

  • Bias.png
    Bias.png
    40.1 KB · Views: 336
That's correct, 10.5/270= 0.039A
To calculate power dissipated by the tube subtract Vk (10.5 volts) from your B+ (310 to 315) times 0.039A= 12 Watts.

It would be more accurate to measure the voltage at the anode of the tube (it will only be a couple of volts lower than B+) then subtract Vk. But B+ will be close enough.

Most Russian-made EL84s will dissipate approx. 16 watts (14 anode + 2 G2). So if you are running your EL84s at about 12 watts you are OK but personally I wouldn't go much higher.

JJ EL84s are rated at about 14 watts (12 + 2). These would be run a little too hard in your rig if they draw more 35mA. It's not serious but you should keep an eye on it.

Cheers, S.
 
For converting this amp to Triode (without replacing the OPT)
The required Z pri for triode connected valves are usually higher, see attached. For your B+ etc a Z pri of 10k is recommended. Using a less than optimum Zpri will alter power out,THD etc.


Re choosing the right Rk, you could just temporarily wire in a 500r pot and tweak it whilst monitoring Ik, power out, THD etc. If you wire the pot in series with a 100r fixed resistor you can use the 100r as a "sense" resistor to measure Ik. Make sure the pot is capable of handling the power and wire it as a rheostat, EG wper connected to one end tag.


Andy.
 

Attachments

  • EL84 data.png
    EL84 data.png
    188.4 KB · Views: 268
Last edited:
Not just any 500R pot... a 5W WW rheostat would be needed if you don't want the pot to smoke.

026TB32R501B1A1 CTS Electronic Components | Mouser Canada

They are so much cheaper than I expected, why not just use one per tube :D
If you want to save a bit more, Digi-Key has a free shipping option for the US and Canada with no minimum purchase. Shipping can really drive up the total cost, especially on small parts orders.

Free Shipping On Parts Orders | Audiokarma Home Audio Stereo Discussion Forums
 
Member
Joined 2015
Paid Member
Thanks all for the additional information.
To provide some context, if I were keeping the amp I would leave it as-is in UL since I think it sounds really really good.
That said, I put it up for sale and got a serious offer from an individual that wants it triode wired.
They are using horn loaded Lowther's, and mentioned that even 1V rms is enough for their preferred listening level.

BTW-- this is my favorite amp that I've built to date. The only reason for sale is because I am also working on a Baby Huey EL84 with custom OPT's from Sowter, so once that one is done I don't know if this EL84 will get much use.

Given that, I want to avoid putting any adjustment pots for ease of use for the non-DIYer.

I have Jan's Autoranger along with ARTA and will measure the output power, THD and frequency response once wired as Triode, with the existing ~8K OPT's and 270R resistors.

Changing the OPT's is not an option for the sale, too much loss there.
So the only option to mitigate some of the issues with the 8K OPT would be to increase Rk, or reduce B+ a bit?
We'll see based on the measurements.


Link to the completed build: https://www.diyaudio.com/forums/tubelab/157491-pictures-tubelab-amp-56.html#post6436039
 
Last edited:
Member
Joined 2015
Paid Member
Custom, from Landfall Systems -- Landfall Systems
Not as cheap as stock item, but cheaper than every other option I've looked at that included CNC work.
IIRC, this case was about $150 total, which included the chassis (3mm alum), brushed finish, CNC work on top/rear panels and shipping.

I've used them for the SPP, 2x TSE's, a Pass Pearl and others.
Really great product and service overall, highly recommended.
 
I operate my Tubelab SPP in triode with no feedback. Output transformers are 6600R Hammonds. By my calculation it operates in class A; about 3.5W to clipping on an oscilloscope. Sounds very good.

I changed nothing on the amp to do this (i.e. I kept the 270R cathode resistors).

I did need to lower the B+ voltage to 300V (to keep the EL84's in spec) which I did by putting a resistor on each leg of the power transformer secondary's before the PCB terminals.
 
Member
Joined 2015
Paid Member
Thanks for the info cnau. Which value resistors did you add to the secondary?
Since rectification is done with SS diodes, the B+ was fairly high to begin with.
To reduce it, I added another RC filter with a chassis mount resistor.

I've rewired from UL to triode (no further changes) and the tubes look to be running a little hot. I measured about 11.6V drop across the cathode resistor on the 6P14P-EV tubes.
With B+ at 318V, this is 43mA bias and 13W plate dissipation.

I was thinking of increasing the 270R to 330R to reduce the stress.
 
It's within the spec since it's running as triode and you can add plate and screen dissipation together but by all means try it and see if you like it.

FWIW: I used to run 6P1P at 18W and they last for about a year 24/7. I dropped it from 56mA to 40mA so they last longer.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.