• 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.

EL84 BIAS - how much ??

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello Forum
I have a Beard Integrated tubeamp, using 6 EL84's pr. side.
About 6 months ago, i had the output tubes replaced by new ones, and the guy who did it turned the bias down, I remember.
But, the amp dosent get all that hot anymore - in fact, I can touch the tubes during operation now - witch was a guaranteed burned finger before ;)
I also think that the amp dosent sound as good as before.

I want to have my old, good sounding heater back !

I read 2,35 mA between the metal case, and the bias input on the PCB - right after the trimpotmeter that reads Bias.
How much can I push theese tubes - any suggestions would be most welcome.

Thnx in advance.
Hans Houmøller - Denmark
diyAudio Senior Member
Joined 2002

Hi Buhl,

This one is a rather tricky one.
So far I haven't seen any technician solving this problem.
Whatever anybody did to your amplifier:
As it is now, it is working the tubes pretty close to cut off (I don't have any exact figures here.)
In other words it is most likely to work fully in the low ranges of class AB1.(i.e. more B than A)
As you state,you can hear the difference.
As far as I can tell:
the problem with this particular design is that you need to match the output tubes and ,to boot, you better match them in the amplifier.
My observation is :the technician probably has tried to bias these tubes but was unsuccesfull.
He always saw one tube or the other glowing cherry red.
He therefore resorted to cutting them off nicely which you now hear as clipping,the tubes are getting overloaded by the previous stage.

Other then the above I have no cure really,

Best rgds,;)
Thnx for the reply fdegrove !

As a matter of fact, the technician, a local hififreak, swapped a lot of tubes in and out of the socekts with the old bias setting, before turning the bias down - he ment that the tubes was beeing runned to close to the limits, and would die again quickley.

Despite the high bias setting, the old tubes ran for about 1½ year, and the amp was used when I got it - I can easily live with 1½-2 years of operating time pr. set of tubes.

So, the tubes are matched to sustain higher bias.
If I would try to up the bias a little, how smalle stemps should I proceed in ??

Is measuring mA between bias and gound the right way to measure the bias ??

again, thnx alot !

Hans Houmøller
memory recall !

Hi Again
I just measured the bias on each of the 6 output tupes pr. side - and get 0.128 mA - 0.129 mA on each - I think to recal that the old setting was near 0.18 - could this be correct ?
Would it be to much you think if I turned the bias up again to 0.18mA / tube ??


diyAudio Senior Member
Joined 2002


Despite all the efforts,I don't have a cut and dried answer here.

Admittedly:if you do not mind the expense of changing the tubes every 18 months,the fact remains that is likely to give you trouble in the long run.
If memory serves me correctly the bias is shared between all the tubes and all I've seen so far is it getting into thermal instability.
(runaway bias).
This,to my mind is a design flaw and the only simple cure I've seen so far is exactly what this techician did.
You can try to up the bias a little bit (say 5 mA at the time) but I can already predict trouble.
A design such as this one should come with individual tube bias and there's just no way I can predict it's behaviour.
It is however very likely to drift off into instability.

My solution at the time (although seemingly expensive) was to replace all tubes with E84L/7320's from Siemens (or other tightly matched tubes).
This presented a not unconsiderable bill to the customer and I guess most of the owners sold their amp later on.
Passing the hot potato as they say.

I'm sorry for you, but I don't have absolute answers for you here and short of redesigng the amplifier this is my best shot at it.
Luckily for you the output tubes used aren't frightfully expensive.
I hate to think what to do with 8417's here.... ;)


Thnx Frank
I can se the problem with the shared bias and the "poorly" matched tubes - the 12 is matched roughley out of about 20.
I will try to up the bias to about 0.15 later this evening I thnik - 12 new tubes are only 50 euro, so I think its worth the try.

Thank you SO much for the qualified and fast response !!!

diyAudio Senior Member
Joined 2002


Before you do anything,can you give me say half an hour to look up the data for that tube?
I got the sneaking suspicion either one of us didn't put the comma in the right place.
Excuse pls.
I'll report back unless Bruce wants to do the hard work here?

Anode voltage

Hi Bruce and Frank
I only know that is is supposed to an 'real' class A amp, delivering about 50 watts pr. side, the model name is Beard BB-100
I doubt the 50 watts though, as I think to recal the EL84 tube only beeing capable of delivering 6-7 watts - so mabye 40 watts ?
But, that dosent matter - it sounds rather good I think.
The anode voltage ? - I've just screewed the amp back together to play music for me during me making supper - I'll unscreew it later...
Can you tell me where to measure to find the anode voltage - between gound and leg nr. ??
Mabye I should try to find the EL84 pinout.......

Thank you all !

diyAudio Senior Member
Joined 2002


While I'm looking at these curves,can you pls.measure the voltage on the anode to ground?
Sorry to have you screwing while making supper Hans...
I had this reply for you waiting in the wings already ,so having them curves here I'll wait for you.

"Bon appetit"
diyAudio Senior Member
Joined 2002


You're 100% right in suggesting this mod.
I don't argue here!
And,yes in my experience this IS a major bugger.
We're not out of the woods here and you can expect a mile long of posts here.
Unfortunately,it is not up to me to decide that.
I do not own this amplifier and I can understand people to be reluctant to carry out this kind of drastic modification.

diyAudio Senior Member
Joined 2002


In the meantime,I did the homework as promised.

When looking at the design I can only conclude the following:

-Tight matching of the tubes,where I much prefer the E84L over the regular EL84 (hoping they all age the same way as well).

-Installing some means of adjusting the bias individually on each tube.(This is what it should have been like IMHO,right from the onset.)
And, as said, Bruce is correct in suggesting this mod.
If you really like these amps,and want to keep them this would be the sensible thing to do.

Best rgds,;)
I am curious...

6 EL84's in Class A. The tube is designed for 12 watt disipation, 300 volt plate voltage, and about 80 mA max current.

In class A the bias voltage would be about 7 or 8. I do not understand where the 0.15-0.18 mA is coming from???

In any event, it would seem to imply class B bias if the cathode currrent were that low. I bias mine (only two) for 13.5 volts at the cathode for class AB. Dissipation is 11 watts, plate voltage 330 volts, idle current about 33 mA per tube.

Can you mean 18 milliamps per tube??

I have heard that paralleling them are a pain. It is because of their high transconductance (10,000µS, if memory serves).

But, push pull class A with 6 tubes would probably give you 30 watts. Class AB 45 watts, conservatively speaking. However, you should be able to bias them with about 14 volts without getting bad sound, or crossover distortion, while keeping the tubes within 5% of each other, and keeping idle dissipation low. They cut off at about 15 volts.

Right now it sounds as if they are at cut off, since you can touch the tubes (yet the heaters get pretty hot. Interesting). So yes, turn it up little by little until they start to get too hot to handle, but back off to where one of them starts to glow red in the dark. They will all still likely be hot, but the weak link will be within reasonable parameters.

My 2 cents.

diyAudio Senior Member
Joined 2002


I noticed that too but let it slip.
You're right when biased as stated in the tech manual and did the calcs you arrived at roughly 48W/Ch.
I saw many a technician tear his hair out over getting it right though.They left them running on the testbench only to find yet another tube glowing cherry red half an hour later. :bawling:
The easy way out for them was then of course to turn the bias down.
Customer wasn't worried about the tubes anymore and the tech was hoping the customer wouldn't notice the difference in sound.
You can well imagine the merry go round with this one. :mad:

BTW,you got some nice stuff on your site.


Thanks for the kind words re. my site.

I wonder if the tech bothered to purchase matched quads. This may have eased the problems a bit. Even though they would have had to buy 3 matched quads and hope for the best by splitting one quad between the two channels, I would imagine that the difficulty would be greatly reduced.

I myself have been planning on offering a 35 watt amp using a matched quad per channel. I will use regulation for the bias.

I like the EL84 very much, but haven't honestly listened to an amp with EL34's. The 6L6 is another possible candidate.

At any rate, I hope the poster has success with his amp and lets us know.

This Bias talk is very interesting. Sorry for the late reply, but I just joined and I am reading some incredible things in this forum. I am loving it here already.

Obviously I am new to valve technology and even though I do posess a bit of electronics savoury, I am a virgin when it comes to biasing.

What is it? What is bias adjusting a valve? Is this something I can do with a screwdriver or do I need specialist equipment?

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.