Welcome, Guest. Please Login or Register
Decware Audio Forums
03/29/24 at 08:14:38 




Pages: 1
Send Topic Print
Bias meters mean...? (Read 15718 times)
hifitubes
Seasoned Member
****




Posts: 296
Bias meters mean...?
07/06/13 at 02:33:45
 
I have a question about matched power tubes. I have some from Upsacle Audio, EL34 Mullard re-issue, that are perfectly matched on the bias meters at about 40mA.

My question is, my other pairs of Black Treasures and PSVANE Type II EL34 are about 4-5mA apart.

I have read this is not a big deal and we should try for within 10% plate....

So what do the bias meters actually reveal? If we had an amp that needed manual bias, wouldn't we try to match the mA setting? Are the meters showing what the tubes are actually biased at? Or what the amp's autio-bias "has to work with".

Or is it that the tubes are more mismatched than 4-5mA and amp when biasing lands on that particular value.

Hope I made my question clear. These other Chinese tubes aren't cheap and I know their performance can vary greatly from tube to tube.

thanks
Back to top
 
 
  IP Logged
Lord Soth
Seasoned Member
****


I'm an American
BadAss!

Posts: 300
Re: Bias meters mean...?
Reply #1 - 07/06/13 at 13:20:57
 
I can't answer all your questions but I have some info which might be useful.

1. NEW (either NOS or new production) Tubes tend to drift. They need 24hrs of initial burn-in to have stable readings for the mA to settle and stabilize.
UPscale Audio does this necessary initial burn in before tube readings were made for tube matching, hence your tubes from them were perfectly matched.
Conversely, your other tube seller did not do this, hence the difference in mA reading.

2. Perfectly matched tubes are not really necessary.
A 20% mismatch was still within the acceptable range during the "Golden Era" of tube amps.
The other school of thought is that only a mA difference of at most 1 to 2 mA is the maximum allowable difference.
This is considered to be a modern concept.
In theory, only Push Pull amps require such closely matched tubes.
Single ended designs are more forgiving.

3. Yes, for Fixed Manual bias amps, mA readings are taken with a multimeter and adjustments are made.

Regards,
LS


Back to top
 
 
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #2 - 10/10/13 at 11:16:13
 
Sorry I missed your post.

So can I assume the meters show the readings of the tubes before the bias circuit?

Or the tubes are being biased, and the meters show the result, the amp doing it's best to get them both at the same operating point?
Back to top
 
 
  IP Logged
Syd
Seasoned Member
****




Posts: 1534
Re: Bias meters mean...?
Reply #3 - 10/10/13 at 19:56:09
 
I think the meters show the actual tubes `as is/are`. My Rachaels had 4 identical tesla el34`s @ 38ma. One blew, and the 2 replacements I bought showed 37 and 40ma. And LS is right,  the replacements start off at 38ma and wander slightly, noticeable after the amps have been playing  for over an hour. Makes no difference when changing each pair over l/r amps....so it`s not the amps. So hopefully they`ll settle back to 38ma after more time. But it`s no prob, the auto bias sorts it all out and the sound is well balanced.
I like the meters.

ps. When the tube went wonky I witnessed it in real time by watching the meters. A bit hairy. The two meters would just take off slowly. One would go up to 60+ma and the other drop to 15ma. So it could be that I was seeing the bias trying to compensate....meaning the meters show the bias applied ... the current fed/drawn to them.....I don`t know.
Back to top
 
 

Decware: Rachaels x 2 bridged, C. SP2+, ZP3, ZMC1, DHC-1 pw/cbl`s Michell Orbe + SME V + M.Benz LP s, Arcam CD33, Nakamichi LX5, Lowther acoustas DX2`s, WE 16g sp/cbl`s, Isotek mains substation, M & K subwoofer, Belden 8402 interconnects.
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #4 - 10/11/13 at 06:44:59
 
I watched a bad tube do the same thing!

Runaway!

My Upscale tubes don't wander, probably due to their excessive warmup before testing. I know a friend with an Amplitrex who does the same thing.

cheers!
Back to top
 
 
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #5 - 05/28/14 at 06:42:24
 
I swapped out my Mullards awhile back and have some NOS Tesla where one usually reads 32mA and the other 40mA.

Not sure what happens but every now and then things get wonky (32mA now at 28mA):

Back to top
 
 
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #6 - 06/02/14 at 06:44:38
 
There is a relationship between the two tubes. I have 3 tubes of increasing strength. Adding the 2 strongest pushes the "old" strongest one down to 36mA. Thus the need for tubes matched to within 10% mA ........
Back to top
 
 
  IP Logged
Steve Deckert
Administrator
*****


If the 1st watt
sucks why continue?

Posts: 6234
Re: Bias meters mean...?
Reply #7 - 06/02/14 at 18:58:47
 
The tubes are biased with a resistor on the cathode that connects to the meter which is connected to ground.  So the meters are showing the actual current of the installed tube.

Some of the earlier SE34I.3's (white ones) were wired with a single cathode resistor serving both output tubes.  That means the cathodes were connected to the meters then to the resistor which was connected to ground.  In this scenario, if a tube on one channel changes, the other channel changes in the opposite direction causing the actual difference between the two tubes to double.  This makes matching the tubes more important than ever.  Of course everything that goes out of here is precision matched, so I didn't notice how pronounced this effect was until someone sent an amp in for upgrades that had the customer's own (mismatched) tubes with it.

To reduce this effect, I started making the amps with a separate bias resistor for each tube so that what one tube does has no effect on the other tube.  Most white ones were this way, all black ones are this way.  To find out if your amp is this way or the "old" way, simply remove one of the power tubes and see if the meter on the other tube changes.  It should remain where it was to within a couple mA.  If it jumps way up or down then you're welcome to send the amp to us for the upgrade.  

Against my original findings when the amp was designed, I found the new way also sounds better.  Reason for the change in perception was the coupling caps are better now and made the change easier to hear.

-Steve


Back to top
 
 
WWW   IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #8 - 06/03/14 at 07:12:40
 
Mine is a white-top and behaves as you describe. I'll try to lug it back to the USA this Summer with the MT.

I had some very tightly matched tubes but once one started to drift it really becomes quite pronounced as you describe.

Back to top
 
 
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #9 - 06/10/14 at 14:01:07
 
Steve,

How complicated is this mod? Being in the Middle East now, I have so many things to bring back for upgrade or repair, I am not sure I can do it. I can't even ship electronics out of here I recently discovered. We have a hifi shop in Bahrain that has offered to help. Can you respond via email?

thanks!
Back to top
 
 
  IP Logged
Syd
Seasoned Member
****




Posts: 1534
Re: Bias meters mean...?
Reply #10 - 06/10/14 at 20:05:50
 
hifitubes, my nos teslas took a while to balance out.
Just had a look and one SE3 reads 34 and 40.
The other SE3 reads 38/38.
Another day and they all read the same.
Today is different for some reason, usually I glance at them and they are all the same.
Back to top
 
 

Decware: Rachaels x 2 bridged, C. SP2+, ZP3, ZMC1, DHC-1 pw/cbl`s Michell Orbe + SME V + M.Benz LP s, Arcam CD33, Nakamichi LX5, Lowther acoustas DX2`s, WE 16g sp/cbl`s, Isotek mains substation, M & K subwoofer, Belden 8402 interconnects.
  IP Logged
hifitubes
Seasoned Member
****




Posts: 296
Re: Bias meters mean...?
Reply #11 - 06/11/14 at 06:31:03
 
I don't see a lot of shift at all over time. I had matched Mullards from Upscale that sat at 40/40 for a long time until one went microphonic.

If I change the voltage on my P3 I see the bias go up or down a little. IIRC steve said 115-117v would be optimal for the Rachel.

I'm going to have the mod done as my Teslas keep "pushing" each other around.

I also have 3 PSVANE Type II EL34 which I chalked up as bad because one would send the bias way up to 100ma, figured it was runaway state and the replacement tube caused strange bouncing meters. I won't buy them again.
Back to top
 
 
  IP Logged
Syd
Seasoned Member
****




Posts: 1534
Re: Bias meters mean...?
Reply #12 - 06/11/14 at 11:39:43
 
Yes I`d like to get mine back for the mod and the Jupiters, which are now fitted as standard. I`d wondered why one meter goes up and the other goes down. I thought it was if one draws more current the other gets less to compensate ( auto ).
 The gap appears to show a sizeable difference when it`s probably half this as one tube would remain normal while the other has a little wander. Thankfully the performance is unchanged because of the auto-bias.
Back to top
 
 

Decware: Rachaels x 2 bridged, C. SP2+, ZP3, ZMC1, DHC-1 pw/cbl`s Michell Orbe + SME V + M.Benz LP s, Arcam CD33, Nakamichi LX5, Lowther acoustas DX2`s, WE 16g sp/cbl`s, Isotek mains substation, M & K subwoofer, Belden 8402 interconnects.
  IP Logged
Pages: 1
Send Topic Print