Biasing problem

white room · Post by **white room** » Sat Jun 25, 2005 9:33 pm

I have had a problem for quite a while now where when I set the bias on my power tubes I get quite a bit of difference in readings.

I can set the bias on V6 and V7 to 35mA but V4 and V5 will read around 24.5-25mA. I have checked both 220k resistors and they read 217k, cool. I thought that maybe one of the .1 couplers may be leaky or something so I swapped places with them to see if the readings on V4 and V5 would change places ( so to speak) with V6 and V7 but there was no difference, V4 and V5 still read low.

I'm stumpped. Am I missing something or looking in the wrong direction? Please tell me that my Dagnall clone can't be suspect.

This is something that I have been pulling my hair out over for a while now and I'm all out of ideas as to what the problem could be. It's becoming frustrating.

Chad

Flames1950 · Post by **Flames1950** » Sat Jun 25, 2005 11:53 pm

What about the screen grid resistors (the 1K 5watters) and screen voltages, are they within spec and matching from side to side? Low or missing screen voltage will cool that bias off fast. Unfortunately if the resistors are in spec but the voltages are off it may still lead back to that OT.......

Flames1950 · Post by **Flames1950** » Sat Jun 25, 2005 11:56 pm

Why not check out the plate resistors on the phase inverter 12AX7 too? The stock values can still be unbalanced enough to redplate, so maybe they could cause things to run too cool too? Check the resistors' values, but also the voltages on the plates of that 12AX7 -- make sure they're similar on each half of the tube (one side is usually slightly less than the other as the voltage charts show.)

white room · Post by **white room** » Sun Jun 26, 2005 12:38 am

The screen voltages read ~490VDC across the board and the screen resistors measure 996 ohms on three tubes and 1.2k on the fourth tube so I assume that all looks good.

The plate voltages on the PI read 251VDC at the 82k node and 211VDC at the 100k node. ( I assume those are the correct points to measure). BTW: The plate resistors for the PI measure 81.1k and 98.5k. Why do I get this nausiating feeling that my Dagnall clone is the problem?

I usually run my amp between 3 and 4 and occasionally 5 since I play mostly rhythm and get a good blend of clean/grind mix at these levels so, I didn't think I was taxing the Heyboer OT ( battle tank) that heavy.

Guest · Post by **Guest** » Sun Jun 26, 2005 2:39 am

Aside from the biasing problem, I finally had the time yesterday to install the Heyboer 100 watt PT and F&T 50/50 mains caps that I bought from George and I have to say that the PT has made a difference in this amp. The amp has a tighter, warmer tone yet, a relaxed feel ( due to the F&T 32/32 screens filter caps).

My concern is that the B+ is 490VDC ( which may account for the sweet JTM 100 like tone). I would assume that this is too high? If so, what could be the cause? If not, that's fine with me

I just don't want anything to spontaneously combust.

white room · Post by **white room** » Sun Jun 26, 2005 2:42 am

That was me getting bumped off line

Flames1950 · Post by **Flames1950** » Sun Jun 26, 2005 11:25 am

It might be nice of the PI voltages were a bit closer (the voltage chart shows 25VDC difference between the two plates, I kinda like to see even less.) Would swapping 12AX7's around balance those PI voltages better and help the bias issue? Maybe one side of your current PI tube is way different than the other and loading the B+ differently?

Flames1950 · Post by **Flames1950** » Sun Jun 26, 2005 11:26 am

And the Heyboer PT's seem to like settling in around that 490VDC range -- it's not unusual and as long as you like the sound of the higher B+ then you're cool.

white room · Post by **white room** » Sun Jun 26, 2005 12:32 pm

Okay, I had a GT 12AX7 in the PI before so I swapped it for a GE JAN 5751 and that brought the plate voltages within 9 volts of each other.

82k node = 225VDC

100k node = 216VDC

Quite a difference from the last time. I'm still getting the big differences in bias between the two sets of power tubes and I noticed just now that V6,7 drifted up to 40+ mA and V4,5 where around 33mA so I backed the bias down to get 6 and 7 down to 33.5mA but then of course 4 and 5 go down to around 24.5mA.

I'm getting worried.

white room · Post by **white room** » Sun Jun 26, 2005 4:01 pm

Maybe I'm groping in the the dark now but could one of the 10uf caps cause this problem if it's bad?

Flames1950 · Post by **Flames1950** » Mon Jun 27, 2005 12:30 am

white room wrote:Maybe I'm groping in the the dark now but could one of the 10uf caps cause this problem if it's bad?

Doubtful, since the bias doesn't split off to the separate pairs of power tubes until later in the circuit.
Are you using grid stoppers (the 1.5K to 5.6K resistors on pin 5?) Are the connections good if you are? Are they in spec? A bad connection (or, um unsoldered, in a case of mine) can drop that voltage and current down.
Or what about the tube sockets themselves -- if they're George's parts he had some runs of sockets in which pins would crack off, he had to replace a bunch from the sounds of some guys' comments here on the forum. Sockets are a bitch to replace on a whim but when all else fails it may be a shot in the dark to try out. If they're vintage maybe they need a good retensioning and cleaning.

white room · Post by **white room** » Tue Jun 28, 2005 8:52 am

Flames,

I have the stock 1.5k and 5.6k resistors and they all check out fine ( I don't know what is going on with them under playing conditions though). I have checked the solder joints on them and they seem to be good but I'll check them again and maybe even just resloder them for the heck of it. The tube sockets are stock in this SLP R/I (it's a 95) and I have done alot of tube swapping over the years in this amp so maybe they do need retensioning. I'm not sure how you go about doing that though but I know it has been discussed here before so I will do a search.

This is all new for me since I have never had the bias float around like this before so I'm greatful for the help.

Chad