Bias changing with PPIMV settings

[Cless357]

Hi, I have a Ceriatone plexi clone, with a Lar-Mar PPIMV. It's a 220k double pot (theorically, in reality one pot is 200k and the other one 240k, so in the -10%/+10% tolerance).
I checked the bias at bedroom volume, it was around 30-33 mA for al power tubes (Ceriatone advises 35mA for 65% dissipation). So I cranked the PPIMV (I think the bias should be adjusted with the PPIMV maxed ?) and I see that it is now 35-36mA for 2 power tubes and 45-47mA for the 2 other ones !!!
Is it because of the tolerances of the two pots of the PPIMV ? So what bias should I aim ?

 

[311boogieman]

I'm not the right guy for this but since I took the time to read this and try to understand I will only make a few comments:

1. Generally, a plexi type amp on EL34s will hit the sweet spot around 35% mA.
2. That however depends on the plate voltage. What is your plate voltage?
3. Every amp tech I've spoke to says do NOT play through your amp to read and adjust bias.
4. I put the volume on zero and unplug the guitar when I bias my plexi style amp.
5. Can't speak to Ceriatone as I've never owned one, but if you pull two power tubes then you are running at half power and should therefore drop your output ohms in half and I would not expect the MV pots to affect the bias, but since you are cranking to volume - I have no clue.

tl;dr - sorry you are having trouble, hope you get it resolved, and bias with no signal going through amp

 

[harddriver]

Bias should be checked with the amp at idle with no signal going through it. All volume and tone controls turned to zero and I can't recall if the PPIMV should be turned all the way up out of the circuit while you set the bias or not. Since bias is set at idle with no signal where the PPIMV master is set should not matter...maybe someone can confirm about the Lar/Mar concerning biasing.

The current being pulled through your individual tubes should not be affected by the PPIMV as long as the amp is at idle with no signal going through it while biasing. Bias and B+ voltage will vary if you monitor it while playing.

Now if the amp is a plexi and ran hard at high volume through an attenuator the V4 and V5 power tube elements may glow a little brighter(not redplating) than your V6 and V7 power tubes because the Marshall phase inverter drives V4 and V5 harder by design and that's normal.

I found an interesting topic on the old Metro forum that may or may not be helpful but the problem sounded similar to yours......Lar/Mar bias question.

http://forum.metropoulos.net/viewtopic.php?f=28&t=35220

Excerpt from thread:

I've had a lar/mar in my 100w 69 metro build for a year or so without any problems. Recently, when I cranked up the MV, I got red plating on v4 and v5. After checking it out, I noticed that the lar/mar was controlling the bias voltage to v4 and v5, but had no effect on v6 or v7. So when I cranked up the volume with the mv, I was seeing the bias go way up at v4 and v5 as well.

I am thinking the outer or back pot on the dual gang pot is the problem since this is the pot connected to v4 and v5. The pot tests ok in terms of resistance.

I'm going to try an new pot, but wanted to see if anyone had any other thoughts.

Thanks

After troubleshooting the amp owner found this:

So as a follow up, I was getting ready to change the pot on my LarMar and decided I better just wire it up without the MV to make sure there isn't something else wrong. Well, I did and there is a problem. For some reason, after the amp is on for about 10 to 15 minutes or so, the negative voltage at pin 5 of v4 and v5 starts dropping (or should I say rising since its negative?) to near zero and the redplating begins. Not sure what to look at next, but any advice would be very welcome.

UPDATE: I resoldered the entire bias section and it seems to be working now. I've tried it for a few days without the LarMar installed and have no signs of redplating. I put the LarMar back in and it seems to be working fine as well. Must have been a cold solder joint or something in the solder connection I guess. As a side note, I always has a problem where the most negative voltage I could get on pin 5 was -32v and when I was looking at the bias section, I noticed the bias resistor in series to the pot was measuring only 44K. I switched that out for a 56K and now I can get -50v on pin 5.

 

[glpg80]

The bias taps to the grids don’t conduct current however from the power tube’s perspective, the bias supply does look like small signal ground which is why the volume design works at all in the first place. What you’re seeing are the grids loaded with different shunt impedances to small signal ground which is showing as a change in quiescent bias. Datasheets callout a max shunt impedance but by design it’s expected they match in the power side or else you’ll get varying performance.

Yes both pots should be as close as possible as to load grid pairs equally. To test if this is the case, try disconnecting the lower resistance side and placing a 250k resistor directly in its place and check the bias of both sides afterwards. I’d bet they’re both closer.

 

[GlideOn]

I do think that too extreme a difference in dual pot value can affect the sound quality of the power tubes, yes. I've myself recently noticed with pots like Bourns that have taper issues and still leave a little resistance when they are supposed to be "off" and yet bleed signal, comes in the form of artifacts in the sound especially at low volumes. Yes, pot quality can matter to PPIVMV (I think more taper consistency moreso than value). Btw I settled on a RichMod 500k dual which I think sounds better as the grid leak value changes from fat to lean with the volume sweep.

Otherwise it sounds about as well biased as one could hope for and yes - the amp is supposed to be 'fixed bias' but an amp is a dynamic thing and is never truly fixed. It's still 'fixed' in place by physical components such as solder, wire, resistors, potentiometers if it is adjustable bias - these thing can get hot, form a bad connection, drift in value too just like anything else.

Also as @harddriver pointed out the nature of the phase inverter driving the push/pull power tubes is inherently unbalanced, meaning one side drives the output section slightly hotter than the other.

Greater volume, greater demand on the transformers, hit a complex chord and the voltage, current rises to meet that demand, bias can bump a bit too. Tubes can redplate momentarily during this peak demand, especially new production tubes.

Hell, the 120v coming into to you amp can fluctuate between 110 and 125 some days, I know mine can.

Nothing is ever a perfect scenario for amps. We are sometimes forced to rely on components and tubes being perhaps a bit forgiving in their advertised ratings and sometimes work a bit beyond their specifications. Even major amp makers like Friedman struggled with this for years because of inconsistently made tubes, so you're not alone.

If you really want to get technical, you could potentially come up with a Zener Diode network that limits the bias voltage to the tubes, behaving somewhat like a cathode bias in operation but could potentially extend the life you your tubes more.

I couldn't tell you off the top of my head a specific example, but I've read somewhere in the Merlin pages and HiFi forums about performing this for cathode bias power tubes and theoretically no reason it couldn't work for push/pull either. You could Zener Diode almost anything, really; heaters, B+ line, various points in the signal (as long as the wattage is high enough), albeit the values would have to be calculated pretty precisely for fixed bias...and probably not be made from Germanium

 

[Purgasound]

I got tired of dealing with bad pots. Seems the quality of the dual pots doubles your chances of failure or bad tapers! I've had to swap out too many ones that were brand new. I started checking them with a dual meter to ensure not only was the resistance matched but the taper as well. In the end, I felt it was too much trouble for the results I was getting. Now I just stick with pre-phase masters and I'll suit the preamp to emulate the desired power amp distortion to the best of my ability.

You are reporting the bias reading as current so am I correct in assuming you are using probes? What is the negative voltage present at pin 5 when the tube pairs are showing the disparity of current draw?