Kalamazoo Blue Smoke Fun

[SteveInMN]

Schematic, VDC readings  and Power Tube Data Sheets attached!

I found that the JJ 7591 power tubes that were in Kalamazoo Bass 30 amp were grossly mismatched (or one had outright failed), as the bias balance pot could not even out the MVs running through the two power tubes.

I replaced them with matched Tung Sols, which I 'believe' were running hot at about 50MV. So, thinks I , I will reduce the current by adding a 150 ohm resistor between the bias pot and ground. This reduced the MV in each tube from ~50MV to about 33MV and worked pretty well for a few weeks, till I added a Micro OD pedal which immediately caused the resistor to smoke. I am no expert, but am made to understand this is 'not good'. 

No harm done - the resistor, amazingly, still meets spec (yes, I will replace it).
 
So.

"What type of amp is this?"
You would think this would be straight forward, but there are different bias numbers for different types of amps (A fixed, AB fixed and cathode biased) and I do not know what a Kalamazoo 30 is and internet research has not helped much. In fact people seem to argue about this seemingly basic factual item, which I find astonishing but there it is.


"What should the MV be at the bias probe, given the attached voltages?"
I'm thinking once I know the amp type, this should be self-evident from the Robinette calculator. But, really, what do I know?

Finally, and critically -- 
"How do I set the bias?".
I now 'think' this involves changing resistors before the power tubes. But I also thought that putting one between the balance pot and ground was a swell idea, so there's that.

Any light shed WELCOMED!

 

[Dave]

According to the schematic, the bias is fixed and you would have to modify it in order to make it adjustable. The amp is set up with a "Hybrid" bias system with a pot to adjust/balance the cathodes. The pot (probably a rheostat?) that is present is for balancing the bias by balancing the cathode "resistors" which are each outer leg of the pot and wiper to ground. The old tubes may not have been so far out of match, maybe the cathode pot was just set wrong.

First thing I would check is that you are actually getting negative DC voltage on the grids.

Dave

[SteveInMN]

According to the schematic, the bias is fixed and you would have to modify it in order to make it adjustable. The amp is set up with a "Hybrid" bias system with a pot to adjust/balance the cathodes. The pot (probably a rheostat?) that is present is for balancing the bias by balancing the cathode "resistors" which are each outer leg of the pot and wiper to ground. The old tubes may not have been so far out of match, maybe the cathode pot was just set wrong.

First thing I would check is that you are actually getting negative DC voltage on the grids.

Dave

Hi Dave, right, you might not have seen the attachments, -17.5 VDC at the grids.  Yes, the bias pot is nothing but a balance pot -- I ran out of room to adjust, and the old tubes were still very far apart. The new ones cured that part. Now, I'm trying to get the bias to the point where it does not trash the new tubes!

If this IS a fixed bias amp (I'll assume AB, as A are uncommon - correct if wrong!) --- what  do I modify to dial it back some? I have a decade box, I just don;t know where to apply it!

[passaloutre]

The 47k in the bias circuit would be a good place to start.

[AmberB]

You could replace the 47K resistor to ground in the bias circuit with a 100K pot and take the bias voltage from the wiper on the pot.  That will give you the ability to adjust your bias voltage to find the bias value that you want.  If you don't want to leave it in, once you've found the bias you want, measure the resistance from the wiper to the ground side of the pot and use a resistor of the closest value in place of the 47K resistor.

[AmberB]

6EU7s and 7591s...you'd think Bogan built that...

[Dave]

The resistor that was connected to the bias pot, the one you installed... I would take it out. If you put it in there, you go from having a small resistance there for balancing purposes to having an amp that is truly hybrid between cathode and grid biased.


I would look at the solution that Amber suggested to get more negative bias voltage onto the grids.


Dave

[SteveInMN]

Thanks all. This a truly helpful board, and I do appreciate not being shamed for not knowing stuff.

So the consensus seems to be REDUCE the resistance in the bias feed to provide more negative current to the grid. I can do that, I've got a decade box and lots of resistors.

To be super clear, reduce the value of the 47K -- NOT either of the nearby 4.7K's (4K7 for any Euro friends) - the 47K that parallels a 20U cap (diagram attached!).  And of course, remove the resistor I had installed between the balance pot wiper and ground. Do I have this right?

[sluckey]

So the consensus seems to be REDUCE the resistance in the bias feed to provide more negative current to the grid. I can do that, I've got a decade box and lots of resistors.

To be super clear, reduce the value of the 47K -- NOT either of the nearby 4.7K's (4K7 for any Euro friends) - the 47K that parallels a 20U cap (diagram attached!).  And of course, remove the resistor I had installed between the balance pot wiper and ground. Do I have this right?

NO! Reduce the value of that 390K just to the left of the diode in your schematic.

[SteveInMN]

OK, Steve, thanks for that.

I just took voltage reading from various points around the bias feed area (diagram attached).

It looks like changing the:
* 47K would have no effect, as there is no voltage on the far side of the resistor.
* 4.7K (4K7) would have a max effect of around an additional negative 2 volts, and would change the balance between a couple of cap feeds (if that matters).

So I will:
* Remove the 390K,
* Insert the decade, and
* Adjust until the MV on the bias probe reads what it should for AB fixed with 7591's with 450VDC Plate to K (23.3MV @ 70% -- up to 28.8MV if you believe Blencowe that 85% is OK for AB fixed...).

Sound like a plan?

[Esquirefreak]

The 47k bleeds bias voltage to ground. Higher resistance there means more negative voltage to the grids. It's a voltage divider.

/Max

[SteveInMN]

The 47k bleeds bias voltage to ground. Higher resistance there means more negative voltage to the grids. It's a voltage divider.

/Max

Ok - so  get in the rough area by adjusting the 390K downward - then fine tune the 47K value?

[Esquirefreak]

I'd probably put a 10k-50k pot in series with the 47k resistor ground to adjust the bias voltage.

/Max

[SteveInMN]

Thanks all, I put 268K of resistance where the 390K was, down to 23.5MV current draw.