MC225 Fixed, Adjustable Bias

[rjb5191]

An easy and convenient modification after changing the selenium rectifier to silicon. Just splice in a 5K trimmer between R46 and either solder terminal. No need to mess with the tedious terminal solder joints. Bring up the amp on variac and adjust the variac until the plate voltage reads the schematic indicated 390 VDC. Adjust the trimmer until -21 VDC is indicated at the pin 6 grid.

I know many people would recommend just replacing the resistor with the appropriate value but I happen to prefer this, thus the share.

 

[nerdorama]

I have adjustable bias in mine but it is part of a PC board that completely replaces the power supply. @rustnrot designed it.

 

[ramjet]

I have adjustable bias in mine but it is part of a PC board that completely replaces the power supply. @rustnrot designed it.
View attachment 3331304View attachment 3331305

Still waiting for one.........................................

 

[rustnrot]

Ha! I finally had a bunch of these boards just made. PM incoming.

 

[nerdorama]

I wasn't intending a poke in the ribs or anything like that. Just sharing. I think your board is a slick deal.

 

[rustnrot]

I took the humor exactly as you guys intended.

When I sent off the files for the MC275 main board I decided to do a couple of fixes on the MC225 power supply board (you know, same price for shipping...might as well throw another board in the order). Just today I was looking at that board and thinking since I don't have a MC225 anymore I need a beta tester to be absolutely sure it works. ramjet is the beta tester!!

Anyway, all is good and I certainly can't complain as we are on a mountain ridge here in Asheville and didn't get flooded nor even a tree on our roof. We consider ourselves lucky even though 2 weeks without power and 3 weeks without water.

 

[ramjet]

I took the humor exactly as you guys intended.

When I sent off the files for the MC275 main board I decided to do a couple of fixes on the MC225 power supply board (you know, same price for shipping...might as well throw another board in the order). Just today I was looking at that board and thinking since I don't have a MC225 anymore I need a beta tester to be absolutely sure it works. ramjet is the beta tester!!

Anyway, all is good and I certainly can't complain as we are on a mountain ridge here in Asheville and didn't get flooded nor even a tree on our roof. We consider ourselves lucky even though 2 weeks without power and 3 weeks without water.

Just finished the 2nd re-cap of my Marantz Model 8 in thirty years so I needed to move on the chassis transplant and new power supply for the Mc225. I do feel privileged to be the beta tester. I have Sheldon Stokes power supply PCBs in my two Citation Vs and my PAS3 and like the all in one concept they deliver. Purists, gnash your teeth here.......

I did wonder how you fared in the hurricane, glad you came out the other side as you did. We here on the coast of Virginia are past our due date for our hurricane comeuppance.

 

[rustnrot]

My Cit V biasing board and a Cit II all-in-one power supply with bias transformer on the board (bias windings high failure rate in power transformer)

 

[gadget73]

Having it adjustable is pretty useful, much less annoying than hand-selecting resistors.

Doing it by grid voltage is not always ideal though. I'd probably be going by cathode voltage instead. That will get you the right current flow, even if the tubes are goofy like modern ones tend to be and don't flow what a selected set from McIntosh would have in the 60s.

also just to be the eternal contrarian, I'd put it in series with either R47 or whatever the 22k to ground is (schematic is fuzzy). That way if the pot wiper lifts, the grid bias goes more negative rather than to zero. I like fail safe stuff, and pots are not way up on my list of most reliable components to exist.

 

[rjb5191]

Having it adjustable is pretty useful, much less annoying than hand-selecting resistors.

Doing it by grid voltage is not always ideal though. I'd probably be going by cathode voltage instead. That will get you the right current flow, even if the tubes are goofy like modern ones tend to be and don't flow what a selected set from McIntosh would have in the 60s.

also just to be the eternal contrarian, I'd put it in series with either R47 or whatever the 22k to ground is (schematic is fuzzy). That way if the pot wiper lifts, the grid bias goes more negative rather than to zero. I like fail safe stuff, and pots are not way up on my list of most reliable components to exist.

Thank you, your comments make sense. I think before adjusting the bias, the cathode voltage was lower than the expected 0.8V (closer to 0.3V) with the original mcintosh 7591 which tested great. I'll have to check again now that the bias is adjusted. I agree with your comments regarding the failsafe but I glazed over this in my zeal. The bourns pots are pretty good (usually) so I won't be sweating over this immediately.

 

[gadget73]

Good pots definitely help. I just figure if I'm doing it, I design like I'm using garbage pots that are definitely going to fail, and then don't use garbage pots. Makes it unlikely to fail in the first place, and then just in case it happens anyway there isn't anything catastrophic to deal with.


might be worth confirming the DC resistance from cathode to ground to see if 0.8v makes any sense. Not impossible its a typo on the schematic, it happens.

 

[ramjet]

Still waiting for one.........................................

Happy to finally report that I have finally finished my Mc225 with rustnrot's PCB power supply with adjustable bias. Gallbladder surgery/recovery doesn't always go as planned by the way....
Anyway, it fits inside perfectly and leaves room to work that was never there with the original setup. Amplifier is dead quiet with no signal and with music it is magical, like a Mc225 always is.
Thanks Tom!

 

[rjb5191]

Good pots definitely help. I just figure if I'm doing it, I design like I'm using garbage pots that are definitely going to fail, and then don't use garbage pots. Makes it unlikely to fail in the first place, and then just in case it happens anyway there isn't anything catastrophic to deal with.


might be worth confirming the DC resistance from cathode to ground to see if 0.8v makes any sense. Not impossible its a typo on the schematic, it happens.

I'd have to assume that it's quite a low resistance through that transformer winding to ground so only 0.8v seems reasonable. That would indicate my actual bias current is low even though my grid voltage is right on. They are original Mcintosh branded 7591. In my limited understanding, the unity coupled circuit isn't super dependent on high bias current to meet spec but i may still crank it up a bit to see if i can hear a difference.

 

[ramjet]

I'd have to assume that it's quite a low resistance through that transformer winding to ground so only 0.8v seems reasonable. That would indicate my actual bias current is low even though my grid voltage is right on. They are original Mcintosh branded 7591. In my limited understanding, the unity coupled circuit isn't super dependent on high bias current to meet spec but i may still crank it up a bit to see if i can hear a difference.

I measured 38.6 and and 36.3 ohms for the cathode windings to ground on M220B and 36.1 and 38.2 ohms on M220A.

 

[rjb5191]

Cool, so at .8 volts across the cathode we are looking at 21-22 mA cathode current. There’s likely very little screen current under typical operating conditions. Even if we assume 25 mA plate current, we are looking at cool class AB bias.

My amp is likely running much less current with only .3v on the cathode (depending on winding resistance). I would definitely like to experiment with raising the bias a bit. It will be interesting to take distortion measurements and listening impressions at different bias currents.

 

[gadget73]

My admittedly limited understanding of the Unity Coupled output stage is they tend to run at fairly low idle current compared to "typical" designs. 0.3v / ~37 ohms is about 8ma, pretty close to class B operation. I don't know if thats right or not honestly but I expect the distortion measurements would tell you if that or 0.8v / 22ma makes more sense.