Sherwood S5500 II w/7868s- take two

[oldman55]

My cathodes are tied together and a wire is run across the chassis for a ground. The tube sockets do not have ground tabs.

Could add a wire and run them to the nearest ground but I dont like to run them without an anchor point. Just not laid out where it is practical.

 

[oldman55]

Think this one is done. Singing nice music. The Hayseed 30/30/30/30 allowed me to just add a 50/50 (supplemented) to the HV+ supply to keep cost down.

And have bias adjustment.
Will leave this one bottoms up as I may think of an easy way to add cathode resistors. And they look better that way usually.
Thanks to all. Another great sounding Sherwood.

Edit: PT started smoking. Too hot to the touch. Should have flipped er over. Something not wired right, but HV+ was on the money.

Edit 2: there was a resistor that really wanted to go to ground. I took it home and all is fine again.

 

[gadget73]

Looks good.

 

[oldman55]

Finally realized that I added 10 ohm resistors on this exact same unit (7951 version). And added screen resistors when the cathode resistors started popping when the tube(s) arced.

So...did it again and when biased to -17.35, I am getting .530/.515v and .380/.220v on the other channel. With 410v plate voltage, that would put the two NOS tubes at about 19 W. How difficult would it be to split the bias supply to provide different voltages to each channel? And at what point is the tube worthless where you can't adjust the bias enough without them distorting too much?

 

[gadget73]

not terribly hard to split it, and you can make it individual if you'd like, all depends how much space and trouble you want to go to.

Some Sherwood S-5500II info

thats what I did with mine to allow each tube to adjust. The set of tubes in that amp are very mismatched and it needed all the help it could get. I have another set that ends up with the balance controls all pretty much in the middle though.

Not sure what the magic point is honestly, but there is one. Probably under 15 volts means its getting fairly worn out.

If you just want to split it, basically take R72 leftward and copy it, meaning you'd have two 18K resistors off the diode, and all the other components would duplicate in the same way. Feed one to the left channel and the other to the right channel. You'll have to pull the wire that connects the two channels together for the bias supply too. Its probably in the back at a terminal strip where the grid resistors land if I had to guess.

 

[oldman55]

not terribly hard to split it, and you can make it individual if you'd like, all depends how much space and trouble you want to go to.

Some Sherwood S-5500II info

thats what I did with mine to allow each tube to adjust. The set of tubes in that amp are very mismatched and it needed all the help it could get. I have another set that ends up with the balance controls all pretty much in the middle though.

Not sure what the magic point is honestly, but there is one. Probably under 15 volts means its getting fairly worn out.

If you just want to split it, basically take R72 leftward and copy it, meaning you'd have two 18K resistors off the diode, and all the other components would duplicate in the same way. Feed one to the left channel and the other to the right channel. You'll have to pull the wire that connects the two channels together for the bias supply too. Its probably in the back at a terminal strip where the grid resistors land if I had to guess.

Thanks for link. Good reading but was hoping you would of finished. Think I will try a new pair of 7868s and see what develops. Cant find anywhere where an opinion is given as to when to trash a tube. I have boxes of questionables as I am sure you do too.
Back to my speaker projects which I dread. If it cant bite you....then half the challenge is gone.

 

[oldman55]

If you just want to split it, basically take R72 leftward and copy it, meaning you'd have two 18K resistors off the diode, and all the other components would duplicate in the same way.

In looking at this closer, its just a matter of a couple of resistors, a trimmer and a 50uf/50v cap. I should have enough room (and parts) for that.
Edit: at -15v I can get at least one of the weak tubes up to .5v. Dont know why they didnt build all amps with separate biasing since it only takes a minimum of parts.

 

[gadget73]

Some amps had no adjustments at all. Its a minimum of parts but if you save 10 cents on 10 units, thats a dollar. Economics of scale.

 

[oldman55]

I went ahead and added a parallel circuit for the other channel but I am only getting -8v in lieu of -16. I am starting at the same -28v (previously -30) location as the other channel which is getting the -16v. Resistors and trimmer check out. Capacitor is neg to ground. Cant figure it out.

 

[gadget73]

did you split it before or after R72 ?

 

[oldman55]

Before. At the diodes with another 18k ohm forward. Resistances along both circuits are the same and the capacitor tests at 51uf. Going straight to chassis ground.

 

[gadget73]

that doesn't seem to make sense if the circuit is identical. You should get the same results out of the same components.

 

[oldman55]

This is what I added, pretty simple. Could it be the cap? That is the only thing that differs in the two circuits though the radial one added is on the money and I assume the can cap is as it has design voltages.

Edit:
L Channel has range with 10k trimmer of -16 to -19 (delta 3v). This is the one with the original can cap.
R Channel has range with 10k trimmer of -9 to -10 (delta 1v). This is the one with the new radial cap.

 

[gadget73]

Could be. All this spells either leakage in that cap, the 18K is way high in value, or the resistor and/or pot are lower in value. Its just not that complicated of a circuit, it has to be something simple.

 

[oldman55]

Resolved...I think. Cap goes positive to ground despite what the schematic shows and despite what the parallel circuit does? Since the other channel (original circuit) goes on the "triange" tab of the common negative multi cap can, it is positive in the circuit so why would this circuit be different from the other?

So I am confused again.

Edit: It is singing with little hum. After all the heartburn, the channels are set exactly the same. But I have one weaker tube on L channel. Guess I would need to squeeze in a balance circuit.
Doubt if its worth it.
I tore this circuit apart countless times testing each components and combinations.

Dual resistors was because I was out of 18k ohm. 47mf cap is pos to ground.

 

[gadget73]

yeah, bias supply needs the positive to chassis ground since the supply is negative. Its usually marked as such, or at least indicated on the can but either way it just has to be that way or it won't work correctly.

 

[oldman55]

So the schematic is in error? And the can is common positive?

 

[gadget73]

the bias capacitor is common positive. Not sure what else lives in that particular can, but at least that section is.

 

[kvflyer]

So the schematic is in error? And the can is common positive?

I am not sure about this specific schematic but, I remember reading within the last year about a schematic which had the polarities reversed. And for sure, the positive ground has "bitten" more than one person. HH Scott uses a four section Positive ground electrolytic can.

 

[oldman55]

The copy of the schematic for the 7868s (which is the one I am working on) is incorrect as the bias/heater can is positive common. Glad I didnt order a custom can for this spot also from Hayseed as I would have used the schematic.
My unit is somewhere between the this schematic and the one for the 7951s.


Now that I have twin bias trimmers, my current draw across the 10ohm cathode resistors:

L channel: -17vdc bias .56 and .40 v
R channel: -17.25vdc bias .45 and .44 v
Plate is 406v

Without adding a balance pot, wouldnt it be easier to just adjust one of the two last resistors of the one "out of line" tube by tagging a resistor onto it? I know this would require changing the resistor over time but there is little real estate left to add two more circuits.
Or leave well enough alone until I can match up another tube (would require shuffling among several amps). Didnt know if the one tube was running too hot.
Its my understanding that even a mu tester wouldnt be accurate for operating conditions.