• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Futterman H3aa monoblocks restoration

gadget73 said:
A resistor at the cathode of either tube will read basically the same current since they are in series. Probably a few milliamps difference to account for the second screen and it's current flow but nothing significant.


problem with all of this being direct coupled is that everything affects everything else. If any of the voltages in the driver section are way off, the output stage ends up whacky too.
Click to expand...
Yep that makes sense. Working on the voltages first before anything. There are two Signalite cold cathode diodes and two zeners. Will check them all. And I've ordered all new NOS tubes for the input, phase splitter, and drivers. Those, at least. are easy to come by. I would have done it eventually anyway and doing it now reduces the variables.
 
UGH I need to not type and work... I'll revert when I actually have a minute to pay attention. But what I CAN confirm is my version has two Signalite V83R4 cold cathode diodes. And it appears I am now in possession of three different schematics for the H3aa.
 

Attachments

Last edited:
interesting, looks like a miniature version of the old gas regulator vacuum tubes with finer voltage step options. The tubes are just a gas discharge lamp, the breakdown voltage depends on the gas mix. Probably the same here, just lower current since its meant as a reference rather than a shunt like the gas regs were often used.

anyway, fail mode of neon lamps and gas tubes is eventually they don't strike at the proper voltage. If its glowing steady you're probably OK but if they are flickery its done. I've heard people claim that with neons on DC you can reverse them and get more life out of it, never tried it but maybe worth a shot if these aren't available and you need one.
 
Jfaustus said:
So I've installed sense resistors between cathode and ground on each of the three lower bank tubes. I used 0.5Ω 2W 1%.

Zeners first, which no I haven't done. Maybe then I'll try swapping and/or add sense resistors to the upper bank tubes as well. Also I took the instructions' hint to heart that the wacky reading of the bias voltage in Step 4 pointed at the 6EJ7. So I swapped it with another one I had in the stash which tests meh on the Hickok but better than the one that was in. That's shifted the bias test point from the instructions from 143V at minimum rotation to 76V with the pot at maximum rotation.

ONWARD!
Click to expand...
A better value for the cathode resistors is 1 Ohm 5 Watt... Everywhere there is a power tube there should be a cathode resistor! Also install grid stop resistors if they are not there- 600 Ohms will work nicely. That can really help with stability.

FWIW I use 5 Ohm 5 Watt resistors in my OTL builds. You'd think they raise the output impedance but they don't, since they are so much more effective at getting the individual power tubes to work better together.

If a power tube arcs and fails it is likely to burn up a 2Watt device.

Also the EL509 is a potential power tube to work with. The non-plate cap version takes an octal socket. As long as you don't run really high Voltages on it (and you won't) it appears to have the same cathode current and dissipation as the plate cap version.
 
atmasphere said:
A better value for the cathode resistors is 1 Ohm 5 Watt... Everywhere there is a power tube there should be a cathode resistor! Also install grid stop resistors if they are not there- 600 Ohms will work nicely. That can really help with stability.

FWIW I use 5 Ohm 5 Watt resistors in my OTL builds. You'd think they raise the output impedance but they don't, since they are so much more effective at getting the individual power tubes to work better together.

If a power tube arcs and fails it is likely to burn up a 2Watt device.

Also the EL509 is a potential power tube to work with. The non-plate cap version takes an octal socket. As long as you don't run really high Voltages on it (and you won't) it appears to have the same cathode current and dissipation as the plate cap version.
Click to expand...
Thanks Ralph!!! I'll take that advice and do it right away. You hit the nail on the head for why I went with 0.5Ω -- I was a bit worried about raising the output impedance. Didn't give as much thought to the wattage but will do as you say.

Going the new production JJ EL509S route is my backup plan if I can't get reasonably matched sets of 6LF6s. The issue there is I would then have to either remove and rewire the existing compactron sockets or fabricate comapctron to octal adapters. I am not super eager to do either of those things.
 
@atmasphere -- wondering whether you have a theory for why Futterman chose not to use cathode resistors to help smooth the current sharing? Cost? Heat? Impedance? All of the above?
 
gadget73 said:
guessing all of the likely candidates for compactron sweep tubes are stupid expensive now, thanks to many having been fried in crappy CB amplifiers 40+ years ago?
Click to expand...
You guessed right. There are some singles out there. Strong testing from reputable sellers are $65-$100. But of course even good testers aren't reading at the amp's operating points. So I don't even want to think about how many I would have to buy to get one matched sextet much less two. And then how long would it take to resell the rest? I'm monitoring listings and these things aren't exactly flying off the shelves.

I ain't licked yet though. Once I get to that stage we'll see how well the ones I have do. And then go from there.
 
One of these are handy if you are going to mess with sweep tubes- I've got the Futterman H3aa , H3 and a custom direct drive amp that uses 26DQ5's . There are many ways to approach this depending how far you want to go down modifying the amps.
 

Attachments

  • IMG_0325.jpeg
    IMG_0325.jpeg
    121 KB · Views: 15
Jfaustus said:
@atmasphere -- wondering whether you have a theory for why Futterman chose not to use cathode resistors to help smooth the current sharing? Cost? Heat? Impedance? All of the above?
Click to expand...
I don't. Any ideas I might have about that are colored by my own experience. But the bottom line is there isn't a good reason. Its cheaper to put them in because you have less tube failure. That means that the stronger tubes don't run as hot (so less heat), and of course the output impedance winds up the same or lower. Win win win, so much winning :)
 
atmasphere said:
I don't. Any ideas I might have about that are colored by my own experience. But the bottom line is there isn't a good reason. Its cheaper to put them in because you have less tube failure. That means that the stronger tubes don't run as hot (so less heat), and of course the output impedance winds up the same or lower. Win win win, so much winning :)
Click to expand...
I like winning!!! I would have done it in any case but I was curious -- her certainly knew he could have used them and actually went to some lengths NOT to. But then again he was designing in a era when tube swaps didn't mean much
kcin said:
One of these are handy if you are going to mess with sweep tubes- I've got the Futterman H3aa , H3 and a custom direct drive amp that uses 26DQ5's . There are many ways to approach this depending how far you want to go down modifying the amps.
Click to expand...
That looks really cool! But I guess that now I'm installing cathode resistors I don't need it as I'll be able to measure directly across those. The amps are out of the chassis for the foreseeable future. And once they're ready to go back in I plan to make a new one and preserve the originals in storage. I'll probably pull leads from the cathode resistors to test points so I can easily keep an eye on whatever tubes I end up with.

I don't mind modding to an extent. It's not the work that bothers me; but I want anything I do to be reversible with not too much difficulty.
 
Pin jacks? I install those on amps sometimes, makes it easy to adjust things without having to open it up or hold probes in place.
 
atmasphere said:
We've installed test jacks on some of our amps that allowed you to plug in DVM probes, making testing the tubes fairly easy.
Click to expand...

gadget73 said:
Pin jacks? I install those on amps sometimes, makes it easy to adjust things without having to open it up or hold probes in place.
Click to expand...

Something like that! And we will see what the tube situation is when I get there but with the grid stoppers, cathode resistors, and pulling to test points I’m already halfway to per-tube biasing…
 
gadget73 said:
Ah Rosenberg, he was also involved with New York Audio Labs, who made the Miscode stuff. I have one of their preamps. Sounds good, no documentation to speak of, and not great construction. Had a hell of a time getting that thing quiet with all of the bad solder and cracked traces.

I think George Kaye was the designer on that stuff
Click to expand...
Was that Moscode?
 
Good news! Not there yet but progress.
  • Pulled the cold cathode diodes out and tested them, They clamp like champs!
  • One of the channels is missing one of them and I managed to find a source so I got a handful. Why not? They're cool.
  • Replaced the input, phase splitter, and driver tubes with NOS.
The voltages are behaving (for the most part)! The node (C12) is at 195V where 182-ish is half of B+ where I believe it should be. I was able to bias the input tube at the 95V called for in the instructions, and the upper and lower bank screens measure close enough to each other (2.4-ish V when the instructions call for max of 2). Remaining is the "balance" which is supposed to be 0 but stubbornly sits at 80V. I believe this is the potential between the upper and lower bank.

The instructions indicate that this is a power tube issue (surprise surprise). Unmatched and/or weak tubes would swing the voltage.

So good news is I think everything except the power tubes is behaving (famous last words). Now all I need to do is hope against hope that I have 12 suitable tubes among the 20 or so that I have. If not then it's wait and monitor listings or pivot to one of the backup plans.
 
So I just talked to a tech who used to work for NYAL, and later for George Kaye, and had serviced Futtermans for quite some time. He walked me through exactly how to go through tubes to find ones that work.

So there’s an upper bank and lower bank. The three tubes in a bank are in parallel and then the two banks are in series.

He recommended starting with just two tubes, one in each bank. The ‘balance’ pot adjusts the grid of the lower bank. Get the two tubes in balance, don’t touch the pot, add two more. If they skew the balance one or both are off. Get the second two in order, then go for the final. Same procedure.

I’m out of town for the weekend but will do that when I’m back with the tubes I have. Fingers crossed I can get at least one channel.
 
You must log in or register to reply here.
Back
Top Bottom