Pilot SA-232 Tube Amp Restoration Thread

[BassKulcha]

As I mentioned in an earlier thread, I picked up a Pilot Console and pulled the SA-232 amp, SP-210 preamp, and a Garrard RC 88/4 turntable. I'm still really new at restoring tube gear, so I thought this would be a good opportunity to document my progress and get feedback. I plan to update this thread as I move forward with the project, with completion hopefully within a month or so (as I find the time—)

I've been slowly ordering parts and doing research on restoring the SA-232 and was able to get some work done over the last couple weekends. I'm still waiting on resistors from Mouser as well as a few other parts, but wanted at least get the power supply installed. (I probably should've waited to install the capacitors, but when I received the shipment of K42-Y PIOs, I couldn't help myself. There are a few out of tolerance resistors that share the same connections as the caps, but I guess I'll just need to cross that bridge when I get to it—)

A few questions:

• Even though there are some out of tolerance resistors, is it ok to bring this up on a variac/DBT at this point? I've already done a good bit of work, so it seems smart to stop and check functioning at this point, yes? And I should first bring it up with no tubes installed, correct?
  
  
• Are there any basic voltage checks I can do with a DMM during power up?
  
  
• The original topside two-section can capacitor was 80-40 uF. I replaced with a 40-40-40 uF and bridged two of the 40uF tabs. Originally two wires went to the original 80uF tab—it doesn't matter that now one wire goes to *each* of the new bridged 40uF tabs, right? (I'm assuming they are combined anyways...)
  
  
• Anyone know exactly which of the preamp out plugs need to be shorted in order to operate independent of the SP-210?

A few before pics:





And some mid-restoration shots:

 

[mibag6]

I rebuilt one of those a few years back and had to replace pretty much everything underneath, wiring, resistors, caps. Made a few upgrades that I don't remember off hand but you might want to contact Jim McShane for that info. I remember I used up just about all the available space under there. Yours is looking good so far but it's way too clean and hardly any challenge to work on compared to the toasty amp I had.
I don't know how many out of spec resistors you have but I would be inclined to replace any that are way off, like more than 20%.
Pins 4-5, 7-8 for shorting plug.
That plug is shown on the schematic if you have one. If you don't, you can get a copy at the Yahoo Pilotone group or Radiomuseum.com.

 

[BassKulcha]

Pins 4-5, 7-8 for shorting plug.
That plug is shown on the schematic if you have one. If you don't, you can get a copy at the Yahoo Pilotone group or Radiomuseum.com.

Funny, right before you responded, I went back into the schematic and re-read the shorting plug info and sorted it out. I made little copper shorting wires from stripped romex that fit snugly and then used heat shrink to cover up the exposed parts (next time I would just keep the exposed section of romex un-stripped). I think it'll work fine (and be safe) until I can find a real shorting plug.

From my resistor readings, I already know I'm going to need to replace at least eight (and possibly more...) I ended up ordering a bunch of resistor values from Mouser, so I theoretically could replace them all. It's pretty challenging working in such a confined space, though, and especially with the leads wrapped up like a Christmas bow, lol. I've already nicked a few other wires and I'm worried about overheating resistors (even when using a clamp as a makeshift heat sink...) I'm going to have to think hard on whether I should just be conservative and only replace parts that are clearly out of tolerance...

 

[mibag6]

You can always clip the leads off long enough to put little hooks on the end then hook the leads on the new parts to the old leads and solder. If you squeeze the hooks fairly tight it doesn't look bad and saves you from cooking the terminals and components.

 

[gadget73]

I've done the hook and solder method a lot of times, especially on things that need really long leads to connect stuff.. I prefer to not do it, but sometimes you just don't have an option. If the choice is splicing or risk damaging a tube socket, I'll splice.

 

[BassKulcha]

You can always clip the leads off long enough to put little hooks on the end then hook the leads on the new parts to the old leads and solder. If you squeeze the hooks fairly tight it doesn't look bad and saves you from cooking the terminals and components.

I've done the hook and solder method a lot of times, especially on things that need really long leads to connect stuff.. I prefer to not do it, but sometimes you just don't have an option. If the choice is splicing or risk damaging a tube socket, I'll splice.

Yeah, I really want to do a thorough, clean job on this, but I'd rather it look not as tidy as I'd like than risk burning stuff up. The hook and solder method may be in my future unless I come up with another solution...

(I was flowing new solder into the old, dry joints, but then I was dealing with 3 or 4 more passes with the soldering gun and either the solder sucker or a solder wick in order to get to places where the clunky sucker tip couldn't get to...not good and waaaay to much heat I think...)

 

[century tek]

FYI,

For capacitors connected in parallel, the voltage equals the lowest voltage rating of either capacitor. For example, wiring two 22-mfd/150-volt capacitors in parallel results in a 44-mfd/150-volt capacitor. Both voltage ratings are equal, so the resulting voltage is 150 volts. If you wire in parallel one 22-mfd/150-volt capacitor and one 22-mfd/35-volt capacitor, the resulting capacitor will be 44-mfd/35-volt. The capacitance is doubled and the lower voltage rating is 35 volts.

 

[BassKulcha]

I haven't replaced any resistors yet, but I wanted to see if the work I'd already done was solid, so I ended up bringing this up on the variac and current limiter last weekend, first with no tubes, then with all tubes in place (and with an 8Ω dummy load).

The good news: No smoke, no explosions, woo hoo! So, I took the DBT out of series and brought it back up with only the variac.

The bad news: At full wall voltage there was redplating again in one the EL-84s. Bummer. I'd assumed that the original issue was likely a bad cap leaking DC, so now I'm not exactly sure how to proceed...

Next steps:
1. From this Blueglow video on restoring an SA-232, it appears that out-of-spec 270Ω resistors between the pin 9 EL-84 screen grids might be the cause, since these also play a role in biasing the tubes. If so, should I go ahead and replace those resistors first, before moving forward?

1. My plan is to go back and measure voltages using the process outlined by @s-petersen in this recent post about starting up a Pilot SA-260. I'm not exactly clear on how to check voltages at the plate (or what the voltages should be), and how to get the bias voltage at the tube sockets, but I'll go back into the schematic and try to sort it out. I'm still a newb with checking voltages, including filament and B+, so it might take a while lol.

Sound about right?

 

[gadget73]

Try moving that tube and see if the problem goes with it. That will tell you if its a tube issue or a component underneath.

High resistance on the screen will make it conduct less current, so that wouldn't cause red plating. The way that is drawn in seems very strange to me though. It looks almost like the resistor is only between the supply and screen on one tube, but the other in the channel is directly wired. Either way, if its not open and somewhere in the general ballpark, it'll work. The value isn't that critical. Ideally you do want good resistors there and just as a general rule components should match reasonably from channel to channel. Much more critical are the grid to ground resistors on each output tube. I can't read the value, but the lower tube has one to ground, the upper has two in series with a connection back to the 12ax7. The upper pair need to be what they are for correct phase inverter performance.

These don't have adjustable bias, its a cathode bias system. It develops across the resistor connected to pin 3 of the output tube. Don't have a real clean copy of it, but it looks like a 130 ohm. Guessing you should have around 12v from pin 3 to ground.

 

[century tek]

The 270Ω should only be replaced if it needs to be replaced. Does it read in spec? Also, the red plating tube is probably the tube itself. Have you swapped tubes to see if the issue follows the tube or the socket?

 

[BassKulcha]

Try moving that tube and see if the problem goes with it. That will tell you if its a tube issue or a component underneath.

Also, the red plating tube is probably the tube itself. Have you swapped tubes to see if the issue follows the tube or the socket?

Will get on this today if I get a chance, thanks.

The 270Ω should only be replaced if it needs to be replaced. Does it read in spec?

R70 reads in-spec at 266Ω, R89 is at 251Ω (270Ω is what is called for in the schematic). I'll need to replace them, but it's good to know that they aren't the cause of the red plating.

 

[gadget73]

Those are within the typical 10% tolerance. I wouldn't worry too much about it honestly unless you're on a re-resistoring path anyway. Its a worthy goal, just a PITA. I've done it and after I was done have not regretted it, I just tend to curse my choices during the process.

 

[BassKulcha]

Ok, I just brought it up again with the variac and the DBT, and unlike the first time the bulb glows bright (not quite the brightness of a short, however—)

I took the DBT out of series and just used the variac to bring it up to 115v. All tubes seemed to be very bright, and V11 started to red plate. Swapped it out with another 6BQ5, brought it up again just with the variac and V11 again started to red plate. I don't know that if I had kept it going more output tubes would've red plated or not. As I mentioned, all the tubes seemed pretty juiced up...

I checked DC voltages on pin 7/plate of all output tubes and they were slightly high at ~347v (345v on schematic), but I figure that is in the range of normal (especially if the variac isn't spot on accurate.)

I don't know whether the next move should be to get a whole new quad of 6BQ5s and try again? The brightness of the DBT seems to point to a different issue...

Should I be measuring voltages anywhere else when I bring this up again? I'm stumped...

 

[BassKulcha]

Much more critical are the grid to ground resistors on each output tube. I can't read the value, but the lower tube has one to ground, the upper has two in series with a connection back to the 12ax7. The upper pair need to be what they are for correct phase inverter performance.

These are 120Ω 2W resistors. Pin 3 on V8 reads 104.9Ω, pin 3 on V11 reads 107.5Ω. Assuming these will need to be replaced? Could these be causing the red plating issues?

 

[gadget73]

Could be, but usually resistors go up in value, not down. If they are wirewound, they rarely shift value at all. Do you have any new resistors in the same general range to confirm if the meter is accurate?

Also if anything is across those resistors it will read low.

Try unhooking the coupling cap from V11 and see if it still red plates. Also worth making sure the grid to ground resistor is somewhere reasonably close to value.

 

[BassKulcha]

Ok, So I went ahead and just replaced some of the resistors that I knew were out of tolerance:

1. R71 on V8 (120Ω 2W / grid-to-ground resistor/6BQ5)
2. R87 on V11 (120Ω 2W / grid-to-ground resistor/6BQ5)
3. R63 on V6 (100KΩ / 12AX7)
4. R67 on V6 (100KΩ / 12AX7)
5. R80 on V9 (100KΩ / 12AX7)
6. R84 on V9 (100KΩ / 12AX7)
7. R60 on V5 (1MΩ / 12AU7)
8. R76 on V5 (1MΩ / 12AU7)

I decided to take a series of voltage measurements with and without the DBT and variac, and with and without the 12AX7 and 12AU7 tubes installed in order to get a real sense of how the addition or subtraction of these factors impact electrical functionality. So, if your eyes start glazing over that's ok, I'm mostly just including all of these measurements to organize and centralize the data for myself...

EDIT: I ordered a new matched quad of 6BQ5s today, so I'll take output tube measurements as soon as I can after they come in...

First:
With no tubes installed I brought the amp up to 115 VAC, with the variac and DBT in series to make sure there were no issues or mistakes that I may have made in replacing the resistors.

Then:
1. I installed the GZ34 rectifier tube and took voltage readings. (+ variac + DBT + GZ34 )
2. I removed DBT and took all readings again. (+ variac - DBT + GZ34 )
3. I installed 12AX7 and 12AU7 tubes, reinstalled the DBT in series, and took readings again. (+ variac + DBT + GZ34 + 12AX7 +12AU7)
4. Keeping tubes installed, I removed the DBT and took readings once more. (+ variac - DBT + GZ34 + 12AX7 +12AU7)
5. Finally, I removed both the variac and DBT and took readings with wall voltage only. ( - variac - DBT + GZ34 + 12AX7 +12AU7)

-------------------------------------------------------
Voltage Readings
-------------------------------------------------------

6BQ5: (schematic = 345 VDC)

1. V7 pin 7: 387.7 VDC / ~448 no DBT

2. V7 pin 9: 387.8 VDC / ~449 no DBT

3. V8 pin 7: 387.3 VDC / ~448 no DBT

4. V8 pin 9: 387.5 VDC / ~448, no DBT

5. V10 pin 7: 387.3 VDC / ~449, no DBT

6. V10 pin 9: 387.4 VDC / ~449, no DBT

7. V11 pin 7: 387.6 VDC / ~449, no DBT

8. V11 pin 9: 387.5 VDC / ~449, no DBT

-------------------------------------------------------

12AX7: (schematic = 145 VDC)
1. V6 pin 1: 278.6 VDC / 323.5, no DBT / 155.7 tubes installed, w DBT / 189.6 tubes installed, no DBT / 175.9 wall voltage only

2. V6 pin 6: 278.6 VDC / 323.8, no DBT / 150.3 tubes installed, w DBT / 181.6 tubes installed, no DBT / 168.4 wall voltage only

3. V9 pin 1: 278.5 VDC / 323.8, no DBT / 154.5 tubes installed, w DBT / 187.9 tubes installed, no DBT / 175.5 wall voltage only

4. V9 pin 6: 278.4 VDC / 323.8, no DBT / 148.1 tubes installed, w DBT / 180.7 tubes installed, no DBT / 169.1 wall voltage only

-------------------------------------------------------

12AU7: (schematic = 70 VDC)
1. V5 pin 1: 278.7 VDC / 321.5, no DBT / 77.9 tubes installed, w DBT / 94.8 tubes installed, no DBT / 88.3 tubes installed, wall voltage only

2. V5 pin 6: 278.2 VDC / 322.1, no DBT / 79.9 tubes installed, w DBT / 96.6 tubes installed, no DBT / 90.0 tubes installed, wall voltage only

-------------------------------------------------------

GZ34: (schematic = 310 VAC)
1. V12 pin 4: 295.6 VAC / 272.2, tubes installed, w DBT / 334.5 tubes installed, no DBT/ 306.2 tubes installed, wall voltage only

2. V12 pin 6: 295.6 VAC / 271.9, tubes installed + DBT / 334.2 tubes installed, no DBT / 306.4 tubes installed, wall voltage only

 

[BassKulcha]

Try moving that tube and see if the problem goes with it. That will tell you if its a tube issue or a component underneath.

Also, the red plating tube is probably the tube itself. Have you swapped tubes to see if the issue follows the tube or the socket?

Matched quad came in today, so I just installed them. DBT glowed pretty bright when brought up to 115 V and got slightly brighter once the rectifier kicked in. Not quite the brightness of a dead short, but clearly something is up.

Don't know if this makes a difference, but I tried placing one tube in at a time into each socket to check if the issue was the socket, but got the same results for each. Then tried adding one tube at a time until all populated and got the same result.

Didn't get to the point to see if tubes red plated (would a DBT in series even allow for that?), but I at least know that the 6BQ5s are good. I also didn't check voltages as I was uncomfortable with keeping it on with the DBT glowing bright (EDIT: I just checked voltage on the output tubes with the DBT in series and they are all somewhere between 111 to 114 volts. Clearly the DBT is doing its job...)

Last comment—I just noticed that the fuse is a 4A 250 volt, while the SA-232 calls for a 2A slo-blo...am I playing with fire here?
Any ideas on how to troubleshoot? Thanks in advance for any help you all can give!

 

[gadget73]

What size bulb are you using? If its got a fair bit of glow even with no rectifier it may not be enough. Probably want 100w or so for this.

With the plate supply at half voltage its probably not ever going to conduct enough to cause a problem. You need to power it from the wall in order to really test this.


Yes, put the 2A fuse in.

 

[BassKulcha]

Ok, thanks gadget73. I don't have a 2A s/b on hand, so I'll need to pick that up (hopefully using the 4A won't cause problems in the meantime...)

What size bulb are you using?

I currently have a 72W/120V bulb in the DBT—

If its got a fair bit of glow even with no rectifier it may not be enough.

On a scale from 0 – 5 with no glow being 0, and a dead short being 5:

• With rectifier in, it's a 1.0. With both the rectifier and the 12AX/AU7s in it's a 1.0 to 1.5. With output tubes in its a 3.5 to 4.0. Output tube filament glow is minimal, tubes warmed up slowly

Probably want 100w or so for this.

On hand, I only have a 200W incandescent, so I tried that (not sure if the higher wattage is an issue...)

• With the rectifier installed, and with both the rectifier and the 12AX/AU7s installed, it was a 0. Adding output tubes brought it up to a 1.5 – 2.0. Output tube filament glow was robust, tubes got hot quickly.

With the plate supply at half voltage its probably not ever going to conduct enough to cause a problem.

I'm trying to understand—does this mean that because the DBT is limiting current, it's only supplying half the voltage to the plates?

You need to power it from the wall in order to really test this.

I was thinking that perhaps there might be an issue taking the DBT out of series because of the brightness—if not I'll definitely give this a shot.

Thanks for all your help with this—it's been challenging trying to figure all of this out, but also satisfying to know that this SA-232 is *so close* to functioning again. I've learned a ton with this project, and I'm excited to further this knowledge moving forward...

 

[tcdriver]

...I currently have a 72W/120V bulb in the DBT—...

It that a typo? Are you using a 75W incandescent bulb? Only incandescent bulbs should be used with a DBT.