One more Eico ST70 rebuild thread

As far as balance
Take a look at bias PS Lower right on schematic.
They share a single supply spit into 2 identical halves. If you see the output, each half feeds each channel's output section (also identical values).
If you are getting same thing (voltage) out of each channel's bias pots (you could check this with output tubes pulled) then your problem is by the output tubes themselves (balance pots, passive parts, resistors (possibly failed, broken part or cold solder joint).
Before you dig into this, i think we asked before. Did you try swapping output tubes from one channel to the other, to see if problem moves with tubes?
Even if a tube tests well (or is new) it can still do weird things in circuit. Conceivably your balance ,bias ,sound issue. Also if they aren't conducting (your output tubes) possibly account at least partially towards your high supply readings.
I know i dumped a bunch of stuff on you. Also probably not nearly as cohesively and clearly as i could have.
Read it a few times. Ask if something isn't clear.
If anybody wants to correct me or clarify, feel free.
 
+1 on the advice in the last two posts by nj phoenix. It's particularly important to check simple things like meter operation and battery. (Great progress old; this is after all your 'sophomore' project. I managed to miswire the filament wiring with my first rebuild so badly that the insulation melted off.)
 
Lot of good info.
Voltages were taken with tubes in and at 117v from variac.
Meter tested on AC voltage (wall outlet always 125), DC on an adjustable DC source as well as compared to my analog vom. So I have to conclude that it is fine.
Swapped out tubes, same exact results.
The 12ax7 flashes on 117v but will disregard for now.

In addition to swapping caps and resistors, I changed the line stage per Lafferty's article (which required changing the 7247 to a 12ax7 but dont know if this is the place to look. Will check the circuit changes closely today as well as the bias PS.

Just dont know where to start checking voltages to trace down the problem. If there was a voltage chart or diagram with voltages, it would be helpful. Wouldnt votages at the pins along the way be the route to go?
 
It sounds to me like one (or more) of the power tubes is not conducting, or is biased way too cold. This would lead to higher than expected plate voltages, too.
With the amp off check to see if all four cathodes, pin 5, of the 7591s are 10 ohms away from ground. If not fix before proceeding.

With the amp on, check for negative voltage on the grids, pin 6 of each power tubes. Note that these voltages are made variable by both the overall bias level pot (one pot for each pair of tubes) and one balance pot (also one pot for each pair of power tubes). It might seem straightforward but it is also easy to lose track of which is doing what. I use a pencil to mark the chassis, reminding me which pot is doing what.

Once that you have verified the range of negative voltage on the grids, you can use the test points, one for each power tube, and the bias method described in the manual. Dialing these in, when wired correctly, should get your power tubes to conduct as expected and tame your high voltages
 
With the amp off check to see if all four cathodes, pin 5, of the 7591s are 10 ohms away from ground.
10, 10, 10, 11.4

With the amp on, check for negative voltage on the grids, pin 6 of each power tubes.

-16v, -15v, -30v, -22v
This is with the current pot adjustments. I tried to balance earlier

"CH1 will not balance (4.5v) and will not bias (4.5v vs spec of .38v)
CH2 will balance but can only get bias to 15v."

I adjusted the pots all the way and the above is the best that I could get. My Sencore TC136 says tubes are good. Will double check once they cool.


 
Last edited:
Getting that much voltage across the 10 ohm resistors should have the tubes glowing red, the resistors smoking, and the voltages to be way low. 4.5 volts across 10 ohm is 450 milliamps, a 7591 probably is supposed to bias at 35 ma or so. I don't honestly know if one is even capable of conducting near a half amp when set to full melt-down mode. The bias voltages seem approximately right on one side and overly low (too negative) on the other. Somewhere in the -15 to -20 volt range is about where I'd expect.

Something just does not add up here. Are you measuring directly across the 10 ohm resistors, or is this via some sort of test point that could possibly have a sketchy switch or connection involved?
 
At the risk of repeating myself. Your tube tester doesn't tell the whole story.
If you swap the 2 output tubes from one channel to the other and the problem stayed with the same channel we can rule out the tubes ( you did say that right?)
 
For clarity sake. What Gadget meant about red plating. Under normal conditions the filament should glow but nothing else in the tube should. If that tube is truly biased at 4.5 v , its flowing about 1100% of what its designed to and will self destruct (possibly spectacularly). Most of the internals on that tube would be glowing quite brightly. I tend to agree with Gadget something doesn't jibe. Are the pin sockets of the tube sockets clean and tight?
 
Last edited:
Thinking aloud again (that's why you don't hear anything :D).
I think if everything is the right component in the right place ( not saying that's the case) you have a faulty connection. (Tube not making contact to pin, cap or resistor with hairline crack which appears intact, cold solder joint....?:idea:
 
Tubes are supposed to be hot.
The entire internal structure of the tube shouldn't glow orange (red?)
If i wasn't on my phone I'd pull you up an image. Google red plating. You likely can find what I'm describing
 
In addition to what NJ says above, I would do the next checks with the power tubes removed. You need to test your (negative) grid voltage on all the pin 6s and should be able to balance them out and have a usable range. This means both pairs should have approximately the same amount of negative voltage available, and should be able to be balanced. EVEN WITH TUBES REMOVED. If not, something in the wiring to that pair needs to be corrected. Follow the negative voltages from the diode towards the grids, and see how one pair of tubes compares to the other.

Your findings are consistent with your earlier description of the symptoms, and believe it or not, I think you are closer to the solution than you might feel.

Be careful even when the power tubes are out because the voltages are still lethal. I use Pomona mini-grabbers to clip onto socket pins, to test live voltages.
 
Humor me.
I'm winging this on my phone on a commuter train
Give me voltages of R98 and R99 . Both sides of each.
Center the bias and balance pots and give me voltages at R66,68,69,70 both sides
 
In addition to what NJ says above, I would do the next checks with the power tubes removed. You need to test your (negative) grid voltage on all the pin 6s and should be able to balance them out and have a usable range. This means both pairs should have approximately the same amount of negative voltage available, and should be able to be balanced. EVEN WITH TUBES REMOVED. If not, something in the wiring to that pair needs to be corrected. Follow the negative voltages from the diode towards the grids, and see how one pair of tubes compares to the other.

Your findings are consistent with your earlier description of the symptoms, and believe it or not, I think you are closer to the solution than you might feel.

Be careful even when the power tubes are out because the voltages are still lethal. I use Pomona mini-grabbers to clip onto socket pins, to test live voltages.



Wow
First time.
Testing with amp on side. Dont know what I touched but sent me across the room. Typing difficult right now but was able to get readings.
Broke my vom when the amp landed down hard.

Ch1 -14v, -19v
Ch2 -16v, -16v
Arms and hands still numb. Think its when I went to unclip the vom.
Wish I could get the voltage down so it wont be as bad next time.
 
NJ
Sent PM, cant let the ex know where Im hiding.
Ears ringing and face stil tingling. One hand on the chassis, the other on something that bites. Chest a little tight. May need to take a break. Hope the data yields something.
Next time i'll measure thru the socket.
 
Most of us have been hit before (sorry, i hope you're alright)
Sorry about the meter too. Some install bleed down resistors. B+ to ground 470k or so. Drains down caps after shutdown.
What do those #s represent?
 
I sometimes test upside down (the amp not me. Rule of thumb. Probe with one hand in back pocket. No metal jewelry
 
Back
Top Bottom