Didn't see the EDIT.. . .Rethink . . . .
If T4's base is the same voltage as its emitter. It should not be on.
That's the first thing you need to sort out. Disregard what I said previously until you've dealt with that,
That 28.6V on T2 base is suspiciously 0.7V below its collector. Exactly what it would be if that transistor were PNP.
T1's base is a Vbe + a schottky below V+. Consistent with the above.
If T3 was NPN, then current would flow through R1 and base-collector junction of T3, giving the voltages shown.
Again if T5 were PNP and T6 NPN then they would be non-conducting, and the output driven to 28.6V via the feedback resistor.

 

What's the value of Rbias?

Could you update properly your schematic?

 

Wow, thank you guys for all the great help and info here. I'll try to address everything.

If T4's base is the same voltage as its emitter. It should not be on.
That's the first thing you need to sort out. Disregard what I said previously until you've dealt with that,

Well remember, T4 now has the 2k2 resistor from base to emitter as suggested in #10. I would assume that if T4 is off, then the base voltage would be 29.35V.

I've checked and triple checked that the parts are in the right place and oriented correctly, so I'm more inclined to believe what is happening is what you suggested in post #19. And trust me, I have been very, very tempted to remove those protection diodes cause D2 is a clear path for the positive voltages to appear on the input.

What connects to the output of this amplifier? And where does the other end of whatever connects to the output connect to?
Please draw the circuit that gives you NEG.VDC.IN

I'll draw this up for you guys. But in the meantime, in an effort to attempt to isolate the problem I first removed the input voltage feed so the input is ~20k to ground (2 trimmers and limiting resistor). My reasoning here was if the amplifier is working properly then with a grounded input we would get a grounded output...no change. Then just now, I removed the reverse biased diode on the output leaving only the feedback path...again no change. So as of now the circuit is completely isolated and behaving this way on it's own.

I find it very odd that both of these circuits on the board are behaving the exact same...that seemingly rules out a mistake or failed component. Unless, is it possible that the high current drawn in my first post with this circuit damaged the same part(s) in the same way.

 

What's the value of Rbias?

Could you update properly your schematic?

The value of Rbias is currently a jumper. When I first powered the device on I was smelling burning and tracked it down to this circuit drawing way too much current using 2 diodes as Rb. Ian0 suggested shorting it to make the output stage class B, which was fine because I wasn't ready to test this circuit at the time. I'll update the schematic this evening.

 

Where is this current coming from? T1 or T2?

Looking at the base voltages on T1 and T2:
T1 Vbe 28.46-28.6 -> -0.14 -> T1 is off
T2 Vbe 28.6-28.6 -> 0 -> T2? strange, base-emitter short?

The 0.27 volts across R1 is driving T3 & T4 but 0.27 isn't enough to
turn on even a single Si BE junction. So there's more mystery there.

Yes, the current has to be mostly through T2. I thought for sure you found the mystery with a b-e short on T2...I didn't think I checked it for shorts. But I was measuring like 150k or something high when I checked it just now.

You ain't kidding about mystery...I have a talent for finding the most obscure failures and problems in extremely simple circuitry.

 

What happens if you break the connection between T1 and R1 (leaving R1 connected to T3)?

 

What happens if you break the connection between T1 and R1 (leaving R1 connected to T3)?

Just tried and no change. I find that curious because shouldn't T1 be driving the bus, so to speak?

 

Exactly. You definitely have a fault with T3/T4/T5. Remove t5 and T6 and connect the collector of T4 to the output.

 

Exactly. You definitely have a fault with T3/T4/T5. Remove t5 and T6 and connect the collector of T4 to the output.

Boom! -76V output!! I went and looked at T5 and it's b-e diode didn't register on the DMM. But something else was very, very strange...T5's b-c diode registered but with the positive lead on the collector and negative lead to base. That's opposite what it should be for a NPN, no? Part is marked 2N6517.

Ok, so now the question is why? Did T5 blow because of the excessive current drawn from my first problem with this circuit? Or was that reverse junction diode the cause of the excessive current? Maybe On Semi screwed up and sent a batch of 2N6520s out as 6517s (which fits your earlier analysis)?

 

A quick simulation with a 2N6520 as T5 with 2 bias diodes added back (in place of RBias) shows 35mA (in other words "way too much) in the output stage and a Vout of 30V as we observed. Shorting the diodes drops the current significantly but also makes Vout 30V...again exactly as we observed. Hmmmmmmm.....

 

So, how about that drawing of what’s connected to the input and output?

 

My apologies I got sidetracked on this because I pulled out the bag of 2N6517s and it turns out they are are all mislabelled. According to the datasheet, the 2N6517 is E-B-C looking at the flat, whatever I have is E-C-B or C-E-B!

 

That is what happens when you buy cheap fakes from ebay, AliExpress or Amazon.

 

That is what happens when you buy cheap fakes from ebay, AliExpress or Amazon.

Does Arrow count?

 

Does Arrow count?

I do not know if they know how to count or if they sell cheap Chinese fake parts.

 

I do not know if they know how to count or if they sell cheap Chinese fake parts.

I actually contacted both Arrow and On Semi to see what they had to say. On said that there is a C version of 6517 that is E-C-B (if I recall) but I can't find any mention of it in the datasheet or on their site. I did buy the C version but On said the markings on mine don't have the 'C' at the end like they should. In any event Arrow was nice enough to refund them and On was offering to replace them for me too. So, that at least turned out OK.

 

https://www.onsemi.com/products/dis...ow-vcesat-transistors/2n6517c#tab=tabOverview
but the datasheet still says that base is in the middle!

 

The C stands for "Chinese":

 

Ok new non-C version parts are on the way which means I'll have a few days before I can do anything. In the meantime, a big "thank you" is in order for everyone who helped, especially Ian0. I really learned a lot from your suggestions, approach and calculations. That gave me a way to sort of break this problem down into smaller sections making it a much more manageable; especially since LTSpice wasn't giving result anywhere close to the real-life observations (and now we know why). I feel like I somewhat understand what is going on now; like how the Darlington may not be the best idea because R1 is going to have to drop enough voltage across it to overcome 2 Vbe junctions instead of the normal 1. So thank you so much!

 

Ok new non-C version parts are on the way which means I'll have a few days before I can do anything. In the meantime, a big "thank you" is in order for everyone who helped, especially Ian0. I really learned a lot from your suggestions, approach and calculations. That gave me a way to sort of break this problem down into smaller sections making it a much more manageable; especially since LTSpice wasn't giving result anywhere close to the real-life observations (and now we know why). I feel like I somewhat understand what is going on now; like how the Darlington may not be the best idea because R1 is going to have to drop enough voltage across it to overcome 2 Vbe junctions instead of the normal 1. So thank you so much!

You'd be better with a current mirror instead of R1 (then you don't have to think about what value to make it).
You might be better with a MOSFET instead of the Darlington, but a Darlington may be better than a single transistor, as you have much more Gm.
If the load is single ended (i.e. doesn't have to go both negative and positive) then the MOSFET could drive the load, and you don't need T5 and T6 and even to think about bias.