What is the current status of your amplifier schematic? Would you be willing to post it - albeit with some additional re-work required?

No aspersions meant concerning your wiring ability but have you thoroughly checked the circuit layout? Maybe it's something overlooked.

 

This is interesting, but I do have one question. If I did connect the emitters and have the load to ground, how would this reduce the reverse voltage on the transistor bases? If the two emitters are at the same voltage, does that reduce the reverse voltage at the base?

I have no clue why the diodes aren't working for you. They should. To answer the above question look at these plots, with special attention to VM1 & VM2.

For the record VM2 wouldn't look like this because the EB junction would break down. The reason this plot doesn't reflect that is because my version of Tina Spice doesn't provide for "Smoke Analysis". That option (Tina Pro) costs more than I care to spend. On the up side the plots clearly show the excessive Vebo on the upper plot, while the bottom plot eliminates it.

 

What is the current status of your amplifier schematic? Would you be willing to post it - albeit with some additional re-work required?

No aspersions meant concerning your wiring ability but have you thoroughly checked the circuit layout? Maybe it's something overlooked.

Thank you for your interest.

At this point all aspersions are welcome. If I merely wired it wrong, I really would be happy, because then the world would make sense again...and I could wire it correctly and all would be well.

As of this post, I reviewed the physical set up of the circuit to make sure that it compared to the attached circuit diagram.

The tests were made with a signal from 2V peak to 20V peak. The frequency from 60kHz-900kHz. 4 different sets of transistors were used. In every single case the transistors failed to shut off as shown by the oscilloscope inserts.

 

Have you tested the BE junction of the transistors that failed? Your DMM has a Diode function that is suitable to test it with.

 

I did some editing on an earlier post that I think you should read. It may be applicable to the problem or it may not.
http://forum.allaboutcircuits.com/showpost.php?p=480059&postcount=12

 

In every single case the transistors failed to shut off as shown by the oscilloscope inserts.

The oscilloscope "inserts" lack any discernible detail. What are the peak values for instance? Is the shape a true sinusoid or something more distorted?

 

I would suggest a missing ground connection at the junction of R5 & R6 would give you a straight sinusoid at the Q2 & Q4 emitters. One can simulate this and see just that result with a 1MΩ probe at either emitter. I used TIP48 and MJE250 in my simulation. The V1 / V2 common point was the only ground applied in the simulation.

 

The oscilloscope "inserts" lack any discernible detail. What are the peak values for instance? Is the shape a true sinusoid or something more distorted?

It is a large image, but if you enlarge it, it should clear up.
The waves are perfect sine waves. The peak values tried were everything from 2V to 20V.

 

I can't see any oscilloscope plots in the attachment.....???

Did you pick up on my comment in post #27?

 

thermal runaway???

 

Have you tested the BE junction of the transistors that failed? Your DMM has a Diode function that is suitable to test it with.

My DMM doesn't have it, but I tested the resistance in all combinations on the transistor and it looked good. There was complete resistance in the proper directions.

As I was doing this, I was thinking that the transistor is blocking current when I test it. Why can't it do it in the circuit? It is absolutely crazy.

 

My DMM doesn't have it, but I tested the resistance in all combinations on the transistor and it looked good. There was complete resistance in the proper directions.

As I was doing this, I was thinking that the transistor is blocking current when I test it. Why can't it do it in the circuit? It is absolutely crazy.

Is the ground between the two emitter resistors really connected?

 

Is the ground between the two emitter resistors really connected?

The longer this thread runs it would seem the evidence is pointing in this direction.

 

Is the ground between the two emitter resistors really connected?

This is interesting. To check this possibility, I disconnected the ground wire from the bread board and nothing changed. So this seemed a distinct possibility. However, checking the power supply connections, it should be OK.

I have a mastech 1-50V dual power supply. The serial mode is selected; the negative of one is connected to the positive of the other. My ground wire is connected at that point (as seen in the schematic). There is no earth ground (even though spice requires it). The mastech manual specifically states not to use the grounds when running in serial mode.

I also checked the entry points to the bread board to make sure all the voltages were present.

 

This is interesting. To check this possibility,(1) I disconnected the ground wire from the bread board and nothing changed. So this seemed a distinct possibility. However, checking the power supply connections, it should be OK.

I have a mastech 1-50V dual power supply. The serial mode is selected; the negative of one is connected to the positive of the other. My ground wire is connected at that point (as seen in the schematic).(2) There is no earth ground (even though spice requires it). The mastech manual specifically states not to use the grounds when running in serial mode.

I also checked the entry points to the bread board to make sure all the voltages were present.

(1) What ground point are you referring too, the power supply common Gnd or R5, R6 ground? Which one did you lift?

(2) Since you referenced Spice, are you using a Spice based USB Scope? I'm asking this because the input circuitry to USB Scopes share some common attributes to a real scope. You are using the probe ground lead, aren't you?

Why haven't you posted your scope plots? I think you were asked twice.

 

(1) What ground point are you referring too, the power supply common Gnd or R5, R6 ground? Which one did you lift?

R5,R6 ground

(2) Since you referenced Spice, are you using a Spice based USB Scope? I'm asking this because the input circuitry to USB Scopes share some common attributes to a real scope. You are using the probe ground lead, aren't you?

I am using a regular analog oscilloscope. The probe ground is connected at R5,R6.

Why haven't you posted your scope plots? I think you were asked twice.

Scope plots were posted in the initial post for 3 points...and again when someone asked.

 

... For a 1.8A load, your oscillator will not run due to loading. You will need a high-power buffer between the oscillator and the emitter followers....

This is a good point Ron,

I thought a buffer was provided by the emitter follower. If this circuit had worked, I was going to cascade the emitter followers, each one allowing more current while not causing the voltage to drop (too much) at the oscillator.

However, this does not seem to work. No matter how many emitter followers I use, when the final one's resistance gets too low, the oscillator fails.

Do you know a good method to isolate the oscillator? Is it possible?

 

Scope plots were posted in the initial post for 3 points...and again when someone asked.

I was the one who asked.

I guess you are referring to the small sinusoidal like sketches as annotations on the schematic...???

If so, those annotations really aren't technically useful scope plots by any stretch of the imagination.

One would expect time scales, actual voltage values including DC offset and so forth. The relative phase relationship between the input signal and the emitter voltages would also be of interest. Presumably you have a dual channel scope to accomplish this.

Again referring to the "earthing" connection at the junction of R5 & R6 - if you do a low resistance range continuity test (with Mastech DC power supply off and CRO leads removed) from the R5/R6 junction point back to the common point of the power supply do you see a "short" - as one would expect were there a physical connection from the R5/R6 junction to the power supply common?

 

I was the one who asked.

I guess you are referring to the small sinusoidal like sketches as annotations on the schematic...???

If so, those annotations really aren't technically useful scope plots by any stretch of the imagination.

They are sufficient to explain the problem.

One would expect time scales, actual voltage values

Which time scale? The one at 60kHz or the one at 900Khz or one between.

Which voltage? The 2V peak or the 20V peak.

All these were given by words. A perfect sine wave at these values is completely defined. As described there were no spikes or oddities with the trace...other then the whole portions that should have been absent...which is the problem.

including DC offset and so forth.

The DC offset can be clearly seen as 0 by the "sketches" and also was described precisely by words and also would be expected by the schematic.

The relative phase relationship between the input signal and the emitter voltages would also be of interest.

At last.. something that I actually did not provide. The phase shift between the base signal (A) and the emitters (B,C) was zero.

Presumably you have a dual channel scope to accomplish this.

I have.

I suppose I could have taken a photo of the oscilloscope traces, but the question is, given what is already known....do you think photos would yield something else?

This is not some subtle problem involving a 1% variance in a signal. There is a whole half wave existing where it shouldn't.

Again referring to the "earthing" connection at the junction of R5 & R6 - if you do a low resistance range continuity test (with Mastech DC power supply off and CRO leads removed) from the R5/R6 junction point back to the common point of the power supply do you see a "short" - as one would expect were there a physical connection from the R5/R6 junction to the power supply common?

I hadn't, but it is a good suggestion, so I did the test and the connection is good.

 

You think you eliminated capacitance as the problem, but you have underestimated the base-emitter capacitance of these transistors. They are not spec'ed, but the spice models have Cje on the order of 3nF for both of them. This makes the capacitive reactance on the order of 1kΩ at 60kHz, which is 10% of the value of your emitter resistors. Try testing at 100Hz (you may have to use a larger input capacitor, or just direct-couple the input signal). At low voltages, you will see the half wave rectification you are looking for. At higher voltages, look closely, and you will see what looks like crossover distortion, due to base-emitter breakdown. I ran some sims and saw exactly what you did at 60kHz, and saw what I am predicting you will see at 100Hz (I added Veb breakdown diodes in parallel with the BE junctions, because the model does not include Veb breakdown).