What Am I Seeing?

[zendragon63]

Greeting gents and fellow tube building amp enthusiasts. Not building much lately but rather trying to learn the finer points of tone shaping and such as time allows.

While I see stuff on the scope that I can identify, I come across a few that I don't know and should. Attached is a scope pic of pretty severely overdriven pair of 6AQ5s measured at the speaker out. There is a spike on the leading edge (circled). I see this in some amps and not in others at this drive level. Anyone have an idea?

Many thanks in advance. Regards

dennis

[sluckey]

That is a high frequency overshoot. A square wave is actually made up of an infinite number of different frequency sine waves.

It may not be caused by your signal at all. If you are using a quality scope probe, connect it to the calibration output test pin on your scope and adjust the probe for minimum overshoot while viewing the cal out signal.

[eleventeen]

If you are measuring at the speaker output >>with the amp driving a speaker<< it's entirely possible that you are seeing an interaction between the inductance of the speaker as it "gives back" from an overdriven excursion. In other words, whatever it is you are seeing is a combination of (a) what the amp is producing and (b) what the speaker is "reflecting" (definitely not the right word) The speaker is a reactive component; as is the secondary of the output transformer. As the speaker reaches a full excursion one way and starts to return, it starts to give energy back to the output circuit. Your scope probe sees that energy. Since that is mashed together with your amp output, you are seeing multiple things.


If that is how you are configured, what you should do (as part of your learning/observations) is to replace the speaker with a resistor. That way, no weirdnesses come back from the speaker and you would then be looking at the pure output of your amp without your amp driving (and therefore also receiving energy back from) a reactive load.


Another reason why a resistive "dummy load" is/can be useful.

[zendragon63]

Thank you gents. First thing I will check probe calibration to see we can dial it out.

I run one of those Weber mass 50 attenuators with a speaker so it is not a pure resistive load--I might be picking up some of that 'extra' energy as it gotta go some where.

Your thoughts on this one are appreciated. Regards

dennis

[shooter]

On the systems I worked on we pulsed high I hi V DC (basically square waves) , in order to decrease slew rate (time from 0 to max DC) the power supply had to generate more than "max" to keep the slew rate low, the problem, overshoot, your spike.  The PS load was a coil also, so we had to cal out the spike without sacrificing slew rate, with 3 supplies it could take 12hrs!  Coils don't like change

[Glennjeff]

That could also be borderline HF instability. IMHO it is very wrong.

[PRR]

I would not be *too* concerned about the pip. Though it is odd to find it on the secondary.

What might bother me more is the gross asymmetry on "a pair of 6AQ5s". The ideal push-pull(?) amplifier makes a nice symmetrical over-drive and maximum loudness.

If I assume the mid-glitch is the crossover, then you have 1 unit down and 3 units up, which is pretty rad and kinda misses the goal of push-pull.

Yes, in extreme overdrive many of the simpler drivers make asymmetric drive signals, and this is part of the charm of some amps.