Incorporating Burst Power and Slew Rate Tests for Amplifier Reviews

[Cbdb2]

I understand that the time domain and frequency domain are mathematically linked, but not entirely interchangeable.

Than you dont understand. Some one here measured there speakers impulse response and freq. response than used software to convert one to the other, they where the same.
Much of your DSP would not work if this wasnt true.

Edit: Freq. response must include the phase, which most FFTs dont do

 

[Cbdb2]

Consider an amplifier with a slew rate of 0.5 V/μs and an input signal at 10 kHz with a peak amplitude of 10 V. The required slew rate for this signal can be calculated based on the frequency and the voltage swing. In this case, the required slew rate is approximately 0.63 V/μs. Since the amplifier’s slew rate is only 0.5 V/μs, it cannot reproduce the signal accurately, even though 10 kHz is well within the audible range and the amplifier’s bandwidth.

Which will be extremely obvious from a standard distortion test.

 

[AdamFrandsen]

Than you dont understand. Some one here measured there speakers impulse response and freq. response than used software to convert one to the other, they where the same.
Much of your DSP would not work if this wasnt true.

Even if they are identical under all operating conditions, it still would make for a much clearer and more digestible result if done separately… If there are no tests that can show TIM, recovery time, dynamic headroom, power supply stability or thermal/current limitations better than what is currently on display, so be it… these were just my suggestions. I also still think 124 multi-tone test would be preferable.

 

[Cbdb2]

Found a definition for laymen.
"In an op amp, the slew rate is the maximum speed at which the output voltage can change; if you instantaneously change the input, the output voltage will "slide violently or uncontrollably" to the new stable point, and the speed at which that happens is called the slew rate.

 

[SSS]

Found a definition for laymen.
"In an op amp, the slew rate is the maximum speed at which the output voltage can change; if you instantaneously change the input, the output voltage will "slide violently or uncontrollably" to the new stable point, and the speed at which that happens is called the slew rate.

Good point. When output is slower than the input signal then negative feedback does not work.

 

[Cbdb2]

Good point. When output is slower than the input signal then negative feedback does not work.

Exactly, so the amp just ramps as fast as it can till we get feedback and control again.

 

[Doodski]

This is good information. I had no idea that slew rate can be indicative of so many things and cause issue with feedback. I'm getting a much better handle on it than before when I equated it with a FFT graph and max frequency response slope.

 

[Cbdb2]

Yes this seems to be hard to grasp for many non-technical. When slew limiting the devices looks more like a digital logic gate (switch) than an analog amp. It just tries to reach the higher voltage as fast as possible.

 

[Spkrdctr]

Great engineering discussion! I have enjoyed reading this thread. It seems at the end of the day though, this thread is just a hop and a jump to "rolling OP amps". Nothing like getting a brand new $2000 amp, taking the cover off and unsoldering that cheap opamp from the factory and soldering a new one in that your buddy at work gave you. Specs? Who needs specs, just a good listen will tell you the new second hand op amp is better. LOL

Seriously a good discussion. Even if I am trying to throw some humor in here.....
Ok, I am going off to hide.....

 

[kemmler3D]

Consider an amplifier with a slew rate of 0.5 V/μs and an input signal at 10 kHz with a peak amplitude of 10 V. The required slew rate for this signal can be calculated based on the frequency and the voltage swing. In this case, the required slew rate is approximately 0.63 V/μs. Since the amplifier’s slew rate is only 0.5 V/μs, it cannot reproduce the signal accurately, even though 10 kHz is well within the audible range and the amplifier’s bandwidth. This shows why actual slew rate testing is necessary to ensure proper transient handling within the audible spectrum.

Plausible that an amp could fail under some dynamic condition or another. Do you have any examples of amps failing to do this?

Slew rate tests above and beyond the normal tests are not performed because they aren't considered informative, not because it hasn't occurred to Amir that amps might fail under various conditions.

The tests he performs pretty much cover the bases, I think the task is to understand what information they are all delivering... or provide a known example where the standard tests would have missed something important.

 

[SSS]

Great engineering discussion! I have enjoyed reading this thread. It seems at the end of the day though, this thread is just a hop and a jump to "rolling OP amps". Nothing like getting a brand new $2000 amp, taking the cover off and unsoldering that cheap opamp from the factory and soldering a new one in that your buddy at work gave you. Specs? Who needs specs, just a good listen will tell you the new second hand op amp is better. LOL

Seriously a good discussion. Even if I am trying to throw some humor in here.....
Ok, I am going off to hide.....

Thanks. That's because some engineers are here in the forum. But I like also black humor, especially when an amp burns down.

 

[SSS]

Plausible that an amp could fail under some dynamic condition or another. Do you have any examples of amps failing to do this?

Slew rate tests above and beyond the normal tests are not performed because they aren't considered informative, not because it hasn't occurred to Amir that amps might fail under various conditions.

The tests he performs pretty much cover the bases, I think the task is to understand what information they are all delivering... or provide a known example where the standard tests would have missed something important.

Yes, amir does an excellent job. I appreciate it much. And no testing does cover all aspects. Further, often non-technical persons cannot interprete findings which development engineers measured. That is why snake oil products exist and are purchased.

 

[777]

Slew rate can't be measured on an amplier that was assembled.

 

[777]

You don't want to measure burst power . You want to see the damped oscillations with the burst signal. You want to see stability of the amplifier.

 

[Titurel]

While I am no Galileo, I catch your drift

Better to not be a Galileo than end up a Giordano Bruno.

 

[SIY]

I also still think 124 multi-tone test would be preferable.

Great, show us one and explain what additional information it provides.

All this was a criteria for tube power amps which tended to such effects when not proberly designed.

Interestingly, tube amps almost never showed slewing. They did show bandwidth limitations and transient ringing when not compensated properly.

 

[SSS]

Great, show us one and explain what additional information it provides.

Interestingly, tube amps almost never showed slewing. They did show bandwidth limitations and transient ringing when not compensated properly.

You are right since tubes by itselfs are very very fast. It is the circuit around which can make them slow. I meant primarily the ringing and bass droop and possible rf oscillilation.

 

[AdamFrandsen]

Plausible that an amp could fail under some dynamic condition or another. Do you have any examples of amps failing to do this?

Slew rate tests above and beyond the normal tests are not performed because they aren't considered informative, not because it hasn't occurred to Amir that amps might fail under various conditions.

The tests he performs pretty much cover the bases, I think the task is to understand what information they are all delivering... or provide a known example where the standard tests would have missed something important.

I was not accusing anyone of negligence, nor was I trying to force anyone to conduct tests they object to. All I did was make a suggestion.

 

[AdamFrandsen]

Great, show us one and explain what additional information it provides.

Interestingly, tube amps almost never showed slewing. They did show bandwidth limitations and transient ringing when not compensated properly.

Unfortunately I do not have the tools nor the time… I don’t see how this would put a strain on anyone, or even add much to the workload of testing the amps. Just makes sense to me…

 

[SIY]

Unfortunately I do not have the tools nor the time… I don’t see how this would put a strain on anyone, or even add much to the workload of testing the amps. Just makes sense to me…

The software to do this is free, and the only equipment needed is a basic audio interface. So get to work. Use 124 frequencies, or 110, or 42, or whatever gematria appeals to you, show everyone what we're missing.