Testing Amplifiers - Speaker or dummy load

[Rob11966]

Hi Folks,

Just a quick technical question - when looking at an amplifier output signal on a 'scope, does matter if the amplifier is connected to a speaker or dummy load? Specifically, will the trace be identical for the same test signal sine wave?

Cheers,

ob

[nigelwright7557]

A speaker is inductive so will present a very different load.
Some speakers can be quite high impedance higher up the frequency range.

When I test my new amps I power it up with the output connected to a scope and with no speaker connected. I check for a DC offset and/or oscilation.
I then set the dc offset on a preset.
I the apply a sine wave with a speaker connected and turn up the bias slowly until crossover distortion goes.

[Loren42]

Originally Posted by Rob11966

Hi Folks,

Just a quick technical question - when looking at an amplifier output signal on a 'scope, does matter if the amplifier is connected to a speaker or dummy load? Specifically, will the trace be identical for the same test signal sine wave?

Cheers,

ob

No, it will not. That assumes you are comparing a non-inductive resistor to a typical loudspeaker.

The reason for that is, loudspeakers are inductive and the real impedance seen by the amp fluctuates with frequency due to changes in the inductance of the driver with frequency.

A non-inductive load is the best way to test an amp because the impedance is constant across the audio spectrum and provides a good benchmark to measure an amp that is consistent from amp to amp.

The other advantage is if the amp goes into oscillation or in the case of a transistor amp shorts to a rail, it may fry the resistor, but that is a lot cheaper than a loudspeaker.

[EC8010]

I don't know about you, but I don't fancy having to listen to a full amplitude 1kHz sine wave (maximum power test) or even a low-level 1kHz square wave (stability test), let alone a full power 10kHz sine wave (slew rate test) that would probably quickly fry a tweeter.

And, as the others have pointed out, a true resistive load might not be representative of a real loudspeaker, but it's definitively and repeatably unrepresentative.

[SGregory]

Ever listen up close to 20W 1kHz Sine Wave on a 90 dB speaker. After a few minutes there won't be a need for HiFi.

A non-inductive resistor is the way to go. If you want to simulate real life then simulate it with a dummy load that doesn't change after a few seconds of power like speaker does and can be repeated.

The signal will not be the same.

[chrish]

Along the lines of this discussion, I have tried to source some reasonably priced high wattage 8R resistors, but my usual sources are always out of stock. I do have a pair of 6R 50W that I have been using. Is this going to induce big errors in my testing?

G'Day Rob, I have moved again and I am literally just around the corner from you now if you feel like sharing ideas over a beer or need to use any of my equipment!

Chris

[Wavebourn]

Both.

A dummy load as an attenuator for a speaker.

Originally Posted by chrish

Along the lines of this discussion, I have tried to source some reasonably priced high wattage 8R resistors, but my usual sources are always out of stock.

I have a lot of 20 Ohm resistors. If to connect 5 of them in parallel it is 4 Ohm in total, for higher power. If to add computer fans for forced cooling they will dissipate even more power without a problem.

http://wavebourn.com/forum/download.php?id=437&f=7

http://wavebourn.com/forum/download.php?id=438&f=7

[chrish]

Originally Posted by Wavebourn

Both.

A dummy load as an attenuator for a speaker.





I have a lot of 20 Ohm resistors. If to connect 5 of them in parallel it is 4 Ohm in total, for higher power. If to add computer fans for forced cooling they will dissipate even more power without a problem.

http://wavebourn.com/forum/download.php?id=437&f=7

http://wavebourn.com/forum/download.php?id=438&f=7

That looks serious!

Fortunately, most of my stuff has been low power. I have 6R aluminum clad 50W mounted on a heatsink because I found them at an electronics surplus place cheap. Every time I try for 8R or 4R they are not there. I could bite the bullet and buy some from RS when I place my next order, but being cheap, if the 6R is OK I won't bother.

[Rob11966]

Originally Posted by EC8010

I don't know about you, but I don't fancy having to listen to a full amplitude 1kHz sine wave (maximum power test) or even a low-level 1kHz square wave (stability test), let alone a full power 10kHz sine wave (slew rate test) that would probably quickly fry a tweeter.

Yes, the sound issues are also a problem and a good enough stand alone reason for a dummy load. I almost wrote 'noise issues aside' in my post. I was really more interested in the 'scope trace - as many posters have confirmed the trace will be different. I guess in a perfect world, with a non-inductive load a square wave in will produce a perfect (but amplified) reproduction of the square wave. How does an inductive load (speaker) change the appearance of this wave? I will end up doing some experiments myself eventually but does anyone have any links or info?

Cheers,

Rob

[Rob11966]

Originally Posted by chrish

Along the lines of this discussion, I have tried to source some reasonably priced high wattage 8R resistors, but my usual sources are always out of stock. I do have a pair of 6R 50W that I have been using. Is this going to induce big errors in my testing?

G'Day Rob, I have moved again and I am literally just around the corner from you now if you feel like sharing ideas over a beer or need to use any of my equipment!

Chris

Hi Chris, good to hear from you. For sure, sounds great, drop me a PM with your phone number and I will give a call.