HP 8903B

[cncjerry]

I have an HP 8903 audio analyzer.  It looks like it can handle up to 300v AC input, or combined with DC, 425V.  Is there any reason to worry about feeding it off an 8ohm resistor with a 100W amp?  I have a class D amp I want to analyze and only ever used the 8903B for preamp level signals.  The 100W amp is using a 36V power supply, I doubt it will actually hit 100W.

Thanks

Jerry

[aliarifat794]

 With a 36V power supply for the amplifier, the output voltage will likely be lower than 36V RMS due to various factors like power supply limitations and amplifier efficiency. However, even if it's not hitting 100W, you'll still want to ensure that the output voltage doesn't exceed the maximum input voltage rating of the HP 8903 audio analyzer.

[macboy]

When I had a 8903B, I commonly connected the inputs directly to the output terminals of > 100 W RMS audio amplifiers. The inputs are designed for those signal levels, so have no fear.  I have since acquired a VP-7222A, and sold the 8903B. This also has balanced HV-capable inputs which can directly connect to virtually any audio amplifier. I assume that most such audio analyzers have similarly capable inputs.

Be aware that class-D amplifiers will produce a lot of ultrasonic noise, and the 8903B will absolutely include this noise in the distortion measurements. Technically, it measures THD+N or Total Harmonic Distortion plus Noise. This means that the THD+N measurement includes everything in the input signal that is not at the the fundamental (test) frequency. The switching noise of the class-D amp will be included. You can't hear this noise which is usually 30 kHz or higher, but the 8903B has a bandwidth up to approximately 700 kHz (from memory, I could be wrong).  In order to get a more representative measurement of the distortion, activate the 30 kHz low pass filter on the 8903B. You could instead try the 80 kHz filter if your amplifier has a very high switching frequency. When you use the 30 kHz filter, you want to limit the THD test frequency to about 3 kHz maximum. This ensures that you are not filtering out the distortion harmonics produced by the amplifier, which would result in a measurement reading that is artificially lower than reality. You usually want to consider the first 10 harmonics. The VP-7222A has a DSP-assisted mode which can filter out extraneous noise (signal components that are not harmonically related to the test signal) to give a THD-only reading, in addition to the THD+N mode.

edit:
Essentially all class-D amplifiers use a bridge-tied load (negative terminal is NOT ground). So, definitely use the floating balanced input mode, which requires connection of both input terminals, and not ground. There is a switch near the inputs to allow the negative input terminal to be grounded or floating. Set to float. You could damage the amplifier if you ground one of its outputs. If you are still unsure how to connect the amp to the instrument, just ask. With BNC input terminals, it isn't obvious what to do.

[cncjerry]

thanks for all the advice.  I'm just curious about the Class D amps so I bought one from sure electronics.  I has a ton of features.  I've been impressed with their DSP and audio boards in general, never tried their amps.  Their DSP boards use the ADAU1701 chip and it is programmable with SigmaDSP from Analog devices.  I built the demodulator for a DSP using their boards that worked very well.  It had a long path length so it wasn't useable for certain CW work but otherwise, pretty amazing.

thanks again!

Jerry