I use a HP 8903A distortion analyzer I picked up on eBay and repaired. But less fancy gear can be used as well. I generally measure amplitude response at 1 kHz, 1 W out. I measure distortion vs frequency and power.
If you use a DMM for amplitude response, first measure the response of the meter. Many less expensive DMM's don't go much past 400 Hz for AC measurements. Some of the fancier bench top multimeters have bandwidths upwards of 1 MHz.
Before I got the 8903A, I just measured -3 dB bandwidth on the o'scope. It's easy. Tweak the vertical control until the trace hits the 0 % and 100 % markings on the screen (use the little CAL knob on the vertical control for that) at 1 kHz. Then turn the frequency up/down until the trace hits 71 %. That's 3 dB down.
~Tom
If you use a DMM for amplitude response, first measure the response of the meter. Many less expensive DMM's don't go much past 400 Hz for AC measurements. Some of the fancier bench top multimeters have bandwidths upwards of 1 MHz.
Before I got the 8903A, I just measured -3 dB bandwidth on the o'scope. It's easy. Tweak the vertical control until the trace hits the 0 % and 100 % markings on the screen (use the little CAL knob on the vertical control for that) at 1 kHz. Then turn the frequency up/down until the trace hits 71 %. That's 3 dB down.
~Tom
It's a great topic to explore, but does force you to build up your measurement arsenal to extract better information.
If using a sine input, then you have acquired a sine oscillator, or using a digital sample - either way, you need to double check that when it enters the amp under test it isn't itself distorted/clipping as you push up the line level.
Most people test with a continuous sine output, so that any voltage supply sag and cathode bias (if cap bypassed) has settled at a continuous level. The measured output power level may or may not be the same as if you were able to set up a short burst test (as per typical music dynamics) where maybe supplies haven't sagged, etc.
Some people test power output at heavy/heaviest overload (ie. guitar amps) where the inherent power limitations of power supply, OPT, and output valve saturation reach their limits).
Using a software spectrum analyser (ie. via soundcard) is a very cheap way to appreciate the growth of harmonics as signal level increases.
If you test into overload regions then you should double check your voltage meter is a 'true rms' meter with a frequency response including the audio range.
Ciao, Tim
If using a sine input, then you have acquired a sine oscillator, or using a digital sample - either way, you need to double check that when it enters the amp under test it isn't itself distorted/clipping as you push up the line level.
Most people test with a continuous sine output, so that any voltage supply sag and cathode bias (if cap bypassed) has settled at a continuous level. The measured output power level may or may not be the same as if you were able to set up a short burst test (as per typical music dynamics) where maybe supplies haven't sagged, etc.
Some people test power output at heavy/heaviest overload (ie. guitar amps) where the inherent power limitations of power supply, OPT, and output valve saturation reach their limits).
Using a software spectrum analyser (ie. via soundcard) is a very cheap way to appreciate the growth of harmonics as signal level increases.
If you test into overload regions then you should double check your voltage meter is a 'true rms' meter with a frequency response including the audio range.
Ciao, Tim
- Status
- Not open for further replies.