Hi there! I'm looking for some recommendations for bench/test equipment. Minimally I would like to improve my ability to test amplifiers that I build and repair. Stretch goal, I would like to learn more about the measurable electrical signatures of euphonic details in systems I like. Also, I may venture into reactive and acoustic measurement. Currently, the relevant stuff I have collected:
My second question is whether distortion analyzers (like the old HP units) are worth it. That is, given a good audio interface and analyzing in software (FFT), do they offer any advantage? I guess I'm trying to work out if they are worth the money and TLC to keep running well. They look super cool either way, heh 🙂
Last question is if I'm missing anything that has proven useful to you?
May your manifolds be smooth,
L2
- A few Fluke multimeters of various capability
- Tektronix Scope (pretty decent one, though more complicated than my older one that i miss dearly...)
- 24/192 multichannel audio interface
- Small collection of resistive loads
- Calibrated acoustic microphone
- A nice old hickock tube tester
My second question is whether distortion analyzers (like the old HP units) are worth it. That is, given a good audio interface and analyzing in software (FFT), do they offer any advantage? I guess I'm trying to work out if they are worth the money and TLC to keep running well. They look super cool either way, heh 🙂
Last question is if I'm missing anything that has proven useful to you?
May your manifolds be smooth,
L2
The HP 3312A signal generator and its modern equivalent, the 33112A, are very handy. Get one that's calibrated if you buy a used one.
The HP 8903 is an excellent instrument as well. It comes in an A and a B version with the 'B' being a bit more desirable. That said, the difference in practice is subtle. I wrote quite a bit on that here: https://neurochrome.com/pages/hp8903 The 8903 will get you down to -80 dBc THD (0.01%), though the handful of samples I measured reached -95 dBc at 1 V, 1 kHz, 20 kHz bandwidth. For the vast majority of equipment that's plenty for fundamental "is it broken?" testing and may even be enough for performance testing.
The next step up is Audio Precision's analyzers. The older System One requires a DOS PC to work. You may be able to get a System Two 20xx, 22xx, and 23xx to work if you have an older PC as well. They require a special interface card as I recall. These System Two should work with AP's USB-to-APIB interface:
Cascade:
You'll notice that even the used AP analyzers tend to be pretty pricey, but the hardware is amazing and the software is a joy to use (and I don't say that often about software). For a lower cost alternative, the offerings by QuantAsylum look very capable: https://quantasylum.com/
That said. Before you rush out and buy an audio analyzer I suggest that you play with your current audio interface. You may find that a few attenuators and some free software (REW, ARTA, etc.) will cover your needs.
Tom
The HP 8903 is an excellent instrument as well. It comes in an A and a B version with the 'B' being a bit more desirable. That said, the difference in practice is subtle. I wrote quite a bit on that here: https://neurochrome.com/pages/hp8903 The 8903 will get you down to -80 dBc THD (0.01%), though the handful of samples I measured reached -95 dBc at 1 V, 1 kHz, 20 kHz bandwidth. For the vast majority of equipment that's plenty for fundamental "is it broken?" testing and may even be enough for performance testing.
The next step up is Audio Precision's analyzers. The older System One requires a DOS PC to work. You may be able to get a System Two 20xx, 22xx, and 23xx to work if you have an older PC as well. They require a special interface card as I recall. These System Two should work with AP's USB-to-APIB interface:
Cascade:
- 2422
- 2522 (96k)
- 2500 (96k)
- 2122
- 2622
- 2722 (96k)
- 2700 (96k)
- 2702
- 2712
- 2722-192k
- 2720-192k
You'll notice that even the used AP analyzers tend to be pretty pricey, but the hardware is amazing and the software is a joy to use (and I don't say that often about software). For a lower cost alternative, the offerings by QuantAsylum look very capable: https://quantasylum.com/
That said. Before you rush out and buy an audio analyzer I suggest that you play with your current audio interface. You may find that a few attenuators and some free software (REW, ARTA, etc.) will cover your needs.
Tom
In addition to the analyzers Tom listed, there is also the Analog Discovery device and the most excellent E1DA Cosmos.
https://digilent.com/shop/analog-discovery-3/
https://www.openaudiolab.com/measuring-amps/
Over time, I have to learned to find the EE Toolkit Pro app very very useful.
https://digilent.com/shop/analog-discovery-3/
https://www.openaudiolab.com/measuring-amps/
Over time, I have to learned to find the EE Toolkit Pro app very very useful.
Instead of vintage I suggest you take a look at Quantasylum. They have a great analyzer and it's TINY.
I designed and built an audio signal generator a long time ago. The signal generator serves two needs:
1) Low-distortion sine waves. This requires a sine-wave oscillator with a very slow feedback loop to control amplitude. Inexpensive generators have high distortion.
2) Fast rise time square waves. This is needed to test response to >>20KHz harmonics. DAC-based generators have slow rise times.
I never got around to the hardware distortion analyzer, but I have done a lot of work with software FFTs. People on this board are designing and building amplifiers with distortion below the noise floor. Analysis requires a FFT to separate the harmonics from the noise.
Ed
1) Low-distortion sine waves. This requires a sine-wave oscillator with a very slow feedback loop to control amplitude. Inexpensive generators have high distortion.
2) Fast rise time square waves. This is needed to test response to >>20KHz harmonics. DAC-based generators have slow rise times.
I never got around to the hardware distortion analyzer, but I have done a lot of work with software FFTs. People on this board are designing and building amplifiers with distortion below the noise floor. Analysis requires a FFT to separate the harmonics from the noise.
Ed
You are going to need a dummy load if you want to test into anything but open circuit. Dummy loads can be resistive (a bank of power resistors in a bucket of water will do) or a simulated speaker load (RLC circuit with high power capabilities).
A cheap arbitrary function generator can be useful - you won't get low distortion sine waves, but many models go upto 10,20,30MHz or so and will do your fast edged square waves, white noise, etc. They have plenty of level too, so you can investigate overload performance and clipping.
Hi Tom
All of the System One's will run on S1 DOS or ApWin (APIB, APIB USB) some systems will run on SERIAL or GPIB also. New PC's don't have ISA or PCI slots and only have USB. ApWin 2.14 or 2.24 need a special APIB USB ADAPTER (Non AP) to run on USB. I have stock of the APIB USB ADAPTERS.
Duke
Interesting. What I stated above is what the AP rep told me back in 2014 when I was looking at getting a System Two. That was based on the AP USB adapter. Now that you mention it I do recall that there was an effort here to get USB going on the older analyzers. Sounds like you have a solution. Do you have a website set up for this? I'd love to write a page about audio analyzers and would link to your page.
Tom
Hi Tom
Send me Email duke.aguiar@ieee.org I will send you all of the info.
http://www.sound-logic-la.com
Duke
Send me Email duke.aguiar@ieee.org I will send you all of the info.
http://www.sound-logic-la.com
Duke
