Having fun with Analog Device precision ADC evaluation modules, low price compare to competitors. Performance is impressive, 500 ksps 16-bits differential, less than -120 THD with properly sized signal source. Video tells more details.
Unit in this thread is AD7693, I also posted somewhere on this forum few pictures SA based on AD7984 (1 Msps, 18-bits, diff.) Both modules pretty much the same on THD level, but 16-bits easier to process on uCPU with 16-bits SPI only.
DSCF1008 - YouTube
DSCF1009 - YouTube
Unit in this thread is AD7693, I also posted somewhere on this forum few pictures SA based on AD7984 (1 Msps, 18-bits, diff.) Both modules pretty much the same on THD level, but 16-bits easier to process on uCPU with 16-bits SPI only.
DSCF1008 - YouTube
DSCF1009 - YouTube
Attachments
I wonder why devices like these aren't being used in available audio test gear. (To be fair, I don't know what Audio Precision uses...)
Audio Precision uses a combination of AK5394A and AD7760 in the APx555 B-Series: Archimago's Musings: A look IN the Audio Precision APx555 B-Series. Thoughts on price and value in audio products.
There are a lot of spectrum analyzer, voltage network analyzer kits for RF in the ham radio world. The evaluation modules from Analog Devices and Texas Instruments get you an inexpensive package with a lot utility!
FWIW, Audio Precision now offers its software for other analyzers -- at a cost of $3,000.
FWIW, Audio Precision now offers its software for other analyzers -- at a cost of $3,000.
Thanks for commenting.
I should say, that there is no any Notch-filter in the test setup.
Measurements done "directly", w/o any software correction for ADC or analog
front-end non-linearity, that may be introduced if necessary.
Hardware is "open source", with circuits diagrams freely available on-line.
That includes
1. Nucleo-stm32f767zi
2. AD7693
PulSAR ADC PMODs [Analog Devices Wiki]
Module installed vertically on AD is SE-to-Diff converter, mcp6D11 in its basic configuration
out of DS.
I should say, that there is no any Notch-filter in the test setup.
Measurements done "directly", w/o any software correction for ADC or analog
front-end non-linearity, that may be introduced if necessary.
Hardware is "open source", with circuits diagrams freely available on-line.
That includes
1. Nucleo-stm32f767zi
2. AD7693
PulSAR ADC PMODs [Analog Devices Wiki]
Module installed vertically on AD is SE-to-Diff converter, mcp6D11 in its basic configuration
out of DS.
Last edited:
Isn't the AKM part obsolete now? Between that and the fab fire problem, I wonder what they'll be using in the future.
The RF test gear is also very inexpensive. A tinySA is ridiculously good for the price. It'll certainly do all the "sniffing" type stuff that audio experimenters might want to do and is also good for what the majority of hams want to do. It's even very useful for pre-certification testing. As good as new Keysight analyzer? Of course, not. But, do you need that? I don't.
The variations of the NanoVNA are also really impressive. When calibrated properly, they compare very well against an older generation HP/Agilent 8753 series.
The closest audio test gear that seems comparable is the QuantAsylum QA401/402.
For hobbyists, all this is really great progress. I have two camera gear lens/accessory cases filled with a semiconductor curve tracer, an Analog Discovery 2 and accessories, some QuantAsylum gear, a tinySA, a NanoVNA, a remote data logger, and other stuff. Pretty much a complete lab that all fits on a shelf in a closet for the 99% of the time it's not being used. Production suitable? Nope. Building and testing my own stuff? Great. All for less than what that AP software costs, with no yearly subscription.
Not much call for VSWR / return loss / impedance / insertion loss measurements for audio - only frequency + phase response would be useful. The NanoVNA is 50kHz and up, the thinySA 100kHz+ so not relevant anyway.
RF measurement tools don't have the resolution for any kind of audio distortion measurements, they separate harmonics using explicit filtering, not inherent linearity (ie they are narrowband).
Except...
When your amplifier is oscillating at 29 MHz and touching a scope probe anywhere kills the oscillation. Then, a field probe attached to something like a TinySA can be very helpful.
When you are trying to track some systemic noise coming from, say, the microprocessor in the dishwasher.
When - OK, no point in going on.
Neither replace the most basic test instrument (voltmeter). Nor are they a good replacement for an oscilloscope. But, a TinySA can be really valuable. For the price, if it saves me a couple hours of pain it's already paid for itself. But, as you said, the NanoVNA has far less application for audio work.
The by now ancient HP3577a 5Hz - 200MHz has a 1 Hz bandwidth setting (but the dynamic range is only 80dB) You'll risk a hernia lifting it, however! (I's not happy with DC on its inputs either)
Hi All
My NanoVNA is a GREAT TOOL. I also measures R, C & L values and way smaller / lighter than several other of instruments.
Duke
My NanoVNA is a GREAT TOOL. I also measures R, C & L values and way smaller / lighter than several other of instruments.
Duke
