NotAPreppie

I've been torturing myself with Arduinos at work and have come to the conclusion that it should be possible to build your own rotary compression tester using an Arduino.

I'll be detailing how to do this over the next week or two but I have the parts list and worked out all of the mechanicals and electronics already. All that's left is to work out the code.

Intro/Plans:
The basic plan is to use an Adafruit ItsyBitsy M0 Express with 20x4 LCD display, barometric pressure sensor, two $20 200 psi pressure transducers, and a LiPo battery. Everything will sit inside a plastic enclosure with a USB panel pass-through to allow charging and updating the Arduino. USB hardware will also be used to connect the pressure transducers to the Arduino but only to carry analog signals (no actual digital USB signalling will be used for this).

The goal is to have it function similarly or identical to the RotaryCompressionTester.com RCT-V5. That said, I have no plans to assemble these for people regardless of how much money they throw at me (Barring a few friends. If you have to ask, you aren't one of them). If you want a turn-key compression tester that works out of the box, you should absolutely buy one of those units. I have one of the early versions and it's great (that's the reason it's my benchmark). They're even a forum sponsor. The guy behind RotaryCompressionTester.com is a cool guy and deserves your money.

This project will be targeted at people who chuckle at the following joke:
"If you've ever built your own television, Linux is the operating system for you."

Why? Well, you may have to do some tinkering with the code. It will work well enough but there's a chance that you will have to calibrate the sensors yourself (because paying $20 for $100-$200 industrial parts comes with caveats). Also, you will need to know how to solder and might even need some electronics experience.

Parts List

Notes:

• This setup uses pressure 2 transducers to measure both rotors simultaneously. I can save $20+ by only using one.
• You can save another $12 if you leave out the barometric pressure sensor. I included it for thoroughness. If you're at an altitude above a few thousand feet, you probably should use it.
• The Adafruit ItsyBitsy M0 Express uses 3.3V for all of the I/O pins (Arduino Uno and Mega 2560 use 5V). If you're going to use any 5V parts, you have to shift the voltage down to 3.3V or you'll damage the µC. Since I haven't found any LCD backpacks that use 3.3V, I used a bi-directional level shifter to take care of that.
• The pressure transducers on that list are spec'd for 5V. I built a voltage divider and it worked fine but then I thought to just try feeding the transducers 3.3V: that works just as well. So, screw using extra parts: just feed them 3.3V.
• The Adafruit LiIon/LiPoly battery backpack includes protection circuitry and will charge the battery. You can use the LiPoly cell on the list or any 1S LiPoly or LiIon cell you have handy. I have a metric crap ton of 18650's laying around (it pays to be on good terms with the IT guy when laptop turnover time comes around) so that's what I'm using. The ItsyBitsy will output 5V on the "USB" pin even when powered by a 3.something volt Lithium battery.
• I found the transducers were actually reasonably non-noisy (most unexpected). That said, the % relative standard deviation (StDev/average *100%, sample size=20, sample interval=1-50 ms) went from 0.2% to 0.1% by putting a 47µF capacitor between the signal input and ground. I'm omitting that since it's not really required (and I don't even know if 47µF is a reasonable number, it's just the only cap I had around). If there's enough demand, I'll buy some caps and see if I can optimize that.
• In my preliminary tests, this setup is able to sample at roughly 6.2 KHz. That's waaaaaaaay more than enough so I'm not at all concerned about sampling frequency. The old ReCT-04 (used a DataQ DI-145) from the now-defunct Rotary Diagnostics had a max sampling rate of 240 Hz and that was a commercial product.

EDIT: 2019.02.24
Code seems to sort of be working. I had to do some really silly things (like figure out how to make an Arduino output a sine wave {which required learning how RC filters work and re-learning trigonometry} and make another Arduino act like a rudimentary o-silly-scope to make sure that was happening) but I'm at the point where I just need a few more pieces of hardware and I can actually test it on my car.

More to come...

NotAPreppie

Placeholder for assembly instructions.

NotAPreppie

Placeholder for code information
Everything will be open source. My code will be released under Creative Commons 4.0 Attribution-ShareAlike copyright. The libraries and references that aren't mine will be subject to their own copyrights.

NotAPreppie

Placeholder for future plans to make a custom-etched PCB for everything.

limitlesscodes

I'm more than excited to see how this turns out . I built a simple compression tester with a single transducer, non-fouler, and a spare pro mini. Mine also has one of those super inexpensive 1" OLEDs as well, but I mostly use the data fed back to the host PC via serial to do some post processing, outlier detection, etc. If you want any help feel free to PM me.

NotAPreppie

Thanks.

I tested my code and it worked as far as starting and stopping but the pressure data collection was wacky.

The project is stalled for the moment while I wait for another Arduino (I swear these things multiply faster than tribles) to act as a wave function generator. The Adafruit M4 units have SAMD51 processors with 2 genuine analog output ports. Just need to knock together some code to have it output 5 Hz sine waves offset 180° from each other.