The parts got in a few weeks ago, but I just got back from 2 solid weeks of teaching obligations. The last big hiccup on time is my niece showing up. She should have her birthday by the end of the week and after that trek to Raleigh I should have some time to get to work on the Rix. My workbench is done and waiting, and while
@victorzamora will probably argue that I need to finish my DPV project first, I'm thinking that the weekend of the 25th will be dedicated to starting work on the Rix.
Step 1. Get everything cleaned. I have the compressors almost fully disassembled when Victor and I first found out they were in much worse shape than I initially thought, though that embarrassingly was in the fall of 2014. At that point I couldn't justify spending the money on rebuilding them as each of them needed close to $2k in parts, but times have changed. Ultrasonic is ready and the big parts cleaners at work will make it a quick job of getting everything cleaned.
Step 2. assemble the compressor to stock-ish. Since I'm going way out of bounds on what this thing is going to ultimately me, I want to start with a clean slate with the compressor "as stock". That will include the stock fan, but will not be run to stock speeds. My garage does not have 220v power, and at this point I can't put it in there, so I am limited to the going slow. I also don't want to mount a different motor and do the initial run-in at work where I have unlimited 3p power. Outside of the motor, the only real thing that is going to be different is a ball valve and QF4 on the outlet instead of a hose, and an adjustable pressure cutoff switch being added to cut the motor off.
Step 3. I am going to let the compressor run through an empty filter for a while to clean itself out and start watching with a thermal camera to see where the hotspots are and if the stock fan does the job. Once it clears out, I'll send it out for gas analysis without a filter cartridge just to see what I'm getting directly out of the coalescers.
Step 4. Start filling bank #1 to ~4500psi ish. They're 4500psi banks, so that's not the issue, but they may only get filled to 4000psi or somewhere in between. What is going to determine this is watching the logs from the VFD and thermal camera. Depending on what happens with the heat on the final stage, and the speed reduction since the VFD will be set to constant current mode, I may determine that the lower pressures make more sense if we start getting really inefficient or hot on the higher pressures. Who knows. Perks of having nice tools to play with. Noting the RPM at max pressure is going to be critical so I can set the compressor to run at whatever that RPM is for mixing nitrox. If running air I can leave it in constant current to maximize speed, but can't do that while blending.
Step 5. Based on temperature from the fan, convert to electric fans. Does that mean one big fan to replace the existing one? Does it mean several small axial fans mounted to various points on the compressor? Unsure, but will figure it out. Current thinking is a 10" diameter axial fan that will be tied to the input on the VFD and the auto-drains. Will be on whenever the compressor is plugged in to give some extra cooling after shut down in case there is any latent heat that wants to build up and just requires you to turn the fan off manually.
Step 6. Auto-drains. Will be using surplus Asco solenoids from work and will be figuring out the timer mechanism on them. I'm assuming draining in sequence is better than draining all at once. Will have two on the compressor itself, then one for the big coalescer on my filter stack. The drains on the compressor will be mounted to the frame so they don't shake the actual coalescers apart as Iain brought up in a previous thread. Power on the drains and fans will be tapped into the inlet of the VFD for now which is slightly annoying since they will need a separate switch, but oh well. Ultimately I think I will put a relay on there that is tied to the VFD that will control everything automatically, but that's down the road. Auto-drains are very much a requirement to be able to let this thing run without babysitting it, so getting this right is a HIGH priority once the compressor is up and running.
Step 7. Nitrox stick. Ultimately this thing needs to make EAN32 100% of the time. I have no use for banking air at home and anything that I would need compressed "air" for may as well be nitrox. With the nitrox stick, the VFD will have to change to constant RPM mode to match the O2 generators output. Will also be hooking up the normally closed O2 solenoid to the VFD so the compressor controls the flow.
Step 8. With the steps above the compressor
should be running stable. Next step wil be dependent on how motivated I'm feeling at learning arduino/rpi etc. I will either go with a screen that will display values, or just use cutoff switches. I would prefer to use the screen, but that's a LOT more work.
Controls would ultimately.
1st & 2nd stage pressure to look for valve failure.
3rd stage will be looking for duration maintained below a setpoint looking for leaks or excessive hammering.
1/2/3 stage temperatures. The key here is keeping the PTFE o-rings under 400f for longevity
Control of the fan based on the temps above with an auto-cutoff once they get back to some predetermined temperature, likely 100f or so.
Control of the solenoids.
I hope to have step 1 and 2 done by the end of the month barring any issues. Step 3/4 hopefully by the end of June since I have to get the motor mounted and VFD programmed with both tested before hooking up to the compressor. Step 5/6/7 will hopefully be done by end of July just based on vacations and weekend commitments. Step 8 who knows but unlikely before end of the year.
