Testing a 3 Phase Air Compressor
HTML-код
- Опубликовано: 10 июл 2024
- Upshot: It was worth a shot. Requires more chooch!
Topic index:
00:00 - concept and credit where due
02:08 - parts
03:29 - what I tried
04:37 - brief aside on fleabay air compressors
11:56 - summary
12:54 - copper cooling lines
AvE's videos:
• BIG COMPRESSOR PERFORM...
• Return of Dewclaw! | B...
My video on the treadmill controller:
• Hacking a 3-phase trea...
Donations help me make more cool shit:
paypal.me/swolebroshopworks 
Наука
Walaupun agak rumit,tapi penjelasannya bagus.sehingga mudah di mengerti. Sangat bagus sekali👍
Have you ever used setsserial? Mine says it is successful, but it does not change. I followed the tutorial on the man setsserial. I did not understood the //todo// “get”. But it seams important
Good discussion. I had to go back a few times to get all the bits :) Your air plasma uses 9CFM at 90 PSI and ~1% more energy is necessary to increase compressor output 2PSI. How about dialing down the pressure switch to 95PSI and run the compressor constantly while the CNC plasma is going? Your Hitachi VFD may even have an analog input where a pressure sensor could attach. Then the VFD might be able to adjust the speed up and down to maintain 90-95PSI. Maybe without a receiver tank, but with necessary pressure relief safety valves. Would be pretty cool if it works.
I think reducing the cut-out pressure on the current compressor to 95 PSI would be a *small* step backwards. It would definitely get more off time in its duty cycle then, but it would come in frequent short bursts. I can't imagine either the motor or the pump head would be too keen on that.
Hooking up one of these treadmill motors with the Hitachi VFD to the current shop compressor might be worth a shot. Better quality pump head, better flywheel for smoothing out the load, etc. I expect it'll be under powered, but what the hell, I've come this far already, right?
The Hitachi has more features than you can shake a stick at - the manual is about 700 pages! At least having "too much" documentation is a good problem to have, versus the common situation of getting none whatsoever. I know an analog input for PID control is possible, I just haven't gotten that far yet.
theory is when nothing works, but everyone knows why. praxis is when everythings working, but nobody knows why.
i tend to combine both: nothing works and nobody knows why...
I bought one of those compressors off ebay it lasted 2 weeks running the air compressor for my plasma table. The store on ebay refunded my money and didn't even ask for the compressor to be returned. BTW blew oil everywhere out of the air vent before it died .
That doesn't surprise me in the slightest. Hahah. What size compressor/motor did you have, and for which plasma cutter?
Honestly, if the Hypertherm didn't do such a long air purge after every cut, it wouldn't be so bad. With the time involved in rapids, probing, and piercing, only maybe half of a job is spent cutting... but the entire time is using the full amount of air!
@@swolebro it is a 65 amp hypertherm 3 hp 50 gallon tank , your right about the air purge for sure . I now have a 105 hypertherm and bought an ingersol rand air compressor and now using the 50 gallon air compressor tank as a reservoir to have enough volume , I built a 9 ft by 12 ft plasma table with 203v gecko drives linux cnc nema 34 motors and 10:1 gearboxes with rack and pinion - looking to eventually try out your THC design
@@JODY-ej5qy Do you have the 50 gallon reservoir because the Ingersol can't keep up with the 105? Oof.
One thought I've had before, but haven't needed to implement yet, is using HAL to automatically pause the CNC when the pressure gets too low. Use a pressure transducer (~$10) and an Arduino (~$2) to monitor it and output a binary go/no-go signal, which you read on an input pin in LinuxCNC and map over to some "stop the torch, lift the head, and wait" routine. If the Arduino had a current sensor, it could even enforce a duty cycle on the air compressor as well as pressure monitoring, all via the same no/no-go signal.
There's also the option of banking multiple air compressors together. The downside there is that their pressure switches and relief valves won't coordinate with each other, so it would require some more advanced Arduino brainboxery.
@@swolebro man i really need to pay attention to my notifications better, the 50 gallon tank is a precaution - i figure its better to have too much and not need it than to not have enough air and screw up in the middle of cutting. the idea with the pressure switch pause when the air pressure gets to low sounds good but i have had issues with restarting in the middle of a line of code where linuxcnc would not start back without backing up a few lines first.
@@swolebro have you seen this? he eventually ended up using an arduino uno there are those that claim the same can be done with a raspi but i would assume that it would need to be in c++ because python is slower in execution. ruclips.net/video/LxGfTt5eavE/видео.html