2 Mesa Cards

Please be sure to keep us updated on your results. I am planning to run the same VFD + Mesa.

Marvin

I ended up buying a RS-485 PCI card. I also changed the cable to a CAT 6 cable. So far everything is working. I have not run the machine that much though. Some how in the process of working I ended up breaking my mesa 5i20 card. I think it had been going bad for awhile and finally died. There were times where the hostmot driver would not load and I would have to take the card out of the computer and reinstall it to get it to work. Luckily my plasma machine uses the same card so I can use that one until I get a new one. After I have to run an order of 300 parts so I can see if the problem is solved.

If reseating the card fixes things temporarily I would suspect the card/card slot of being contaminated. Cleaning the slot/card edge connector with 99% Isopropyl alcohol should help in this case.
Another possibility is a mechancal intermittant on the 5I20 card (possibly a bad solder joint) If cleaning the card/slot does not fix the problem the card should be returned for warranty repair.

I tried cleaning everything. I thought it might be the slot. but my other card works in both machines and this card does not work in either machine. I will send it in to see if it can be repaired.

Looks like it is the RS-485 card some how breaking the mesa card. Now the other card does not work in either machine. I was running a program in the machine and suddenly it gave me a joint following error and the entire computer crashed. Then when I rebooted, I can no longer load the driver. One thing that I have noticed is that after this happens the board gets extremely hot. there is a component on the board labeled u4 that is very hot.

I will try and get these boards fixed and am going to change to controlling the VFD with a pwm signal.

U4 on a 5I20 getting hot means the local 3.3V power supply on the 5I20 is shorted out. This can be due to shorting a large number of FPGA pins to ground (backwards or bad flat cable is one way to do this) or a crowbarred FPGA due to exceeding its I/O pin voltage limits (This may be permanent = fried FPGA)

Any I/O pin pulled to more than 5.5V or below -.5V can cause the FPGA to crowbar which will short out the local 3.3V from U4, possible crash your computer and possibly cause permanent FPGA chip damage

Since this got worse when you installed the RS-485 card it may be that you have a ground loop in your system that is shunting current through the PC/FPGA card

Spindle drives are notorious as noise generators so this may be related

I do have a rats nest that I will have to sort out. If I got the 7i42Ta would that potentially avoid this problem in the future?

Yeah if you are using bare FPGA pins you have to be really careful with what connects to them (even for a microsecond)

The you tube video I just posted is:


Thanks for the comment. It works quite well mechanically.

I finally just finished the turret and live tooling motor setup.

Unfortunately, electronics is not my strong point and with all the solenoids VFDs and everything else, I have managed to create quite a rats nest of cables. (That in the end I think destroyed my mesa card). While I wait for the new card I am going to clean up the wiring. I also ordered the mesa cards to opto-isolate the PCI card. (Should have done that in the beginning but ....) I am also thinking of opto-isolating the RS-485 connector to the computer. I have only found one "black Box" that does that and it costs $160. So, I am not sure that it is required.

One big problem I believe I have is I am not using shielded cable from the VFD to the motor. I cannot seem to find shielded cable from any of the local electrical supply stores - so will have to order it.

I am in the process of making a tool that will drill radial holes that mounts to the turret. (Unfortunately or fortunately I guess other work for customers is taking priority.)