1988 Fadal 4020 Conversion Thoughts: MESA 6i24 + 7i49 + ?

hilo90mhz wrote:

andypugh wrote: Here is the code:
github.com/LinuxCNC/linuxcnc/blob/master...mot2/resolver.c#L354

Edit: Okay I realized my mistake - this is not the firmware source but the driver source that runs on the LinuxCNC computer and interacts with the firmware on the Mesa 7i49..

Yes, you need to look here for the Mesa source:
github.com/LinuxCNC/hostmot2-firmware/bl...ter/src/resolver.vhd

(And this firmware would run on the 6i24, the 7i49 is essentially passive)

andypugh wrote: Yes, you need to look here for the Mesa source:
github.com/LinuxCNC/hostmot2-firmware/bl...ter/src/resolver.vhd
(And this firmware would run on the 6i24, the 7i49 is essentially passive)

I had just found that repository - I notice the most recent commit is 4 years old, does that mean there have been no updates to the firmware since then? or just not merged from some upstream Mesa source?

andypugh wrote: (And this firmware would run on the 6i24, the 7i49 is essentially passive)

Of course you are right! and I knew that just wasn't thinking about it fully..

Do you recommend using shielded 50 pin IDE cables? for connections between FPGA and daughter boards? I can shield them myself with conductive fabric style tape but not sure its necessary.

-Chester

hilo90mhz wrote: I had just found that repository - I notice the most recent commit is 4 years old, does that mean there have been no updates to the firmware since then? or just not merged from some upstream Mesa source?

Not merged from upstream. You can find the latest source under "Specifications" for each card in the Mesa web site

Do you recommend using shielded 50 pin IDE cables? for connections between FPGA and daughter boards?

If you keep them fairly short there appears to be little trouble.

I have taken apart my Y motor again as I wanted to get to the bottom of the roughness in the shaft. I believe all the bearings are shot in this motor + the tachometer commutator is not concentric with the shaft and I can visibly see it moving the brushes up and down .5mm or so out of round, it also makes a clicking noise at a certain spot on the shaft that varies with rotation.

Also the resolver is a more accurate 3 arc minute variety it turns out. But bad news is the bearings inside resolver are totally gone.. super rough spinning. I have disassembled it and the rest of it looks okay but I would at least have to replace these bearings.

The bearings are 0.5" OD 0.125" ID and 0.109" wide which is a very unusual size. I could make spacers to adapt a standard size 0.25" OD bearing to 0.5" OD..

Again re-evaluating options

hilo90mhz wrote: I the tachometer commutator is not concentric with the shaft
...
But bad news is the bearings inside resolver are totally gone.

Interesting. Generally those two components would never fail, they are zero-load bearings, so I would be looking upstream for what has given them a hard time.

It's actually a bit unusual to have resolver bearings, generally the rotor is mounted directly on the motor shaft and the stator to the backplate.

If these are brushless resolvers there is nothing to worry about dismantling and reassembling. No springs to go ping or magnets to be desaturated. Bearings should be an easy fix, but the question is _why_ they failed.

They fail quite commonly on these Fadal machines.. It is usually because of improper shaft alignment where the flexible aluminum coupling puts extra stress on the bearings. It is also likely original and 30+ years old and I believe this entire machine has had a rough life with high duty cycle production. According to all the Fadal message boards the resolvers are a common failure item for the Fadal and they recommend you keep a spare handy.

At least the other two axis motors seem to have been replaced recently along with their resolvers / tach.