Advanced Search

That wire is on a 5V regulator pin, so maybe related, hard to tell.

Yes, NCV7608s are obsolete, we do have spares if you need a few.

Thanks, Tommy.
The shop is heated and very dry, especially during the German winter, so moisture is probably not the issue.
-
Max

Picture 2, that looks like arcing across those two parts, moisture can do that, but the rest of the boards look like there is no moisture in quantities to cause that.

Thank you for your response, Peter.
Is the NCV7608 still available? I only found one listing on Aliexpress, which makes me wonder if it’s becoming hard to source.

While taking a closer look at the board, I found a small piece of cut wire lying around in that area:
 
 

Could this have caused the issue?If not, what else could I check to resolve the 5V overvoltage?
The 7i85S is powered via the 25-pin connector from the 7i76E.
Thanks again for your help!
–
Max

Not sure about the 7I76E outputs, but the 7I85S seems to have been damaged by an overvoltage on the the 5V
supply, possible related to the original fault. It appears that both the IS32174 step/dir driver and U11, the LMV342
have been damaged.

7I85 U11 is part of the encoder circuitry when TTL mode is used, it will not affect differential mode encoders
so the 7I85s will probably continue to function for encoder inputs if they are differential, BUT the 5V situation
needs to be checked.

On the 7I76E, the damaged driver chip (NCV7608) should be replaced as it may fail further and perhaps cause
further damage.

Hello everyone,

Yesterday, I managed to blow up two Mesa boards within just a few minutes, and I’m still not sure why it happened or how to avoid this in the future.

The machine is a retrofitted Schaublin 125CNC lathe, which had been running flawlessly—until last Friday’s coffee break. When I came back, I found the central lubrication pump had burned out, apparently because it had been running continuously.

I discovered that output 6 and output 7 were permanently high as soon as the card was powered up. These outputs are connected to 400V 3-phase contactors (for the lubrication and coolant pumps), each with a 24V coil and a flyback diode directly across the coil. Each contactor draws about 275 mA.

I’m not sure which output failed first—or whether both failed at the same time. The coolant pump is already configured in HAL but hasn’t yet been wired to its contactor, so I can only confirm that the lubrication pump was not being controlled by LinuxCNC at all—it isn’t connected in HAL.

After powering down the machine and disconnecting the wire between the lubrication pump’s contactor and output 6, I turned the power back on. While sitting in front of the control cabinet and moving some wires around, I saw smoke coming from the 7i85S board next to the 7i76E. It appears that capacitors C22 and C24 have blown, and one IC also looks suspicious.

I’m only using the encoder inputs of the 7i85S, and those still seem to be functional. I suspect that one of the unconnected step/dir outputs may have been damaged.

While I can accept that switching contactors directly might not have been ideal (even if within spec), I don’t understand what could have caused the damage on the 7i85S—especially since nothing was connected to any of its outputs.

During my investigation, I also realized that I had used the VIN pin on the field power connector instead of pins 1, 2, 3, or 4—so field power was supplied through the internal jumper. Could that have caused issues?

By the way, this part of Norbert’s connection sheet (page 2) may be a bit misleading or unclear in that regard.

Both boards are still working, aside from the two damaged outputs and possibly one or more differential outputs on the 7i85S. Also worth noting: the 24V supply to the cards is not switched on the DC side; the PSU is switched on the AC side.

Should I replace the boards anyway, or are they still safe to use going forward?

Thanks in advance for any help or insights!

— Max
 
         

The 7i96s  looses some inputs to run MPGs. I would go with a Mesa 7i76E or the EU version. It has two MPG inputs. Perhpas PCW will be able to add a third. With the 7i76E the machine will run as fast as you can wind the MPG. There is a parllel port example in the docs so its pretty easy to adapt to the mesa card as its just changing the pin names.

That's what I'm starting to realize! Thanks for adding to my confidence on this route.

I have found some good encoders - I'm thinking of going with 10000 PPR for X and Z and 800 PPR for the turret and the live tooling motor as the latter won't need much in terms of resolution, while the Z and especially X will need as much as I can get/afford (maximum count rates of A-B Ultra3000 2.5 mHz).

 

I had red encoders on the two servos I was using but changed them for 1024 a b z encoders. They weren't compatible with the mesa card I bought- i thought they would be.  Was a lot simpler with standard encoders even with the startup positioning.

I have them in stock as well.
mesaus.com/product/7i76eu/

JT

yes, that is the "spindle-released" sensor. As mentioned above I have to convert it to a binary value as the 14V is below the (60% of 24V) digital threshold of my 7i76e.

I can see the "spindle-released" going high when the drawbar is pulled (so, no error when releasing the tool) but get the error when clamping the tool.

Hi,

I am currently integrating my ATC. However, I get the Error "Failed to release Tool" message when I clamp the tool.

My HAL looks like the following. I get a 0-14V Signal which is converted into a HIGH LOW for the Motion controller.

This very quickly is going to turn into a how to destroy some Fanuc encoders thread so stay tuned for that   

I've decided to replace the encoders... this is probably not the answer that future readers are going to want to read, but its the least expensive and most straight-forward approach I can come up with. Either I have a digital problem or a mechanical problem to solve, and the latter is much more manageable with my skill set. 

I'm trying to get better at knowing when to take a different path to the finish line. I once spent months trying to design a DC motor driver (90v 200amps) that ran on 600v 3phase for a Hardinge CHNC lathe based on open source electric vehicle schematics. Long story short, I didn't succeed and blew up a lot of expensive components along the way    Once I finally decided to swap the motor for an AC one and add a VFD, the whole thing was turning in in 2 days! 

More to come! 

The 7i76E is in (or eventually will be) the same enclosure (and same power cord) as everything else, except the PC, which is elsewhere. Putting just it (24V) on a UPS would be a technical challenge.
As for the motors, to keep the encoders powered that means another power supply (16V) in that enclosure that needs a UPS, and I'd have to get around to testing the following-error code
The motor's big power supply has pretty beefy capacitors - so much so that I have a discharge circuit on it's contactor for when I power it down.
My machine homes pretty accurately, at least enough for my purposes (woodworking) so a following error is not much better than a reset.
The spindle vfd is already monitored for faults (modbus).

Now I get to research tiny din-rail UPSs for 24V and 16V...

Maybe you could power the 7i76E from the UPS too? The spindle VFD can probably handle a sub-second power interruption on the bus capacitors.
The servos may stall, but there is some chance that they might recover position if the encoders remain powered. And if not, then you will get an f-error report, and still have a live system to recover with.