7C80 + 7i74 and 7i84

Hello,

we are working on a small gantry mill with xyz +b (table index) axis and a tool changer.
Starting with a Rpi4 and a Mesa 7C80 we have to expand the system with a 7i74 and a 7i84...

The 7C80 is running fine (after some hours of commissioning), but we need the 7i84 to access additional I/Os for the tool changer.
Now we think, that we have to load/create a *.bit file for this combo..?
Is there an easy way to create this files or does someone have a file for this combination of cards?

Kind regards, i hope someone can help us

If you just need the one 7I84, it can connect directly to the 7C80 with out adding the 7I74.

If you need more than one SSerial channels, here is a 7C80 bitfile that has 7I74 support:

Great - the idea is, to use the other serial channels of the 7i74 for encoder feedback. We want to connect additional SSI/BISS multiturn absolute encoders for all axis. Is there anything special we have to look for, before ordering the encoders?

17bit absolute with 12bit multiturn 10-30VDC SSI/BISS would be the product of choice...

Thanks a lot, for your quick and reply and the bitfile

OK if you need SSI or BISS encoders, that needs a different bitfile

How many channels do you need?

I would suggest SSI if you need many channels because it uses less FPGA resources

7C80 configuration with 8 SSI channels on 7I74:

Thank you very much - I think we are going to use 5 channels in the near future.

Actually we have some problems with the voltage regulation on the 7C80 for the rpi4 - the 5.03Vdc seems to be to low, to power the rpi4. The connector TB15 can probably solve this issue, but there is no documentation how it is ment to be used

Is it possible to connect an external power supply with 5.2Vdc to this connector? Is there a jumper to disconnect the internal regulator?

Kind regards

No, its meant to be an output, What is the 5V rail voltage at the RPI?
Are you uses a short cable flat cable (1.25 inches is standard)?

You could probably get away with supplying 5.2V at TB15 because the local regulator
will just shut down if the output voltage is above its setpoint, but this also means
you are supplying 5.2V to every 5V device on the 7C80 card. I would check cabling
and the voltage at the RPI4 first.

we have checked the cabling - we use a very short 40pin cable (length < 30mm) - the rpi is directly mounted to the motherboard...

The 5V line is at 5.03VDC - all over the board - the rpi4 voltage (measured on the 40 pin header) is 4.95VDC.
We think, that it might be a better way, to disconnect the rpi4 from the internal 5V source (GND should stay) and power it separately, instead of pushing 5.2V to the mesa card. It is a pity, that the rpi4 needs a higher voltage than 5V to operate safely...

We are also searching for a failure of one "ISL32174 EIBZ" - we lost this part of the mesa card on stepgen2/3, while commissioning the servos. We use inverters from cs-lab (simdrive 400W) with standard differential ttl optocoupler inputs... we have checked the cabling several times, have done an analysis with differential probes on all lines (see screenshot) and we can not find any "physical" problem. Perhaps you have a clue...

It actually sounds a bit like a defective RPI4, At least I have never seen this issue
or heard of it from another customer, and have at least a year of uptime on my test system.

If the step/dir outputs do not swing a full 5V when unloaded, the driver chip has been damaged. This can occur if the outputs accidentally contact something with a voltage outside the 0 to 5V range. A max3042 is a replacement (ISL32174s are hard to find currently)

ok - we will replace the rpi4 and see what happens
The idea with the external power supply will be tested, if the under voltage message is still there, after the replacement...

Thanks for the max3042 clue - the ISL32174 is available in 2022...
Do you have a number for the version with 3.9mm width..?

Kind regards,