Interface between Mesa 7i77 and Granite Devices VSD-E servo amplifier

More
8 years 10 months ago #103188 by DeckelHead
This is a somewhat generic request for information on how people interfaced their 7i77 Mesa card with a Granite Devices VSD-E servo amplifier. The specific area of interest is in the I/O, aka the COMMAND port, area.

Let me set the stage a bit... I'm bringing up a system that has been sitting for several years and I'm finally getting back to it. Most everything is done, but when I mothballed the project, I was working on the interfacing of two signals of interest: ENABLE and FAULT. The Mesa card has a Field voltage of 24VDC and the VSD-E generally has I/O of 5VDC

The ENABLE signal is connected to an OUTPUT port on the 7i77 and to the HSIN2 port (pin #5) of the VSD-E. That is an opto-isolated port using an HCPL2631. My thought there is to drive the port with a supplemental series 10K resistor. Unless I'm missing something, that should be sufficient.

The FAULT line is a bit more problematic. The output port on the VSD-E (pin #11) is an HCPL2531. Granite Devices' documentation indicates that IO-Vcc for this is 5VDC. Reading the documentation for the Mesa card, it seems like the only way to get the INPUT pin there to read TRUE is to drive the pin to the field voltage, or 24VDC. To perform this action, it seems like I will need to use a small opto-isolator board such as the GeeTech G213 ( www.geeetech.com/d213-optoisolator-break...jau1i5b1274026i6sm97 ), a few of which I have laying around.

Is this consistent with your design? I really don't want to blow up my VSD-E drives!

Any thoughts or ideas are always welcomed. If you have a question, please feel free to ask.

Thanks
Alan

Please Log in or Create an account to join the conversation.

More
8 years 10 months ago #103207 by andypugh
You don't have to use 24V, but it does seem that you need to use at least 8V, so a lower Field Voltage supply isn't the perfect answer.

The VSD-E manual does suggest an external resistor for widening the input voltage range of the opto-isolated inputs.
They are shown with an internal 220R resistor. The manual also says that the current threshold for activation is 6.3mA. Rather helpfully they also specify the input device used. ( uk.rs-online.com/web/p/products/1241346/ ) At 5V the current through the opto (1.4V fwd voltage from the datasheet) is (5V - 1.4V) / 220R = 16mA. To keep the same current at 24V you would need a total resistance of (24V - 1.4V) / 17mA = 1400R.
There is already a 220R in circuit, and the working current of the inputs is actually at the top end of the recommended range, so I think I would choose a 1.5k additional series resistor for a working current of 13mA.
Also, to be cautious, I would first activate the inputs through a multimeter in mA mode to make sure that the input current through the opto was >6.3mA and <20mA.



If you set the card to Mode 3 then four of the inputs (16, 17, 18, 19) have their logic threshold set to 2.5V. If you are not using MPGs on those pins then this seems like one answer.
Alternatively you could use the analogue inputs (in modes 1, 2 or 3) to read the fault output voltage and convert to true/false in HAL.

Or add a TTL logic expander board on the smart-serial expansion connector: store.mesanet.com/index.php?route=product/category&path=83_88
The 7i73 is cheapest, but the 7i69 might suit better.

Please Log in or Create an account to join the conversation.

More
8 years 10 months ago - 8 years 10 months ago #103211 by PCW
You can use 5V field voltage on a 7I77 (VIN must be separately powered by a 8 to 32 VDC supply in this case)


You cannot use the MPG input to get 2.5V thresholds with 12V or 24V VField because these have 2.5v thresholds for the MPG encoder logic only, not the digital inputs.

Also if you need lot of TTL level I/O a 7I90HD is probably a better choice than a 7i69 (72 I/O vs 48 and $59 vs $69)
Last edit: 8 years 10 months ago by PCW.

Please Log in or Create an account to join the conversation.

More
8 years 10 months ago #103215 by andypugh

PCW wrote: You cannot use the MPG input to get 2.5V thresholds with 12V or 24V VField because these have 2.5v thresholds for the MPG encoder logic only, not the digital inputs.


I was hoping you would pop in with a definitive answer on that one.

Also if you need lot of TTL level I/O a 7I90HD is probably a better choice than a 7i69 (72 I/O vs 48 and $59 vs $69)


It looks like he might only need 3 x 5V inputs, so maybe the 4 x analog inputs will be enough?

Please Log in or Create an account to join the conversation.

More
8 years 10 months ago #103216 by PCW
Sure, if you don't need the analog inputs for other uses.

Please Log in or Create an account to join the conversation.

More
8 years 10 months ago #103226 by DeckelHead
Ah! Just the dialog I was hoping to spark! Thank you Andy and PCW.

You guys have jogged my memory in a few areas, and that is great.... One thing I can't really do is change the output side of the Mesa because it is driving 24VDC relays. And, yes, my napkin math was wrong. I meant 1K not 10K, along with the the 220 series resistor already present. Your idea about validating an in circuit representation with a DVM is something I would have done but it is also good of you to suggest the same for someone else reading this thread later on; so, thank you for bringing that up.

An area that I had totally forgotten about is and that I believe you are driving towards is splitting off the input and output voltages used for I/O (sorry, I am intentionally avoiding Vin and Field because I don't have the documentation in front of me right this second). I vaguely recollect that this is possible and it may be the best solution for me. Ironically, I inserted an opto-isolator board into the limit switch area that takes the native Hurco 5VDC output and made it compatible with the Mesa 24VDC input. But, I could easily change the output target for that opto board to 5VDC if I want.

My next task, then, is to re-read the section regarding Vin and Field. If I can make the input side compatible with 5VDC while keeping the output at 24VDC then I think I'm set. I'll change the existing output side of my isolators for the limit switches to 5VDC and then wire up the Granite Devices' output (FAULT) opto isolators directly to the Mesa inputs. If it works the way I think it will, then that would be the easiest solution (still keep the 1K resistor in the ENABLE output as that would be limiting the current from the Mesa 24VDC).

One final note, although it sounds academic at this point... I do want to stay away from using the MPG Mesa I/O. I expect that I will be using those for the stated purpose.

Cool. You guys gave me some great ideas to research. The help is appreciated.... Have a great holiday season.

Please Log in or Create an account to join the conversation.

Moderators: cmorley
Time to create page: 0.323 seconds
Powered by Kunena Forum