SSR-RACK48 and PCIe-DIO96H with external PSU

[KoltonS]

I have an SSR-RACK48 populated with output relays being controlled by an PCIeDIO96H and am running into an issue. The PCIeDIO96H is powered the computer PSU and the SSR-RACK48 is powered by a separated PSU in the electrical enclosure it is mounted in. The two GNDs from the separate PSUs are referenced.

The problem I am having is that the pullup/down configurable resistor in the PCIeDIO96 is not active unless the board is powered (aka the PC is on). In my current setup it is possible for the PC to be off but the electrical enclosure to be on. I would expect that all the outputs would default to the off state but instead without the pullup/down enabled I see random outputs turn on which is not desirable behavior.

I am looking for a solution that avoids this situation. Currently I see a few options.
One would be to run 5v from the PC PSU to the system electrical enclosure and use that to control a relay and in turn control the power to the SSR-RACK48.
Second is there is 5V on the ribbon cable from the PCIeDIO96H but it's a N/C on the SSR-RACK. A bodge wire from the N/C line to the 5v input on the SSR-RACK48 would provide power but I am not sure is this was not done in the design intentionally.
Third option??? I am not sure, neither of the options I have proposed feel like the intended answer. Have I overlooked something or are these boards really just meant to both be powered off the PC or this odd behavior ensues?

Any feedback/input would be greatly appreciated. Thank you!

[KoltonS]

Yesterday evening I tried installing the bodge wire just to see if that would improve the situation and realized I seem to have a bigger problem. Even with the two being powered together I am getting erroneous outputs coming on before our software starts. From what I understand this is because the 82C55 emulation via the PCIe-DIO96H defaults all ports to inputs after a reset and the pull-down resistors (configured via InstalCal) should prevent the high impedance inputs from floating high and turning on relays on the SSR-RACK.

When I change the configuration of the resistors to pull-up all the outputs on the SSR-RACK turn on as expected. But when configured for pull-down the erroneous outputs come back again. Any ideas?

[Fausto]

Hello @KoltonS.

You have not missed something.  The SSR-RACK48 50-pin I/O connector was not designed with a +5 V line, unlike the PCIe-DIO96H 100-pin I/O connector.  Also, given the 82C55 emulation on power-up and reset, I would favor your first option, but allow the PC to be up and running before powering the SSR-RACK48.

22 hours ago, KoltonS said:

One would be to run 5v from the PC PSU to the system electrical enclosure and use that to control a relay and in turn control the power to the SSR-RACK48.

Does that scenario work for you?

The SSR-RACK uses inverting logic.  A low signal (0) output from a connected digital I/O board causes an output relay to activate, while a high signal (1) causes an output relay to deactivate.  The chips on the SSR-RACK which invert the logic are socketed and can be replaced by a user with chips that do not invert the logic.  (Note: Digilent does not provide a service to swap the logic chips.)  Please note, the IC's can be damaged by breaking the legs off of the IC's. Use an IC puller or similar method to remove them to avoid damaging the legs.

The standard IC that is installed on the SSR-RACK is a 74LS245. This IC needs to be changed with a 74LS640 to switch the logic on an SSR-RACK from inverting to non-inverting (standard logic).

Regards,

Fausto

[KoltonS]

Thanks for the info @Fausto

We have swapped the chips for non-inverting as you mentioned but now, I am wondering if I should swap back and just invert the software controls since it seems the pull-down option is unable to keep the SSR-RACK in a clear "off" state on power up, but the pull-up option seems to create the expected output.

As for powering the SSR-RACK (5v only) via a relay that is certainly an option.

One last question, I see a lot of the MMC parts have been marked legacy and kind of buried on the Digilent website, are the SSR-RACK48 or the PCIe-DIO96H going end of life anytime soon?

Thanks!

[KoltonS]

EDIT: Double post. Site froze up.

Edited January 9 by KoltonS

[Fausto]

@KoltonS,

3 hours ago, KoltonS said:

As for powering the SSR-RACK (5v only) via a relay that is certainly an option.

Consider this option given the issues you are having with the other two scenarios.

 

3 hours ago, KoltonS said:

One last question, I see a lot of the MMC parts have been marked legacy and kind of buried on the Digilent website, are the SSR-RACK48 or the PCIe-DIO96H going end of life anytime soon?

Not aware of the SSR-RACK48 and PCIe-DIO96H devices going EOL anytime soon.  

Regards,

Fausto