GRV-IDC24 Inputs Failing

Hi,

For the first time ever I am experiencing failures from a GRV-IDC24 module.

It started with a input failing, I then moved it to another input and it lasted a week or so then that also went. Changed the module, this worked for approximately 3 weeks and then also failed, moved the point, worked for couple of hours then failed again, moved to another and now working for 5 days with no failure

In the field I have a simple proxy sensor, connected to a 24VDC coil relay. On the N/O of the relay I bring a 24VDC supply from my power supply in the Groov panel and return it to the input on the GRV-IDC24 module.

Sensor is triggered, relay switches and input on Groov goes high.

When I measure the input voltage I get 24VDC, however when the relay switches this jumps to 30VDC then returns to the 24VDC.

Is there something I can place inline to stop these ports from packing up, eg a resistor ?
Assuming it is this slight surge that is causing it.

On the same line I have 2 of the same sensors, wired through the same make, type of relays in the same manner and they do not have this problem where the inputs fail. Only the one does this.

Advice would be appreciated

Regards
Robin

You mention that you have two other sensors of the same time that are working ok, but I wonder about their lead lengths / paths through the machine?

While that module says its protected for 1500 volts for 1 minute, I’d be testing a TVS diode on that input to see if it helps.

Is the NO contact and the IDC-24 the only “load” on the 24VDC power supply?

My guess is it could be poor output regulation on the supply or it the supply output is heavily inductive. Perhaps a small parallel dummy load on the supply that doesn’t get switched off may prevent a large spike (I bet the spike is a lot higher than 30VDC to be taking out the module). The IDC-24 current should be around 2-3mA at 24VDC, so a 20mA load should be good to see if the problem goes away. Or get a better power supply.

Thanks Beno, replied via email

System is using approximately 8 inputs, and 8 outputs. All 24VDC.

Inputs consist of three proxy’s, limit, selector switch, push button and run signals from the motor contactors.

Output’s are wired from the ODC to a slimline single contact relay then to the field.

All are wired through a unshielded multicore to the main panel (11m), only place where it will be close to the robot and motor control is where it enters the bottom of the main panel.

Connected the the power supply permanently is a scale indicator. 24VDC

I had a Weidmuller DR-75-24 Power Supply in there and have now changed it to their PRO EC) 72W 24V 3A

Have used these components in combination in all my installations never to experience this. On another sight I am running 70 in and 64 out with no problems.

I therefore lean in favour of Beno’s assessment, ESD coming from somewhere in the field.

Therefore considering the route of:

  1. Installing the TVSD on this channel that is causing the line failure
  2. Linking my 0V to the GND line in the Groov Panel to eliminate any risk of floating

And then if it reoccurs replacing the unshielded cable with a shielded cable with the shield grounded on one end

Other option is to take my focus and work these points in reverse and try and eliminate the problem from the field starting with the shielded wire.

Regards
Robin

Hi Beno,

Can you have a look at this spec and advise if this will be a suitable starting point for the TVS diode.

P/N : 1.5KE18CA
Description: TVS DIODE 15.3VWM 25.2VC DO201
25.2V Clamp 60.3A Ipp Tvs Diode Through Hole DO-201

Regards
Robin

Looks to be the same part number and yes, that’s exactly the sort of thing I had in mind to protect and hopefully eliminate that induced spike.

Thanks, let you know how matters progress

Update, this installation has not dropped a I/O for two months now.

At the end I added a 4.7k ohm resistor to each input line , this appears to have eliminated the problem.

Regards
Robin

4 Likes

Thanks for the follow up. Did you put this in parallel or in series?

2 Likes

Hi Phillip, they were fitted in series

1 Like