Repeated Prox Sensor Failure

Bunt23

Member
Join Date
Jan 2019
Location
Ankeny, IA
Posts
25
Hello all,

I have a mixing machine that measures a specific amount of water, adds it to a mixing chamber and creates a homogenous slurry. The way we measure the amount of water going into the mix chamber is with a paddlewheel water meter. Omega FP-5063 is the specific part number. It has a paddlewheel with a metallic piece embedded in it that makes a prox switch each time it rotates. The number of rotations are sensed using a high speed counter input on a PLC. I’ve been asked to automate a few steps in our process, and to do so I’ve added 2 solenoid air valves and 1 water valve (essentially, the machine now has 4 more outputs [1 of the air valves is double acting]). Since I’ve made those modifications, our paddlewheel water meter failed. Thinking it was a coincidence, I replaced it with a new one, and that one failed shortly thereafter (completed 2 machine cycles prior to crapping out). Thinking maybe THAT was a coincidence I found one more, hooked it up to a bench top power supply, and manually spun the paddlewheel by blowing air into the meter while measuring voltage on the meters signal wire. It worked exactly as it should, so I put that one on the machine. Once again, I got through 3 or 4 cycles and the water meter quit working. I removed it and went back to the bench top. I hooked it back up to a power supply and blew air in it to manually spin the paddlewheel to see if it would output a signal. Nothing. Also did that with the other 2 that had failed. No signals from either of them as well. So they definitely failed somehow. I then proceeded to check all my connections in the electrical cabinet related to the water meter, and everything is wired per the manufacturer specs. I measured the voltage it sees, and it is 24VDC, per the manufacturer spec. I have this same water meter on 18 other machines that are wired the same way that work with no problems. The only difference between the problem-child machine and the rest is that I wired in a few more valves that the PLC turns on and off during the process. Other than that, the machines are identical. I decided to remove the valves I added to return the machine to its original state. Then, I put a new water meter in and ran some cycles. I think I’ve run over 40 cycles now and have had no issues. How can wiring in a few extra solenoid valves to some outputs on the PLC cause a paddlewheel prox sensor to fail? It just doesn’t make any sense to me. Can anyone shed some light on what may be happening?
 
Back emf from the new solenoid valves switching?


(BTW - Good first post)
 
Last edited:
The valve coils when disconnected generate a voltage surge, that is probably what spoils your meter, you should install surge suppressors on all the valve coils.
 
Wire a flyback diode in parallel with each solenoid and see if the problem goes away.

Not sure if a typical multimeter could catch the high voltage transient, but maybe if it has a peak hold feature. Otherwise you would need some sort of scope.

What kind of solenoids are these?
 
2 SMC SY3000 series solenoid valves (specifically, SY3100-5U1 and SY3200-5U1) and an SMC water valve (VXE2330L-03N-5D1-B)

The air valves are wired directly to the embedded outputs on the PLC, but the water valves are fired by relays, which are controlled by outputs on the PLC. Everything is 24VDC
 
We have a Fluke brand multimeter, and it does have a min/max/avg function... But I don't know whether or not that would be sufficient for picking up the high voltage transient.
 
They should all get surge suppression, but I'm betting the issue is primarily from the water valve, as that's a large solenoid. The 3000 series valves are piloted and have a much smaller coil, more like a small relay.

You can even put a diode on the water valve's relay to suppress that to be safe. What PLC and what output type (relay or transistor?)
 
Automation Direct PLC (part no. D0-06DD2-D) and the outputs are just 24VDC sourcing outputs. The water valve is wired to a relay (automation direct part no. RS6N-DE).

By looking at the wiring diagram of that relay, there seems to be a diode on the input side, which is what the PLC output is wired to, but no sort of protection on the output side, which is what the water valve is connected to.

There's +24VDC wire from our 24V terminal block bus connected to pin 15 on the relay.
One side of the water valve solenoid is connected to pin 12, and the other side of the solenoid is connected to the -24VDC bus. Could it be possible that when the water valve switches it is providing a surge back to the 24VDC bus, which makes ONLY the paddlewheel flow meter go bad? The meter receives it's power from the same 24V bus...
 
Yeah, it's possible those proxes just don't handle voltage spikes as well as the other components. Most industrial stuff doesn't care much, but maybe these are more sensitive.
 
Could it be possible that when the water valve switches it is providing a surge back to the 24VDC bus, which makes ONLY the paddlewheel flow meter go bad? The meter receives it's power from the same 24V bus...

Probably. I suspect the issue is that the electronics in the water meter are just too sensitive for what is a pretty typical industrial electrical environment. It might make sense to isolate the meter to its own power supply and/or put transient protection on it.
 

Similar Topics

I have a Allen Bradley L35E compactlogix with the following hardware: [0]1769-L35E (firmware V16.3) [1]1769-SDN/B DNET_R00_S01 [2]1769-MODULE...
Replies
5
Views
7,134
Hey guys, I am encountering a recurring failure of 1762-IR4 module for the past few months.I have replaced 3 modules in last 6 months.This...
Replies
4
Views
1,764
I am trying to us a analogue input signal, via a Micrologix1400 which is repeated as an analogue output signal that can be used by another...
Replies
7
Views
2,293
I have an array of 500 strings. Is there an easy way to check to see if there are any repeats without using brute force with a loop? Thanks. FNC
Replies
4
Views
2,897
This is probably a newbie question, so bear with me. I've been noticing that some things need a one-shot (ONS) so that they only get executed...
Replies
6
Views
3,015
Back
Top Bottom