DeviceNet E3+ Starting Unexpectedly

[jmalter 1]

We experiencing some unexpected starting of motors that are controlled by E3+ overload relays. We are using one of the inputs of the relay for a start signal from a hand switch in the field and I suspect there is an induced signal on the control wires when another motor starts. The 120 VAC control wiring is in the same cable tray as the 480 VAC motor leads and some of the runs are a couple of hundred feet long. I don't know how close the cables are in the tray. The E3+ as an EIMD interface to convert from 120VA to 24VDC. We tried installing a 33k ohm resistor from the input to common but this did not help. Our next option is to install a interposing relay. The buckets are crowded and it will take some time and skill to install. There are about 20 starters with this configuration although not all have experienced this problem. We couldn't measure the voltage as the signal is probably too fast. The problem is intermittent. Does anyone have any recommendations or suggestions? Rockwell was not very helpful.

[CoolBreeze 0]

The ones that are starting unexpectedly, where are they in location to the ones that are working correctly? Are all of them grouped together? What I'm getting at is try swapping one that works correctly with one that is acting weird, and see if the problem follows? No matter what the outcome is it atleast gives you a point to start looking at. Edited 25 Feb 2009 by CoolBreeze

[Ken Roach 194]

I had a similar problem with DSA modules a few years ago where the I/O wires were run for 200+ feet in the same conduit as the motor output leads. RA actually raised the load drawn by the 100-DSA modules in the next Series of hardware for exactly this kind of reason. How are you getting a Digital Input of the E3+ to act as a start signal ? I thought you could only use them for remote reset or 2-speed starter control. Are the motors starting for only a moment, or are they staying on ?

[BobLfoot 301]

Had a similar problem years back where an Omron G2A Relay was used as the MCR for a conveyor drive VFD Start circuit. The coil wires of the G2A ran alongside the 208 motor leads for almost a 1/4 mile. The induced current was 15 ma enough to keep tje G2A from releasing. Check for current not just voltage. And try an inline resistor.

[paulengr 44]

Why not simply rewire the start/stop switch (it is a contact, right?) as 24 VDC and forget about the 120 VAC logic since DC is pretty much immune to inductive pickup? Obviously you already have a 24 VDC power supply so the extra relay is just that...an extra point of failure.

[jmalter 1]

Thanks for the reply. I'm not sure how to size an inline resistor. We can try measuring the current but I'm not sure a standard Fluke meter will catch the signal.

[Ken Roach 194]

I went back to check a Series C user manual to see if something changed and I'm still a little confused. How exactly does an input on the E3+ start the motor ? When the voltage on a Digital Input goes high, it can: 1. Turn on a bit in the controller's DeviceNet scanner data table 2. Reset an overload fault 3. Command the overload to trip 4. Alter the overload's thermal calculation for 2-speed motor 5. Turn on a DeviceLogix bit that might be used to start the motor under DeviceLogix control. If you are doing the most common thing, which is to use the PLC's input data table to command the E3+ relay to close (which starts the motor over the network, then the PLC can help you debounce a momentary input signal. Can you describe how you are starting the motor with this discrete input ? I might be totally missing the point, like if your discrete input is wired in parallel with the E3+ output relay.

[jmalter 1]

Excellent point. We are using the input to the PLC. Assuming the cross talk is short duration we can add a timer in the PLC to make sure the start push button is held in for say one second. Anything shorter will be treated as noise. The operators may need to hold the switch a bit longer but the motor is right there so they can hold it until the pump starts.

[Ken Roach 194]

Experiment a little bit to see what the duration of these inputs appear to be (from the controller's perspective). Here's an example of how I've tested for debounce timing (RSLogix 5000 v16). jitter_logger.ACD

[DavidR 0]

Starting a large plant. Have aproxdimately 20 E3+ Series C and B throughout lighting like Christmas trees. Control cabling not shielded or near VFD outtouts seem to be the noise source and voltage inducing culprits. This theory has been played out and problem resolved by using shiel;ded VFD output cable to motor (long, and expensive run), re-running cable trays prendicular to 480 motor trays. and in a couple of cases interposing the input contacts. Still have a few that are geting noise frome somewhre. Had the same answer from Allen-Bradley as the rest of the posters...not much help. This is an issue. Our intellicenters (approx 250) are bridged by a Pyramid devicenet to Profibus and distributed to ABB 800xA controllers for plant wide DCS. Our equipment uses field stop lock out devices when pumps /fans are off. This prevents in advertant starting,but, the DCS still sees data from the E3 that field or network stars have been enabled. This is an issue regardless of what A-b claims. We are trudging through this. Noise seems to be the issue. We use the 120-24v input converters. I am curious about the "new and different block?" mentioned in a previous post. The terminator hasn't proved to be a problem in our testing. More than happy to discuss this by phone at 989-301-5960 or e-mail david.robins@hscpoly.com. Most fixes suggested have been tried . Cable shielding and separation have given us the best results. Dave Robins