Allen Bradley 1794 AENT Flex Module

Hi Guys,

I have a slight problem, and a curious problem which you might be able to assist me with.

We have a ControlLogix 1756-L62, this is connected on the backplane to ethernet card 1756-ENBT/A. This in turn connects to the field I/O, which is Flex I/O modules 1794-AENT.

The module I have a problem with has 2 input cards 1794-IB16/A and 1 output card 1794-OB16/A.

Every so often the module fails, or should I say the network fails, but only on this node.

I think it is a network issue because of the LED's on the front of the Flex Adapter. When the network fails, the module status LED remains green, but the network status LED flashes red, and the link LED flashes green.

I have read the manual and all it says is that there is a Connection Timeout. What connection is this implying.

I have removed the input modules and the output modules and this gives a different fault, so I know those connections are good.

When I look in the I/O tree on Logix 5000, I get the little yellow exclamation marks on all modules in that 'node', but when I pull up the properties the status stays as running.

After 2 or 3 seconds the modules all come back online and everything works fine again for 20 minutes or so.

Now the problem I have is when the modules fault, all the outputs drop off, which I assume is for safety reasons.

Now the curious problem. Although all of the outputs drop off there are a couple of double solenoid controlled pistons. If these pistons are extended when the modules fail, they retract as if the output comes on to retract them.

And vice versa if the pistons are retracted, they extend, as if the output comes on to extend the piston.

I have watched the output LED's when the modules fail, and they all go off.

So I guess my first question is why the modules might be failing.

And my second question is why would the pistons act this way when the modules do fail.

Sorry for being so long winded, but i just wanted to give as much detail as possible.

Mark

pistons

it may be that the valve needs voltage applied to one side or the other to shift positions. when no voltage is applied the pressure tends to equalize thus leting the piston move.

I,m no expert but sounds like e-net problem do you have managed switch,how long is the cable how heavy is the traffic is this a new problem? or has it always been this way?

BCE123,

This is a totally new problem. We used to have quite a few issues with our ethernet, just because of the way the cables have been run.

But around 8 months ago we had a company come in and change everything to fibre optics, and all our problems went away, until now.

There are 8 1788-ENBT bridges on the network, 6 panelviews and 10 'nodes' with Flex I/O Modules so I wouldn't say the network is that busy

Mark

you might try to unplug the e-net and see if you get the same results I:E flasineg led (if this will not cause any saftey or process issues!!!!) you may just have a loose connection due to vibration ect.. agian I am no expert on e-net also you could force output off to sov. to test the response of the piston

I have unplugged the ethernet and that concludes in different LED's flashing and again after reading the manual, the LED's which do flash when disconnected are the correct ones.

I have changed the ethernet cable and still it happens, the only thing I haven't done is change the Flex I/O adapter, mainly because I haven't got a spare.

Mark

are the sov 24v? could be a bad one or short in wiring if the same power supplies the card and the sov. output.causing the voltage to drop. but i dont know much about ab flex i/o. just seems to me assuming no program changes this has to be a hardware/wiring/noise issue. but you know assuming does

Thanks for all the suggestions.

The Flex Adapter has its own power source, as do the 2 input cards and the output card.

I am starting to think I need to bite the bullet and get a new adapter and if it turns out its not that, then at least we have a spare.

Mark

when you say 20 min or so is this on a regular basis?

It happens quite regularly, between 20 minutes and an hour.

anything in your process that happens at the same interval? a spare would be nice, but could you swap thise mod with another to test? it would be nice if someone who knew more about the flex i/o was awake

if i new more about process and the saftey of doing so i might sugest going to the module properties and in the fault idle options selecting hold last outputs or apply safe state on comm fault. this might could discover if you have a field wiring issue. also it might be a more desireable state to leave the outputs on when this happens. but you would have to consider the whole process and the !!!!!potentil Hazards to this action!!!!!!!

I used the fault idle option and set it to hold the last outputs, and this is working fine, and should work fine until my new modules come tomorrow.

Mark

I had a similar problem and it turned out to be bent pins on the bus connection between the problem module and the one before it. Someone ham-handed the things together.

MarkNightingale said:

BCE123,

This is a totally new problem. We used to have quite a few issues with our ethernet, just because of the way the cables have been run.

But around 8 months ago we had a company come in and change everything to fibre optics, and all our problems went away, until now.

There are 8 1788-ENBT bridges on the network, 6 panelviews and 10 'nodes' with Flex I/O Modules so I wouldn't say the network is that busy

Mark

Click to expand...

Without seeing how many controllers and I/O connections and update rates, I would say it looks pretty busy to me.

Can you monitor the status of the ENBT from Logix5000? Tell us how it is configured? How many switch hops? IGMP Snooping turned on?

Just WAGS, it might be a bent pin or bad module, but a timeout error would have me verifying the comms and stats first.