URGENT - High Speed Counters not working all of a sudden ML1400/Logix 500

[TimD]

FLOW?

So I removed all pull-up resistors and flow element leads. All resistors out of circuit read approximately 2.2k ohms

I hooked one meter back up, and I am reading flow!

It must be one or more bad sensors pulling the rest down.

Process of elimination time. I'll keep you all posted!

 

[mordred]

I would also strip and reterminate all connection points if all the sensers are sharing the same neutral or common start there

 

[TimD]

update

so, one flow meter appears to work, but after disconnecting it, the HSC accumulator keeps going, even if the meter is stopped. Worse, when I hook up two flowmeters, about 4 accumulators start randomly counting as if I had pulses on totally disconnected inputs...

Getting somewhere, but not very fast...

 

[Gil47]

Do you have a good earth link onto your ML 1400 earth terminal.

You mentioned your filter was at fastest setting of 5us try taking it to a higher number say 50us in case you are picking up a high frequency ripple on your power supply.

 

[TimD]

yes, tried the filter.

I am beginning to think my flow sensors are shot. How most of them got that way, I don't know. I am going to order a new sensor and give it a try.

Thanks all for staying tuned. hopefully I can resolve this bugger.

 

[geniusintraining]

What kind of meter/sensor are they?

I had something like this and 3 failed in one day, they were paddle type and we were measuring water, the problem was we were filtering the water a lot (DI/Carbon/IR/micron)

The paddle on the meters were being lubricated by the particles in the water, so when we cleaned the water they failed

This took about 6 months for them to fail

Just a thought...

 

[milldrone]

so, one flow meter appears to work, but after disconnecting it, the HSC accumulator keeps going, even if the meter is stopped.Worse, when I hook up two flowmeters, about 4 accumulators start randomly counting as if I had pulses on totally disconnected inputs...

Getting somewhere, but not very fast...

I'm curious about the disconnecting of the flowmeters in the quote above. Did you do this at the PLC terminals, or did you do it out in the field so that there was still wiring and possibly conduit still connected to the PLC.

 

[TimD]

I'm curious about the disconnecting of the flowmeters in the quote above. Did you do this at the PLC terminals, or did you do it out in the field so that there was still wiring and possibly conduit still connected to the PLC.

Wires were still on PLC going to terminal blocks.

Random HSC pulses from non-connected devices are removed whrn pull-up resistor is installed.

I have a clue... I have two 24VDC power supplies. One powers high-current DC pumps at 24VDC, the other powers all the instrument loops, including the flowmeters. If I take a voltage measurement from panel ground to 24v+ on my pump supply, my meter is solid at 24V... If I do the same thing to my I/O supply, voltage starts at 24 and quickly falls. Sometimes, I only measure 8v...

Now, it gets trickier... all my I/O is working, except for these flowmeters. The 24v supply in question outputs 24v measured direct off the supply using 24v +/- terminals, so this panel ground... gets me thinking. I switched supplies just in case, same effect. Any electrical whizes have any ideas on this one? I am not over-drawing current.

Where is my panel grounded to? Not earth, at least not nearby. It is grounded to a breaker panel located about 100' away through conduit of course.

Please remember, these meters worked flawlessly for 8 months...

I am almost positive (no pun intended) that it is an electrical issue. I have a new flow sensor cable coming tomorrow. This will QA my sensor question. I;'m expecting a new meter will not work.

Thanks All for staying tuned!

 

[TimD]

A ML1400 has an IQ series input, which can be either PNP or NPN, depending on whether the common for that group is Negative or Positive.
In my opinion, using NPN on that group would be a better than a pull up resistor.

Gil47,

Please tell me, if still following this, if the 1400 CPU has the IQ inputs. All I see is a group of 1771-series add-on cards for the Micro. I don't think I want to buy a card for something that used to work

I am digging through the hardware manual. Thanks!

 

[Mickey]

I believe this is what Gil47 was talking about. See pictures below.

 

[TimD]

I believe this is what Gil47 was talking about. See pictures below.

Thanks Mickey! Very much from the manual! I tried doing this early this AM, but will give it one more try, I tried isolating one common input group but wasn't thinking straight.

I've been reading up on Giant MagnetoResistive Devices as of late, and found some useful articles about pull up resistance and pulse-width issues...

Trying to see if/why a voltage spike or sag could invalidate the inputs completely. I still suspect a grounding/power change caused all of this.

 

[Gil47]

By IQ inputs I was meaning that the common can be either positive or negative in its IO groups
The first 4 Inputs can have a positive common while all remaining inputs could be negative
This allows you to mix and match your PNP and NPN sensors to the PLC.

 

[shooter]

You connect the zero voltage to ground, that is wrong.
the loop for the power should be +24 volts and - 24 volts. or connect the minus to ground hard.
and the shield to ground obvious.
2k2 at 24 volt means 10 mA times 24 volt is 250 mW and that is max of resistor. so check the resistor.
is the sensor in hot environment?
then the transistor inside can be hot.

 

[milldrone]

Wires were still on PLC going to terminal blocks.

Thank you for answering this. You have just given us some power supply info. And it sounds like one power supply is "common" to all flow meters. And I'm guessing that there are other "common" (to the flow meters) items, like conduit and 24V+ and 24V- wiring. These items that are "common" to all the affected flow meters are to be considered suspect.

Random HSC pulses from non-connected devices are removed whrn pull-up resistor is installed.

This sounds like noise. And may or may not be a clue.

I have a clue... If I take a voltage measurement from panel ground to 24v+ on my pump supply, my meter is solid at 24V... If I do the same thing to my I/O supply, voltage starts at 24 and quickly falls. Sometimes, I only measure 8v...

I'm making a presumption here (sometimes a bad thing) When you are saying "ground" are you meaning 24V- ? If your power supply is 24 V and you have less than say 22 volts. Whoa! Stop right here! This needs to be addressed before anything else can be done.

Note: If my interpretation of what you meant by "ground" is incorrect, then my suspicions are all wrong.

I am not over-drawing current.

Is this statement driven by your knowledge of the fact that a fuse did not "blow". If it is. My guess is the power supply is a "switcher" and is self protected. It will never "blow" that fuse.

Presuming what I'm guessing is the Problem is indeed true. then there is something that is drawing down the I/O power supply. (Including the power supply itself)

 

[Ron Beaufort]

from post #38 ...

I have two 24VDC power supplies. One powers high-current DC pumps at 24VDC, the other powers all the instrument loops, including the flowmeters. If I take a voltage measurement from panel ground to 24v+ on my pump supply, my meter is solid at 24V...

so far, so good ...

If I do the same thing to my I/O supply, voltage starts at 24 and quickly falls. Sometimes, I only measure 8v...

now that sounds flakey from where I sit ...

does this mean that the VOLTMETER is "loading" the power supply circuit enough to pull down the voltage? ... if it is, then you need a new power supply ... but I'm guessing that the "load" of the meter is probably not causing this - and that the power supply is OK ...

then the next most likely cause for the "8 volt power supply drop" effect that you reported would be that the power supply is "floating" – and the meter connection is merely providing a path to "ground" ... in other words, it doesn't sound like the power supply is tied to a firm "ground" potential ... in most cases that's a recipe for "noise" and "fluctuation" from the input signals ...