One Input Two Different Values

[ASF]

Apologies if this has already been answered, but given what I've read so far, the most obvious thing to check (for me) is that both analog input cards have been configured the same in Logix 5000 (or Studio 5000). For example, one card could be configured to read 0V as 0 and 10V as 10000, while the other might be configured to read 0V as 0 and 10V as 12000. This would give you an error like this.

Before I got to that stage, I would (as others have suggested) put a high-precision voltmeter on both analog inputs and compare the readings. There is no use any of us, the OP included, taking shots in the dark when some simple tests could provide a lot more useful information.

Also, as I asked previously - has this only started happening recently, or has this problem always existed, since the transmitter was first installed? That answer will also narrow the field of possibilities a lot more.

 

[goodoboy]

Apologies if this has already been answered, but given what I've read so far, the most obvious thing to check (for me) is that both analog input cards have been configured the same in Logix 5000 (or Studio 5000). For example, one card could be configured to read 0V as 0 and 10V as 10000, while the other might be configured to read 0V as 0 and 10V as 12000. This would give you an error like this.

Before I got to that stage, I would (as others have suggested) put a high-precision voltmeter on both analog inputs and compare the readings. There is no use any of us, the OP included, taking shots in the dark when some simple tests could provide a lot more useful information.

Also, as I asked previously - has this only started happening recently, or has this problem always existed, since the transmitter was first installed? That answer will also narrow the field of possibilities a lot more.

Thanks,

Yes, we check the configure on both cards. We also got desperate and just replaced both AI cards with new cards.

We have not attached a voltmeter on both AI cards to make sure the voltages are the same.

I do believe the flow oscillates so quickly that one AI card reading is different from the other due to the sampling time of each AI card and PLC scan rate.

Thanks,

 

[ASF]

I do believe the flow oscillates so quickly that one AI card reading is different from the other due to the sampling time of each AI card and PLC scan rate.

To test this, I would connect a 4-20mA source in place of your resistor, and check the PLC reading from both cards at 4mA, 8mA, 12mA, 16mA and 20mA. If it truly is just a sampling time issue, then you should have more or less exactly the same results on both PLC cards with a stable source.

And again - you should test with a voltmeter, and answer the question about whether this is a new or old problem. We can only do so much if you don't do the legwork at the other end.

 

[DSMR]

Out of curiosity, why not wire both analog inputs in series as current inputs? Voltage inputs can always be a bit flakey.

If wired in series with no external resistor, the inputs would have to be the same as long as both cards were calibrated the same.

Next question is how safe this. Either way you cut it, this is still a single point of failure. This would not meet SIL3 as a voltage of 0 would read as a valid input. For SIL3 you should be using a safety analog input card like a 1734-IE4S (not sure what SIL you are shooting for).

 

[Lynx777]

I've worked on a problem a little similar to this before. It did indeed have to do with the sampling time difference in the PLC. It also had to do with a difference in the filtering settings on each channel. Ch-A had no filter and Ch-B had a filter enabled.

Our best results were obtained by enabling a filter on both channels.

 

[jawolthuis]

If the problem is in fact a difference in scan time, then couldn't you set the program to capture the values from both cards at the same time at the beginning of the scan, and then use those "cached" values within the program? This would preserve the redundancy, but would ensure that both readings were taken at the same time.
(I have never used RSL5000 so don't understand the details of how to go about doing this).

 

[goodoboy]

I've worked on a problem a little similar to this before. It did indeed have to do with the sampling time difference in the PLC. It also had to do with a difference in the filtering settings on each channel. Ch-A had no filter and Ch-B had a filter enabled.

Our best results were obtained by enabling a filter on both channels.

Thanks for response,

Where do I check the filter and sample time of the channels?

 

[goodoboy]

To test this, I would connect a 4-20mA source in place of your resistor, and check the PLC reading from both cards at 4mA, 8mA, 12mA, 16mA and 20mA. If it truly is just a sampling time issue, then you should have more or less exactly the same results on both PLC cards with a stable source.

And again - you should test with a voltmeter, and answer the question about whether this is a new or old problem. We can only do so much if you don't do the legwork at the other end.

Thank you,

I really can not say if new or old problem because we just started capture difference alarms.

 

[danw]

Where do I check the filter and sample time of the channels?

It's right in front of you. Digital Filter

You need to read the information that's available to you.