1794-IF8IH/A Strange analog noise

So, pulled my hair out for a while trying to figure out a problem with fluctuating analog signals from multiple 1794-IF8IH modules.

I found something interesting that I thought I would share and see if anyone else has noticed this, AB support didn't seem to know anything about it.

Long story short, RTD's, flowmeters, pressure transmitters all fluctuating +- ~200-800 raw units at 4000-20000 scaling (how can you comfortably calibrate something like that?). Checked electrical grounding, shields, power supplies etc. Couldn't find any problems. Put a o-scope on one of the channels and saw a patch of noise every second or so ~800mv p-p. Disabled hart in the config word and the noise and fluctuations disappeared...

Thanks AB! I guess if you are going to have hart enabled then you need to use the hart.pv data and not the analog???

Heres a trend of the raw data--

I have used this card multiple times and never had an issue. Check your 24VDC power supply and measure its current. It is possible the power supply is overloaded and 24vdc fluctuate a bit as perhaps hart communication require additional power.

Power supply is not overloaded. I isolated one rack to a separate supply with no change. I verified that this issue exists at another one of our plants. I'm going to set up a proper test on my bench on Monday, but I'm pretty sure it is an issue with the module.

Do you use the analog 4-20 data or the hart.pv data?

I bought the same Analog Flex modules, 4 in total, and like you, I encounter the same strange noise fluctuation: that is the reason I get your post on google. I tried, like you, to check any existing CEM noise with scope meter, and nothing more than the HART FM signal itself, but very distorted.In normal condition, the hart signal is a nice sinus form. It's like if the Hart signal make interference with the ADC. I don't call yet Rockwell but after all my attempt to understand what is wrong with these modules, I will call support. I ever try to disable Hart on chanel, but it was because I add on the loop an Emerson Thum Wireless who talk as a Hart Master with the transmitter : The signal noise fluctuating always present . This confort me that the HART signal disturb the ADC inside module. I Will try to disable the hart on one channel and report you if the fluctuation stop...

VERY interesting. Most AI's have a low pass filter for 4-20mA DC that filters out the 2200 Hz and whatever the other FSK frequency is that form the logic one and logic zero for HART comm.

It appears that yes, if you have the HART enabled card, you should be using the HART data, not the analog data.

Please keep us posted on your findings.

Ok , the problem seem to be solved after a quick read on the rockwell automation knowledge base :

For each channel , there is a cutoff input filter how can be configured through configuration word :

B0 B1 B2
0 0 0 470 Hz -> ( factory Value)
1 0 0 62 Hz
0 1 0 19.6 Hz
1 1 0 16.7 Hz
0 0 1 10 Hz
1 0 1 4.17 Hz
1 1 0 n/a
1 1 1 n/a

(See documentation http://literature.rockwellautomation.com/idc/groups/literature/documents/um/1794-um065_-en-p.pdf )

The 4.17 Hz filter with Hart enabled channel suppress all fluctuation. There is residual variation , but less than 10 µA, very acceptable for accuracy measurement.