EPIC Load Cell Input

Is there a timeline on when or if load cell input modules will become available for the EPIC platform?

I checked in with the Engineering group and yup, its well and truly on the roadmap.

Sorry we cant give you an ETA just yet (its a bit too soon), but we are wondering how many channels of how many wire load cells are you looking for in each module?

I don’t have an immediate need but we have many batching systems that use AILC modules. These are typically multi-scale (most 2 channel, some 4 or more), 6-wire systems. I think a dual-channel 6-wire module would be very useful to have.
You’ll always want 6-wires as most of these will actually connect to a multi-cell scale. The scale can be remote so you’ll want sense wires in the cable coming back to the module.

Have you used them with SNAP-PAC in the past?

We have many SNAP-PAC systems with AILC modules. I know we can interface a SNAP-PAC with the EPIC, I was just wondering if the EPIC platform was going to get it’s own version of the AILC.

Thanks for getting back to us. It’s really helpful to know whats worked in the past and what our customers are looking for.

I use a MV/V to 4-20 mA Converter
Like this https://oceancontrols.com.au/AXS-500.html

1 Like

For our application, we have a scale with 4 load cells that lead to a junction box, and from there, the 6 wires + 1 shield go to a Mettler Toledo scale display. From the Mettler Toledo, we pass the scale reading to the EPIC via RS-232.

The forthcoming EPIC module would allow us to skip the Mettler Toledo scale and the RS-232 “fun” (/s). Not sure how the AILC module works, but I presume it’s straightforward to get lbs or kg once we have the wires hooked up.

One thing that I am hoping is possible is to detect when something goes bad in the load cell(s). Due to our process, the scale can sometimes get abused and then when the feeder system (which uses the scale as part of its program) seems out of whack, we check the scale by verifying with a test weight, and then make the necessary fix & re-calibrate. I am wondering if one could inspect or monitor the EXC, SEN and SIG wires, if one could “see” when a load cell goes bad. Any idea?

The AILC provides its own excitation and sense terminals as well as the signal terminals so it’s very much like a standard scale in that regard. You do have provide it 24 VDC.

After you configure the mv/V in the AILC, it calculates the percent of full load on the load cell based on the load cell output. For example, if a cell is 3 mV/V that means the load cell output is 3 mV for every excitation volt. The AILC excitation is 10 V so for this cell, 30 mV = full load (capacity). The AILC reports 0 for 0 mV and 100 for 30 mV. This is a theoretical calibration. In reality, load cell outputs aren’t that exact, though pretty close. Also, cell outputs can change over time as they age. That’s why other scale terminals have a calibration procedure that uses actual weights. They match the weight on the scale with the load cell signal. Do this for two data points, usually zero weight and a span weight, and you have a linear formula.

What I’ve done in our systems is to just use the load cell signal from the AILC (-100 to 100) and created my own calibration procedure in the software and that works pretty well.

Something else you need to look out for is noise, either electrical or mechanical. The AILC provides its own filter channel that I think is primarily for filtering out electrical noise but I may be wrong about that. I also build into my applications an averaging function that averages the last X readings from the AILC. That also helps reduce noise.

When a cell goes bad, the signal usually goes out of range. When that happens, the value provided by the AILC drops to its lowest value, a large negative number. In that case, either a cell is bad or there is an open circuit somewhere.

A bad cell may not go out of range but pretty close. For example, if you have a 1000 lb scale that is empty but the cell output is very high like 25 mV, then you know that something isn’t right with that scale. The output is not what you expect it to be.

A junction box may muddy the waters a bit though. One bad cell in a group of cells does affect the output from the junction box but it may not be as noticeable. Again, you’ll have to use a little common sense to see if the output is what you would expect. If it’s a bit too high or low, then you’ll have to check each cell manually with a voltage meter.

Hope that helps.

This is just an FYI. I ran into this issue this week.

We had an application, one of our standard applications, that involves two scales. We installed two SNAP-AILC modules to handle each scale. However, we found out onsite that one of the scales had six load cells instead of the standard four. The SNAP-AILC module can only support up to four load cells. After that, the excitation voltage gets loaded down.

So, keep that in mind when using the SNAP-AILC modules. We managed to get around our issue by installing a third AILC module. We then connected three of the cells to one module and the other three to the new module. The signals were added together in our software. Turned out to be an easy fix programmatically but it was a hassle getting a new module shipped in overnight and running a third cable through the existing conduit.

Many scale instruments from companies like Rice Lake support up to eight load cells. They need that because many large systems like truck scales require that many cells. Perhaps a new EPIC load cell module should do that as well.


Any new updates when the Epic load cell input module will be out?

Not as yet. Will post here when we get some more information.

Mean time you can try using Epic thermocouple input module and use a precision power supply to power the wheatstone bridge.
typically analog load cell are 2 or 3 mv per each V of the power supply usually 5 or 12 Vdc

I was just wondering if Arun perhaps had a chance to forward our technical specification to R&D for a 16 x loadcell input module? We requested this in 2018 just after our visit to Temecula when the EPIC was launched


I did indeed. I’ll see if I can get an update on the status and get back to you.



Hennie - I have an update from Engineering, and indeed it will be in the next batch of groov EPIC modules. I can’t give you and ETA yet, but will keep you posted.

Thank You for your continued patience.

I use a Wheatstone bridge converter with 4-20mA output and have the input into the iMa 24 card. Plus the converter takes some of the wild bounces out of the Loadcell.

1 Like