HART Module for the Do More

[Henryp]

Hi,
How hard would it be to have Host make a HART communication module. Something like a 4 or 8 channel card that could be multi-dropped.

[BobO]

It's just a question of market demand and available resources. We are definitely entering a phase where we want to do more communications modules, but the number of targets is high and the resources are low.

[Controls Guy]

This raises an interesting question -- if there would be third parties willing to make BRX modules.

AB has long had tons of third parties that supply custom comms modules, modules for vertical applications like welders, and so on for PLCx, SLC, and Control and Compact Logix.  Horner made modules for Generous Electric.  Back in the TI5x0, 5x5, and 405 days, CTI and Facts made premium A/D I/O modules like higher point count, better analog, etc, rather than specialized vertical stuff.  CTC had that weird Access 4000 thing.

Do you think there would be third parties (other than Facts, who uses the same sales channel) who might want to make BRX modules?

[ATU]

If you were designing a new Analog Card, couldn't that be incorporated into the hardware?

[BobO]

I have no idea how you would implement HART without using specialized transceiver chips. That would go well beyond the requirements of a basic analog module, both in cost and complexity.

[BobO]

This raises an interesting question -- if there would be third parties willing to make BRX modules.

AB has long had tons of third parties that supply custom comms modules, modules for vertical applications like welders, and so on for PLCx, SLC, and Control and Compact Logix.  Horner made modules for Generous Electric.  Back in the TI5x0, 5x5, and 405 days, CTI and Facts made premium A/D I/O modules like higher point count, better analog, etc, rather than specialized vertical stuff.  CTC had that weird Access 4000 thing.

Do you think there would be third parties (other than Facts, who uses the same sales channel) who might want to make BRX modules?

I don't guess we are opposed to it, but we haven't had requests. My attitude is that if there is enough market to justify the effort, I'm willing to build the module (as resources permit), and if not, I wouldn't ask anyone else to. I honestly don't have much knowledge of HART. It's a name we hear from time to time, but it hasn't been often enough to give me a feel for the opportunity. We genuinely want to start doing more modules like this though, and if those efforts are rewarded with sales, we'll do more still.

[Controls Guy]

My attitude is that if there is enough market to justify the effort, I'm willing to build the module (as resources permit), and if not, I wouldn't ask anyone else to.

Well, that's just it.  It sounded as if there might be more worthwhile opportunities than there are resources to exploit them.  The third-party system occurred to me as an alternate way for resources to get to the opportunities, so resource leveling.

Second, set aside premium versions of generic A&D I/O. and let's talk about the vertical specialized stuff. "Market" may be defined differently by an OEM selling a specialized solution than it would for Host.  I once installed a SLC module that controlled a welding process, and it wasn't just the module, the solution comprised all these various boards, wiring, proprietary HMI and so on.  Everything down to the gate drive for the power semiconductors, and the whole shebang cost about a brazilian dollars.  So I think they're going to define market differently that the PLC maker would.

I honestly don't have much knowledge of HART. It's a name we hear from time to time, but it hasn't been often enough to give me a feel for the opportunity. We genuinely want to start doing more modules like this though, and if those efforts are rewarded with sales, we'll do more still.

Well, my post wasn't specifically about HART, but that was just a case in point that made me think about the third party ecosystem that a lot of PLC makers have.

[BobO]

Even if a 3rd party does the module, there is still a very significant hit to us to support in firmware and software. There is also a big uptick in systems engineering to make sure that stuff isn't getting broken. And all of this assumes that we had system documentation of a quality required to support 3rd party development, which we don't.

Now if someone said, "Hey Host, I wanna standardize on BRX and will buy it by the metric ton, but I want to do this custom module containing critical IP that only I can possess", they have my full attention.

In my experience, the world is full of guys that want to make modules, but very few will pull in significant volume. I would love to be wrong.

[Controls Guy]

Even if a 3rd party does the module, there is still a very significant hit to us to support in firmware and software. There is also a big uptick in systems engineering to make sure that stuff isn't getting broken. And all of this assumes that we had system documentation of a quality required to support 3rd party development, which we don't.

OK, so it might be that having a third-party ecosystem is something more suited to large PLC OEMs.  Come to think of it, all the examples I mentioned (AB, GE, TI) are/were all good sized companies, so they could afford the manpower to do the support.

Oddly, I think Siemens is the odd man out on this issue, for large PLC builders.  They build a big selection, I assume in house, at least that's the way they're labeled, and I'm not aware of the existence of any third party modules, whether premium generic (Siemens has three or more levels of how deluxe you want to buy a 4-20 AI module for example) or vertical specialized.  Not sure about other big boys like Omron or Mitsubishi.

[ATU]

Moore Industries makes a Modbus to HART converter.

[BobO]

Some PLC architectures can easily be reverse engineered. Some of the 3rd parties you mentioned were not supported by the primary manufacturer. BRX would be considerably harder, and with basically no CPU support they would be limited to simple I/O.

I'm always open, but the opportunity to Host would need to be significant.

[ATU]

It seems like other controller manufacturers have not made their own HART module, but instead sell this stand alone converter by Moore Ind.  They are pricey too, around $1000 each.