Host Engineering Forum
General Category => Do-more CPUs and Do-more Designer Software => Topic started by: Henryp on March 08, 2016, 12:42:14 PM
-
A Do More and HMI all in one would be nice.
-
We use one of them, HornerAB I think, and it's okay but my downside is expandability. Where I work people come up with new ideas for this machine, and I'm reminding them that the PLC is full and we can't add anything more.
I'd rather pay a little more and separate the two units. Plus the way our operators man handle equipment, a broken HMI has happened... Repeatedly...
-
Unless it's primarily an HMI and logic application with maybe a need for less than 10 I/O or something, I've never got the idea behind these. Who wants to run 100 wires to the door?
@deep6ixed: Many of Horner's units will talk to remote I/O (some are even remote-only), so you can put the I/O on the backplane, and expand it, if you like Horner's programming paradigm. The ones I'm thinking of even talk Modbus/TCP, so you can use generic distributed I/O like Terminator or ProtosX, or 50 other distributed I/O systems.
-
I love the idea of Host eventually getting into HMI, but that might be one of those "stay in your own sandbox" things.
-
If you did it, I know it wouldn't be a "Me too" product -- you guys would rock it! :)
-
Yea, an HMI with a built in Do-More engine and remote Ethernet IO... That could be great.
Host makes great products, and I am sure that if you went in that direction it would be an amazing combo!
-
I'd still like if you're going to do an HMI, to at least offer a pure HMI version with no logic engine (for those rare times the customer forces me to use a lesser PLC, plus I like to modularize my apps anyway).
-
I'd still like if you're going to do an HMI, to at least offer a pure HMI version with no logic engine (for those rare times the customer forces me to use a lesser PLC, plus I like to modularize my apps anyway).
So much comm work to become truly general purpose. Might be a reasonable second phase though, once the product was making enough money to justify the investment. And I suspect the politics of embedding Do-more would be far more manageable than another general purpose HMI.
I do seriously love the idea though and would think the opportunity is quite large.
-
The upfront comms work isn't too bad, so long as you employ the 80/20 rule.
Initially, just implement Modbus RTU, Modbus/TCP, and DMP or whatever Do-More protocol is called, and you'll have covered about 90% of the available market. Add other drivers as you can demonstrate demand for them. Seriously, EVERYTHING talks Modbus.
-
The upfront comms work isn't too bad, so long as you employ the 80/20 rule.
Initially, just implement Modbus RTU, Modbus/TCP, and DMP or whatever Do-More protocol is called, and you'll have covered about 90% of the available market. Add other drivers as you can demonstrate demand for them. Seriously, EVERYTHING talks Modbus.
I can see that.
Probably would still embed the engine though, even if we offered without I/O. HMIs really benefit from scripting, and I just happen to already have a nice way to do that.
-
I can see that.
Probably would still embed the engine though, even if we offered without I/O. HMIs really benefit from scripting, and I just happen to already have a nice way to do that.
That you do! ;D I was going to mention that as long as the Ethernet port is accessible for HMI comms with a remote processor, it doesn't hurt anything if there's a DM engine in there, and as you note, a lot of times it's going to be quite valuable (especially once DM does ST). I've done that (used a combo HMI/PLC as a pure HMI) and it works fine.
-
I love the idea of Host eventually getting into HMI, but that might be one of those "stay in your own sandbox" things.
I have mentioned this to some friends before. I would love to see a Host Eng HMI.
-
I have mentioned this to some friends before. I would love to see a Host Eng HMI.
Well, from your new lofty perch amongst the shakers and movers in the automation world, you can begin softening the target for Host's eventual, and dare I say inevitable, entry into the HMI market. ;)
And yes...that was a joke.
Mostly.
-
I have mentioned this to some friends before. I would love to see a Host Eng HMI.
Well, from your new lofty perch amongst the shakers and movers in the automation world, you can begin softening the target for Host's eventual, and dare I say inevitable, entry into the HMI market. ;)
And yes...that was a joke.
Mostly.
Lofty Perch? You never fail to make me laugh, BobO. You're awesome!
-
Lofty perch? That's just a fish with am attitude!
-
I don't like the PLC HMI combo. Break one buy both, wires to the door are stupid.
-
Or it could be done similar to how unitronics does it. The back of the HMI is the PLC rack. You add in a cpu and your io cards.
Not that I am a fan of their hardware, but that style is neat.
-
The upfront comms work isn't too bad, so long as you employ the 80/20 rule.
Initially, just implement Modbus RTU, Modbus/TCP, and DMP or whatever Do-More protocol is called, and you'll have covered about 90% of the available market. Add other drivers as you can demonstrate demand for them. Seriously, EVERYTHING talks Modbus.
I stumbled on this again looking for Protos X with Do-more experiences.
But it also brings to mind another tech wondering about adding an H0-ECOM100 to a D0-06 to tie it in to a brand new CompactLogix with an EA9. Not EVERYTHING talks Modbus, especially in Rockwell land. And no EIP for the D0-06. But we could possibly use the EA9 as a gateway if push came to shove. But you know, that is the problem with HMIs, there is always a new generation just a couple of years away and what if the features change and screw up your scheme? I have one machine in mind that has had Total Control to EZ-Touch to EA7 to EA9. (Ooh, I just looked and it's been 21 years, wow.) Only a TI PLC and a D2-260 though.
I actually like the Do-more engine in an HMI idea, but no local I/O for me. If I were an OEM maybe, but otherwise it is too big of a pain. But I could sure see handling the smarts for ethernet linked components.
-
SLC's (as of about FRN 11 or so -- granted they were late to the party) and most (all?) Micros do Modbus, and I believe there are AOI's for Control/Compact Logii.
-
Modbus is everywhere nowadays, I even found a Cordova plugin for using Modbus in a mobile app. But you still have some manufacturers who have to sell their own special setup to force you into vendor lock in.
-
SLC's (as of about FRN 11 or so -- granted they were late to the party) and most (all?) Micros do Modbus, and I believe there are AOI's for Control/Compact Logii.
CompactLogix $1400 for ProSoft Modus TCP
ControlLogix $2000-$2300 for ProSoft Modus TCP
CompactLogix $250 for ProSoft Modus RTU (Lite)
ControlLogix $2000-$2300 for ProSoft Modus RTU
Somehow I'm not feeling Modbus fits my budget with Rockwell/ProSoft.
-
CompactLogix $1400 for ProSoft Modus TCP
ControlLogix $2000-$2300 for ProSoft Modus TCP
CompactLogix $250 for ProSoft Modus RTU (Lite)
ControlLogix $2000-$2300 for ProSoft Modus RTU
Somehow I'm not feeling Modbus fits my budget with Rockwell/ProSoft.
Yup, I've done that way too back in the day when it was the only way, and yes, I too thought it was spendy (though if you think about the amount of time it offsets it starts looking cheap), but....like I said SLC's do Modbus natively now (no $ required), Micros always have, and I believe Add On Instructions are available to make the 5000 stuff act as master or slave or both, and I think they're free.
So yeah, at this point Prosoft modules are starting to look spendy again.
-
A Digi One IAP can bridge a lot of stuff for $500. As far as I know, it will only support Control / Compact Logic via PLC/SLC emulation mode, and the implementation isn't perfect. (I can't get it to support B-files of longer than one word, N-file support is pretty good but a little kludgey for buttons, etc.), but it might work OK if you MSG from the PLC to the DOIAP.
I have an app at the moment with a SLC that needs to talk Modbus to a bunch of Danfoss drives. I'll use one of these methods and let you know how it works.
-
...I believe Add On Instructions are available to make the 5000 stuff act as master or slave or both, and I think they're free.
I'll throw in my $0.02 and say the AOIs for Modbus are not very stable, flexible, or reliable. They feel like they were developed to show that it could be done, but never fully fleshed out or tested. I started a project planning to use the Modbus/TCP AOIs and ended up installing a ProSoft card because I could never get it to work quite right. I really do wish Rockwell would circle back and get native Modbus support on Logix.
-
Good to know! Are the AOIs open or protected? If open, we could probably finish them (or just start from scratch and do our own).
-
As I recall, they are open, but fairly convoluted.