Sign in to follow this  
Followers 0
jd_hartzell

HMI VS PC

12 posts in this topic

Guys, some one help me out here. I have been designing all my machines with Panelview Plus HMI's we got a new guy at work and he is wanting us to convert all of our HMI's to Industrial PC's with WOnderware on them. the industrial PC's are cheaper in price but you have to do the operating system virus protection plus the Wonderware. what are the pro's and con's? JD

Share this post


Link to post
Share on other sites
PC: flexibility with more complexity HMI: specialized with more dependability I work with the PanelViewPlus, so I feel a bias (even it isn't actually there). However, each application is different. I know I can get a PanelViewPlus running out of the box in less than 10 minutes. But that would not help me if I need the terminal to also run a CAD design (proprietary) software program. I would consider operator needs/use, environment factors, regulatory certifications needed, MTTR, end user support, total cost of ownership (engineering-maintenance), flexibility needs, etc. Off the top of my head, here are some things I can think of: Dedicated HMI terminal (ex. PanelView Plus) -no moving parts; related to wear/tear dust, operating shock/vibration, etc -power failure tolerant; designed to recover from loss of power -does not involve IT (depends on facility/union, but in some cases a PC requires IT to be involved whereas a dedicated HMI would not) -less prone to viruses, malware, or operator intervention (not impossible, but very unlikely a person codes a program to run on a dedicated HMI) -does not offer great flexibility to use other programs (depends on HMI, ex. internet explorer, media players, control code programs, etc.) -ready to go out of the box; runtime software already installed (except if you want to upgrade to different firmware version) PC -standard platform with great flexibility to install other software programs, or hardware (ie. anything for a Windows PC) -requires a UPS backup to minimize failures after power loss -may have moving parts, some computers are available without (ex. solid state drives, no fans) -diagnostic, troubleshooting utilities for the h/w or OS are generally better and more available (well developed for the consumer PC market) -does not require special training/knowledge for setup/maintenance;more familiar platform for setup/mainenance (IT trained staff, relates to consumer PC knowledge)

Share this post


Link to post
Share on other sites
Great information I'd like to hear from more of you on this subject. I'm sure more people have opinions on this subject.,

Share this post


Link to post
Share on other sites
If I get to choose when doing a project, I use a Panel PC with either Windows CE or XP Embedded. The Panel PCs are generally lower cost than a dedicated HMI and you are not bound by the limitations of the off the shelf HMI. Not sure I agree about the HMI being faster to develop out of the box. I can have a Panel PC running an HMI in under 5 minutes: CE doesn't have the flexibility of XP, but is virtually invulnerable to viruses. And that is generally only a concern if its on a network outside of the machine.

Share this post


Link to post
Share on other sites
Hi, I had the pleasure of creating DRPs for every kind of HMI we used in a packaging plant and with the mix of HMIs there were, it was a nightmare. The dedicated HMI terminals do have the advantages listed already, but the lack of flexibility ruins it for me. If one fails you must find a one to one replacement in spares in order to do a rapid recovery. This is fine if you have spares for every kind, ideally only one kind. If not, you must convert the project to the applicable platform, which can get messy. This is where the industrial PC blows an HMI terminal out the water for me. You can properly set up an IPC to join your domain, be subject to the same security policies as all PC on your manufacturing domain or VLAN, have the same virus protection and set it up for auto logon and program launches. You can also set it up to run your PLC editing software to faultfind at the machine itself. The biggest advantage is that you may use your stock standard PC skills to recover the thing at 3 in the morning. It is easy to identify what failed, be it harddrive, motherboard, network adapter etc. If worst comes, grab your PC out your office, install the HMI viewing software (e.g. InTouch) to get you through the night. If an HMI terminal fails, it is like a big solid unit. No spare, no luck. The only thing that you have to worry about with an IPC is that you have easily accessable version of the appropriate, latest backup of your app. I know that goes without saying, but you'll surprised how many people don't believe that. Train your guys to have basic PC fault finding skills and the ability to set up an HMI from scratch including the DAS and you are fine. As I said, you can use your personal laptop to get you going again. Hope this helps.

Share this post


Link to post
Share on other sites
No One has mentioned anything about scale and sizing of the Enterperise so I will. The HMI works nice give one point of operation dependable, hi-speed access to the PLC Data and Process Cotnrol. The IPC works okay for one point of operation dependable, medium-speed access to the PLC Data and Process Cotnrol. The IPC can also be deployed multiple times at multiple locations and provide varying degrees of control based on an access schema. I have an IPC I am thinking of which controls a section of a machine, but can also display operational status for 45 other machines if I need to check on them. Try connecting 50 Panelviews to a single ControlLogix. It will work but you won't like the performance. Now try an IPC Tagserver serving PC Data up to 50 IPC stations and the performance will be more than acceptable. All depends on what you need and where you are going. IPC with Wondereware also opens up thru use of SQL Databases the ability to log status of tags over weeks and months and then trend them when needed. A neat troubleshooting aide for intermittent stuff {think OE}

Share this post


Link to post
Share on other sites
I agree with Bob on the HMI being best for single point use. My only problem with using an IPC is that you usually have to invest in some type of cabinet (sometimes climate conditioned). You also have a much better selection for your software to display your graphics (RSView32, Wonderware and others), as well as all the added features of just having a PC to use. We use both types of system, usually determined by customer desire, cost and application (single vs. multiple points of access).

Share this post


Link to post
Share on other sites
How about cost? I just recently tried to get an upgrade from Wonderwere to go from a Windows NT system to a windows XP system. Keeping in mind, we already the licence, but that didn't matter. We needed to buy an upgrade along with an upgraded licence. That cost was $8956.88. So we converted it over to AB for a lot less. Just a thought.

Share this post


Link to post
Share on other sites
For one of the reasons they call it WONDERWARE SEE ABOVE :) It should be noted though that Wonderware 1 programs which ran on Windows 3.1 can be ported to Wonderware 10 and run on Windows 7 with a minimum of dificulty, albeit the cost is significant.

Share this post


Link to post
Share on other sites
In adddition to what's been stated, I'll broaden it a bit by suggesting non-standalone systems. It is incredibly nice to be able to edit the HMI in real time on the server and have the changes propagated across the plant in real time. There's a lot of up front costs in terms of the servers but once you get that much done, you can amortize it. The breakeven seems to be somewhere around 8-10 stations. Second, I've had a heck of a time armoring up stock PC's even industrialized versions to survive in the environments that I work in. Granted over time it's gotten a lot better but a plant environment is just not the place to be putting that kind of stuff other than in environmentally controlled control rooms. You CAN do it but it's difficult. Ok, and now for the broadening. What I've found to be extremely effective is thin clients. It does nothing towards costs (despite claims otherwise) but you can get fanless, diskless "PC's" that are effectively just terminals with mouse/keyboard/screen (including touch screens) but also you get 40-60 deg. C temperature tolerances, very cheap expendable hardware on the plant floor, and generally much better survivability regardless of the fact that it's PC hardware. Also, you can usually set them up to be field replaceable in minutes with a minimum of hardware. The actual hardware is of course the big fat server in an environmentally controlled area elsewhere in the plant. Again, the cutoff is around 8-10 operator stations before the thin client model makes sense. But in the long run, I feel that the "standalone PC" model has very limited use in the world anymore. As for the concerns about failures and such with this model, it's quite overblown. You need the network infrastructure anyways to support standalone PC's or server/client systems so there's effectively no difference in terms of reliability from that point of view. Where I currently work at though I'd still look at standalone PC's in the long run. We run networks of substations/starters in a mine. They are all connected wirelessly (most reliable communication in this instance). Wireless networking is very reliable relative to running hard cabling BUT it's not 100% reliable 24/7. Thin clients in this environment which rely on solid comm links would be a nightmare.

Share this post


Link to post
Share on other sites
Paul - I agree with you assessment, but would take it a step farther. We got our first batch of 80 Wyse thin clients about 6 years ago and have deployed newer ones since. It's true that they're not much cheaper initially, but your TCO (total cost of ownership) is substantially lower if you're already supporting a client/server model. The effective lifecycle of a thin client is several years longer than a desktop and you spend MUCH less time patching, upgrading, imaging, or reinstalling operating systems. Also, units die over time, but you tend to replace parts/units much more infrequently than desktops. I couldn't imagine thin clients over a wireless connection - what a NIGHTMARE. They can even be pesky over a reliable wired network at times. Edited by Nathan

Share this post


Link to post
Share on other sites
Most of the obvious ones have been covered. There's no "right" answer - either, or even a combination of both may be ideal for your setup. In general, for a small setup the Panels will be more robust. As you scale the size of your project and the number of terminals you hit a crossover point where you get more bang for your buck with a PC based solution, then eventually it becomes the only viable model (what Bob referenced when too many individual stations try to connect to the same PLC). Even with a handful of panels, you'll probably get tired of having to program each one individually. That said, it depends greatly on your application. PC based solutions are gaining capability faster than canned panels. Industrial PCs from many vendors continue to improve at lower prices. General purpose industrial software is getting better, cheaper, and more distributed. It's like comparing the trend in arcade machines or even console games to PC games. There's just more "oomph" behind the PC segment. Sometimes it's just cool to play Pac Man at the arcade, but it's nothing that your PC can't blow away. Edited by Nathan

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0