How do we communicate with aging PCs?

Replacing legacy HMI displays requires programming software.

By Control Design

I’ve been replacing legacy HMI screens that are many, many years past life expectancy, but they still work. The issue though is whether we still have access to a working programming terminal of the same vintage. Virtual PCs can be of some help, but we still have to deal with the communications from PC to device and some of the older technologies don't pass through a virtual PC very well.

The challenge has been to get a version of the application for these HMIs, so I could either convert it or rebuild it in a new platform. Without the programming software and application, I’m left with developing it from scratch. Anyone have experience with this or suggestions on how to address this?

Join the discussion and read the comments below

 

Main image courtesy of jesadaphorn at FreeDigitalPhotos.net

 

 

 

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

  • <p>The inability to access old configuration files is a challenge that will only grow tougher as the years pass. The following is an excerpt from one of my application engineers in response to your situation. "We’ve worked with a number of integrators who were trying to move an older (competitive) HMIs programming to a newer Red Lion HMI. In these cases, even if the customer successfully retrieved the old application file using old software and an old/virtual PC. The customer would still need to rebuild the project from scratch. Unfortunately, there is no magic bullet to this situation and it seems the reader already has a couple methods of accessing older configuration files. If they are able to access the base file and they plan to stay with the same manufacturer, there may be some easier upgrade paths in place. For example, Red Lion’s HMIs feature a smooth migration from Crimson 2.0 to Crimson 3.0 for use on our newer HMI platforms. </p>

    Reply

  • <p>I wish there was a simple answer to this but unfortunately there isn't. But there are some possibilities. I speak with our clients almost everyday that have had one of our legacy machines for 10+ years and now have a pending event to replace/repair. These legacy industrial PC's can have various hardware components that just aren't available anymore. That usually the easy part. Most of the time it is that legacy application. It ran in DOS or Win95 Win98, or WinXP. There are a lot of the situations where the installation disks are either long gone or in a format that isn't easy to port (floppy disk like the old 5 1/4 for the old guys like me or the 3 1/2). Or the installation media is long gone and the original company that built the application is out of business. </p> <p>There are some things that you can do. If you are fortunate enough to not be too far back you can try running the application on a newer (note not newest) O/S like Win 7 in the compatibility mode. There are settings that will allow you to adjust for things like screen resolution and number of colors. A roll of the dice at best but the first try for sure.</p> <p>You also can try loading older version of the O/S as a virtual machine and that can work at times. Missing DLL's or drivers will in a lot of those cases be the nemesis but I have had some success getting things to run. </p> <p>There are also some conversion programs out there that you can try. Old DOS programs sometimes be successfully converted (note Windows has not been built on DOS for sometime and the command prompt is not full DOS) </p> <p>There are things to keep in mind like was this a 16 bit program now trying to do 64 bit? Or did the application do thinks like use the timing of the CPU to make request and wait cycles? </p> <p>So how and application was built will have an impact of if you can get it to run in modern technology.</p> <p>One thing I do know for sure is this will get harder not easier as software and hardware progress and the break neck speed we have today. So I would look at every legacy machine you have and determine a plan well ahead of time. Planning for obsolescence unfortunately is a necessity. And you don't want to do this when things are down and the boss is over your shoulder.</p>

    Reply

  • <p>Engineers have been faced with replacing aging proprietary equipment, and most recently, aging HMI’s for many years. At one time such equipment manufacturers could have been the single “go-to” source for machine and process information and control, simply because of the belief that those companies providing this equipment would support the HMI devices and technology indefinitely. Unfortunately this has proven not to be the case for many reasons. In some cases the equipment manufacturers were purchased by a different company, and the legacy products were simply dropped, since they were no longer compatible with current devices or technology. In other cases, the HMI technology, whether the hardware or software parts, or both, were simply not available or compatible with existing platforms and operating systems anymore. Libraries that were used to compile older versions of programming software could have been abandoned or replaced because of unrepairable security flaws or unsupportable code that was 8-bit or 16-bit based, and being run in a 32-bit or 64-bit environment using a more advanced or entirely different OS. Whatever the reason, at some point, older technology, simply cannot be repaired or replaced without extraordinary costs being incurred to maintain the vintage equipment, along with the constant concern that the vintage repair parts will continue to simply fail due to age or become almost entirely incompatible with current technological solutions. It is at this point that a decision must be made to abandon the older technology for newer, or simply replace the machine or process line (if it is possible). In the case of an expensive, complex, useful machine or process line that is otherwise mechanically sound, having complicated Relay-Relay Logic (RRL), PLC’s and modern computers have successfully been used to replace the failing RRL logic (e.g., everything from telephone exchanges using rotary step relays and switches to packaging or bundling machines using time-delay relays with pneumatic timing bladders). HMI displays and physical interfaces are no different. Vintage HMI’s were initially constructed using monochrome LCDs and CRTs with simple 8-bit video drivers for alpha-numeric-only displays that did not convey very much information about a machine or process, along with computer keyboards or embedded/sandwiched wafer switches that physically wear out. GUI’s today are color, high-resolution displays with embedded screen multi-touch that can have computing ability well beyond the older equipment and can include connectivity to databases, host thin clients and display operational data including trends and process efficiency data. Generally speaking when retrofitting older equipment, a cost-benefit analysis on retrofit vs replacement must be done. This is required before any retrofit is really implemented, because, as the person asking the question has demonstrated, retrofitting can get into a great deal of reverse engineering, and in some cases, much of the information needed to mimic or replace functionality in the machine or process line may not be available anymore, and certainly not available electronically. Many younger engineers may not have the working knowledge needed in Mechatronics to be able to successfully retrofit older machines and HMI’s, so it becomes necessary to remove and replace enough of the older technology to satisfactorily restore proper operation of the equipment. In the very simplest of cases, where a CRT-based HMI, which was previously a display and a keyboard/keypad is being used, its technologies need to be removed back to the actual machine or process controller. If this is a supported legacy PLC, using standard legacy communication protocols, such as Modbus, etc. using software like InduSoft Web Studio in a flat-panel PC or an embedded solution such as Wonderware InTouch Machine Edition in a Flat-Panel using an embedded OS can be a very cost effective solution to this problem. The form-factor of the flat-panel PC or embedded device can be purchased to fit into the same physical environment as the older HMI, and these software packages are extremely simple to use and configure. Additionally they already come with industry drivers for legacy protocols. Additionally, there are utilities to directly replace the older programmed HMI’s such as Allen-Bradley Panel Builder or Eaton’s Panel Mate and import the tags directly into the software. If necessary, serial protocols used to interface older machine or process control logic to the HMI can be mimicked using the TXRX Serial Driver-to-TCP/IP included with InduSoft Web Studio and InTouch Machine Edition. For more unusual HMI implementations such as the one being mentioned in this question, where the actual legacy HMIs are still functional, but finding a working computer to run the programming software is problematic, again your only choices are reverse-engineering and adapting modern technology, or replacement. Since reverse-engineering requires specific expertise and the product design documentation may be sketchy, incomplete, or unavailable, this problem is now compounded immensely. Reverse Engineering a device and adapting or constructing technology to interface to it can be a very expensive, time consuming endeavor. Additionally it makes the equipment “one-of-a-kind” with only one or only a few people who understand how it works. Commonly people who are looking for this kind of shortcut or “cost savings” have not considered the real costs involved with this kind of project, including the months of development time required to get the new interface equipment running well enough to be useful. Additionally, while the HMIs may be operating OK well past their replacement cycles, the might start failing without warning, and without any replacements available, everything that was just previously done to extend their service life was wasted effort and resources. Additionally this situation can ultimately create an “emergency” situation where production just stops until adequate replacement equipment can be secured and installed. Richard Clark, InduSoft/Schneider-Electric Software Automation and Control Systems Engineer</p>

    Reply

  • <p>The challenge of upgrading and modernizing legacy equipment is a top priority for many machine builders today. As others have noted, outdated tools have limited capabilities, and upgrading equipment can be expensive.</p> <p>You could consider doing a cost/benefit analysis of retaining the older application code and moving it to a newer platform versus building something that is partially migrated but written from scratch. Using the latest technologies will allow you to take advantage of new capabilities. For example, if your HMI is using legacy alarming based on an external database, a new alarming approach may be easier to integrate with your controller. With a new alarming approach, you could build alarms in the controller and receive more accurate state-change and time-stamp information in to your HMI. Newer technology also offers re-usable displays and graphics (objects), which help ease the process of developing HMI content. </p> <p>Many machine builders are starting to see obsolescence planning as an opportunity to correct issues from both a platform and application perspective. This creates a more sustainable solution for the future, rather than continuing to battle migration issues with each piece of aging equipment or software individually.</p> <p>- Chirayu Shah, Commercial Program Leader for HMI Software at Rockwell Automation</p>

    Reply

  • <p>As other people have noted, unfortunately there isn’t much a customer can do in a situation like this. It’s up to vendors to enable a smooth migration to their new platforms. When GE developed our new QuickPanel+ operator interface, we made sure that programs that our customers have created in our legacy OI can be ported over to the new model. Upgrading to a new HMI with the latest features such as multi-touch gesture and secure remote connectivity is great, but not a the expense of migrating the customers from older technology. That’s why choosing an automation vendor will become increasingly important as the pace of technological innovation increases.</p>

    Reply

RSS feed for comments on this page | RSS feed for all comments