How are you using RSView SE?

[AndrewB]

I've just stuck another 2GB of Ram in our server taking it up to 3GB total. It never seemed to use what we had but wow, what a difference in speed of the clients....Seems to be more stable too, hasnt crashed in a month now!! (which is unheard of)

 

[harryting]

This seems like a good place for my RSviewSE questions. So here goes:

1. Do I need to buy it with RSLinx Enterprise? It says it comes with RSLinx for RSView (Classic, I presume). So is it worth the extra $ to get Enterprise?

2. I keep getting this annoying error "XXXX is marked read-only" when trying to save graphics. If I use the "Save-As" it works, even if the same name. WTF

3. I only have the Studio right now for development purpose. I can't seem to reference any RSLinx Enterprise shortcut, because, I guess I don't have the license for RSLinx Enterprise. Correct?

Are there other server-client HMI out there that people would recommend based on 1st hand experience?

Oh, yes, I have installed all the patches released thus far.

 

[Aceboyz]

RSView SE V5 CPR9

We implemented an RSViewSE V4 CPR7 system nearly a year and half ago, the system comprised of the following: -

V4 CPR7
Data & HMI Server Redundancy
HMI Tags Used
Alarm Services (Needed Events, Rockwell do not cater for this so we designed our own Event handling)
2 sets of quad turbines (2 Units) each unit had 6 PLC's
Using Linx Enterprise

When we first set this system up in the office with both servers and just a couple of clients it worked fine but on installation at site with all clients connected SE failed on many different problems i.e. Alarms not working, C+++ errors etc.

Eventually we had to force Rockwell to go and visit site so they could implement all of their patches, it has taken 6 months and 3 seperate Rockwell visits to get the system stable, i turely wish we had stuck to RSView32.

Now we have more issues, which i hope someone can help with.

The customer needs to have RSOPC installed, which i think is now called FactoryTalk OPC or something. On installing this software it has highlighted the need for upgrading the RSViewSE to V5 CPR9, this is a little concerning considering how much time and effort has been lost due to the problems on the previous versions and the last thing we want to do is to loose anymore confidence with the customer. Has anybody used the new V5 CPR9 and if so please can you let me know if there are any new issues with this version.


Thanks

 

[rdrast]

I have installations of FactoryTalk SE 5.0, with a couple patches. I'm not seeing any issues with SE, or communications. There is however a problem with the VBA engine, you cannot pass parameters across display boundries, they get dropped. That seems a bug in VBA though.

Then again, we here haven't had much trouble with any installations of View SE, all the way back to Vers. 3.1.

A big problem with SE, is that is is NOT easy to just load and go. There are lots of things that need to be understood about the software itself, operating system compatability, patch levels, access rights, etc.

It's unfortunate, but there are probably more incorrect installations of RSView/FTView SE out there then correct ones.

 

[SLC_Integrator]

I have installations of FactoryTalk SE 5.0, with a couple patches. I'm not seeing any issues with SE, or communications. There is however a problem with the VBA engine, you cannot pass parameters across display boundries, they get dropped. That seems a bug in VBA though.

Then again, we here haven't had much trouble with any installations of View SE, all the way back to Vers. 3.1.

A big problem with SE, is that is is NOT easy to just load and go. There are lots of things that need to be understood about the software itself, operating system compatability, patch levels, access rights, etc.

It's unfortunate, but there are probably more incorrect installations of RSView/FTView SE out there then correct ones.

We manage a system with RSView SE 4.0 - 1 pair redundant HMI/Data servers, 3 clients, 5 CLX and RSSQL.
Its pretty stable now but only after months (yes months) of patches and me dinking around with it with help from Rockwell tech support.

A flash chip fab plant not far from here put in a 14-server installation and had nothing but trouble with it.

There is no way I would reccommend this product to any of my clients. My advice is if you can afford it, rip it out and use something else.

 

[rdrast]

Not taking that advice. We currently have 8 or 9 seperate installations of SE here, and have for years. Still adding new ones on (working on one at the moment actually).

It works, why change it?

 

[jtn]

My experience with the 4.0 was that the people that didn't use the redundancy didn't have that many issues. Everyone I talked to that did use the redundancy had big big headaches, me included. I haven't done a redundant jop with 5.0, but if you try it I would back up the hard drive and be ready to restore to its existing condition in case problems arise again.

 

[Aceboyz]

Thanks for the comments, much appreciated.

The feeling from ourselves was that if we really have to move to SE V5 then we Ghost both the Servers and all of the Clients so we at least have something to fall back too.

We have only just got some confidence back into out customer and really can't afford to loose that again.

Thanks again.

 

[Amazingmg]

FactoryTalk View SE Distributed

I finished a Factorytalk View SE distr.(CPR9)/CLX installation back January. Customer loves it. System consisted of 2 HMI/Data servers (redundant), 4 Clients, Pi Box with Gateway, DC, Redundant CLX processors and I/O. No problems what so ever.

Doing start-up of a duplicate system at same facility right now.

The "tagless" system is the way to go and I love it. The only tags I had to create were for legacy PLC interfaces.

 

[rdrast]

The "tagless" system is the way to go and I love it. The only tags I had to create were for legacy PLC interfaces.

Ironically, after doing some systems with Direct Refrences (tagless) I have found that I will never do so again!

Nice idea, but the inability to edit tags on a display makes this a pain if you aren't doing just a 'One-Off' job, or if you want to reuse displays between jobs.

Even the pain of having the same tag on multiple displays, and then having to a) remember each one, and b) edit each one individually if you change the reference in the PLC... ech.

Plus, I can easily convert my ME and SE tags back and forth, and for the most part automatically generate them with the Logix Tag Import / Export Tool (BROKEN TO SOME DEGREE, But of course Rockwell won't fix it).

 

[Amazingmg]

Hmmm, I'm thinking about what you are saying.

Sometimes it can be a pain, but I just reused an entire application for a duplicate and it wasn't that bad. The wireframing (which is a great feature over Wonderware imo) indicates tags that need to be corrected and as we take control of the new plant (instrument by instrument, motor by motor) a glance at the screen tells me what is online with the new CLX and what is not. Works pretty good, I think.

Different strokes for different folks?

Oh, btw, the speed of the "tagless" system is one of the biggest selling points (or so they say), so if you are using tags for everything.....you are losing some of the benefit.

 

[OZEE]

I don't know that you're really losing anything by using tags. I haven't been too impressed by this "feature" either...

I try to do all of my documentation (at least for IO) before I ever start writing PLC code or developing HMI screens. This is done in Excel. In this spreadsheet, I've written code to automatically generate all of my documentation - both for HMI tags (either WonderWare or RSView) and PLC IO/Tag Descriptions (PLC5, SLC500, ControlLogix). Then I just have to import this automatically generated data into the HMI/PLC app.

 

[rdrast]

Oh, btw, the speed of the "tagless" system is one of the biggest selling points (or so they say), so if you are using tags for everything.....you are losing some of the benefit.

Speed? No speed differences, the dynamic request arrays to Linx, Linx Enterprise are built much faster than either data server can actually respond, so the extra level of indirection doesn't hurt.

Also, you lose the ability to enforce limits, or do scaling (on numeric references), and in quite a few cases, leaving View to decide on the numeric representation doesn't work... Defining a tag lets you explicitly have a value interpreted as a LONG, or FLOAT, or INT, etc. I almost always use the tag database to scale time .ACC and .PRE elements for example, to operate directly in seconds (and limit them to make sure that a negative (crashing) preset can never be entered, even if I forget to modify the limits on the actual entry control.


Re-use... I have many many screens that are common, even if used on very different overall machines.

I just export the tag dictionary to Excel, do a global search and replace on "[Machine_1]" changing it to [Machine_2].

Re-import tags with update, Done.

One other things I do, is generally, I try to develop at least my base screens in RSView ME, and then import them into SE. That way, I have a functional version ready to go if I toss a Panelview in somewhere (since you cannot import SE displays to ME).

 

[OZEE]

One other things I do, is generally, I try to develop at least my base screens in RSView ME, and then import them into SE. That way, I have a functional version ready to go if I toss a Panelview in somewhere (since you cannot import SE displays to ME).

GREAT IDEA!!

 

[Amazingmg]

Speed? No speed differences, the dynamic request arrays to Linx, Linx Enterprise are built much faster than either data server can actually respond, so the extra level of indirection doesn't hurt.

We did notice a speed increase with a complete system using direct addressing to CLX via enterprise. We had top notch hardware (fast proc, much ram, good switches, etc.) though and this could have made a big difference too.

Also, you lose the ability to enforce limits, or do scaling (on numeric references),

By rule always done in the PLC whenever possible. We want the HMI to be as "dumb" as possible with all calculating, scaling, limits, etc. done in PLC.

...and in quite a few cases, leaving View to decide on the numeric representation doesn't work... Defining a tag lets you explicitly have a value interpreted as a LONG, or FLOAT, or INT, etc.

I haven't had this happen, yet. The settings from numeric display ,for example, have been enough.

Re-use... I have many many screens that are common, even if used on very different overall machines. I just export the tag dictionary to Excel, do a global search and replace on "[Machine_1]" changing it to [Machine_2].

Re-import tags with update, Done.

I have common screens too. Duplicate the display, tag substitute the whole thing for new path. Done. No excel.

One other things I do, is generally, I try to develop at least my base screens in RSView ME, and then import them into SE. That way, I have a functional version ready to go if I toss a Panelview in somewhere (since you cannot import SE displays to ME).

THAT is a great idea, even though we don't do much Panelview.