ML1100 and HSC

[gosenbach]

I agree, you have to dig though there and I think it is just a small side note that there is actually only one accumulator.

 

[hounddog]

Question

had a meeting with our local ab rep and he said that the 1100 has multiple high speed counter inputs and that we could connect two encoders to it.
have a project coming up that will require two encoders.
After reading this post, it appears that you only have one accumulator, so is it possible to use two encoders?
Will download the manual in the morning when i get to work.

thanks

 

[hounddog]

just downloaded the manual, and sure enough, the 1100 only has one hsc input.
Now I feel dumb. Should have read it instead of just listening to the rep.
I need to use two encoders, so it appears the 1100 will not be able to do that.

 

[darrenj]

This is one post i will be keeping an eye on.....
If the HSC only updates once per scan..what is the point in a high speed counter?
if the plc only looks at it every say 20 milliseconds but my value is changing every 10 milliseconds then the potential for missing something is high..

I am only wondering because of this...

my first control logix job was a filling machine...everything on the machine was timed according to where the cam was at any given time..(I had to remove and old Giddings and lewis processor because they could get spare parts for it anymore)..the old G&L used a resolver..so i stayed with that..stictly because it was critical where the cam was..if power was removed and turned back on they wanted to be able to fire up the machine right away..not have to set a home position..

all was well till trial run..the i noticed that some cartons were empty..turned out that the instruction i had used was looking for a specific value on the resolver..say turn valve X on at 240 deg...well because it would only update the deg every scan every once and a while i would miss that 240..i ended up having to put in a buffer..say turn valve x on if cam between 210 and 260 deg...this works ok..but not how i figured it would work.. i

I figured if an 20 year old PLC can do it bang on the degree then a new state of the art control logix should have no issue...guess i was wrong..and thats my fault for not understanding

would i have been better off putting an encoder in and using a HSC..i guess not if it still updates only once per scan....

?? :confused?

 

[CroCop]

This is one post i will be keeping an eye on.....
If the HSC only updates once per scan..what is the point in a high speed counter?
if the plc only looks at it every say 20 milliseconds but my value is changing every 10 milliseconds then the potential for missing something is high..

I am only wondering because of this...

my first control logix job was a filling machine...everything on the machine was timed according to where the cam was at any given time..(I had to remove and old Giddings and lewis processor because they could get spare parts for it anymore)..the old G&L used a resolver..so i stayed with that..stictly because it was critical where the cam was..if power was removed and turned back on they wanted to be able to fire up the machine right away..not have to set a home position..

all was well till trial run..the i noticed that some cartons were empty..turned out that the instruction i had used was looking for a specific value on the resolver..say turn valve X on at 240 deg...well because it would only update the deg every scan every once and a while i would miss that 240..i ended up having to put in a buffer..say turn valve x on if cam between 210 and 260 deg...this works ok..but not how i figured it would work.. i

I figured if an 20 year old PLC can do it bang on the degree then a new state of the art control logix should have no issue...guess i was wrong..and thats my fault for not understanding

would i have been better off putting an encoder in and using a HSC..i guess not if it still updates only once per scan....

?? :confused?

CLX updates the IO independent of the scan time.
2 questions up front:
1. What was the RPI of the IO?
2. What was the scan interval set at?

 

[darrenj]

going from memory...the RPI on the resolver module was at 1ms..any lower and it would fault on me..

the scan interval was set as continuios

Just pulled the program..the above are actual values..not from memory

 

[CroCop]

Was the resolver the same from the old PLC to CLX?

1ms is pretty fast.

You know you can have event based tasks in CLX?

 

[darrenj]

no.. the resolver card i got was from AMCI..rather than avoid the old..its the resolver vs the card routine i got a brand new resolver from AMCI..i know about the event based..but i again..lack of experiance..i didnt want my interupts to screw something else up...to be honest i figured if a 20 year old dinasour can do...this would do it while making the coffee!..However after a little research i found that G&L are the bong for motion stuff...(Not disrispecting any company here..just G&L do what they do very well)

 

[TimWilborne]

Alaric. Got the answer. Looks like the accumulator is updated only ONCE per scan OR with the I/O Refresh instruction.

The HSC value is only updated once (every scan) to the HSC:0.ACC word in the ML1100. Any time the HSC:0.ACC value is referenced it will be the same value (per scan). The UA bit that existed in the SLC doesn't exist in the ML1100 however you can use the REF (I/O Refresh instruction) and it will update all the IO along with the HSC.ACC value. Example I can stick a REF instruction at the 1/5, 2/5, 3/5 and 4/5 ths point in a program and each HSC reference afterwards will be updated (not the same as first reading - based on a continuous pulse coming into my counter).

 

[gosenbach]

Thats great info. Exactly what I was looking for!

Thanks for finding that!

 

[Bob A.]

From the early days of analog inputs, the advice was given to buffer the input data such that it could not change during the program scan. I usually avoided this by only using the value once in the program.

If it matters, I would assume that the count can change anytime, until someone can assure you otherwise.

Bob A.

 

[TimWilborne]

If it matters, I would assume that the count can change anytime, until someone can assure you otherwise.

Bob A.

I just told you otherwise two post ago
http://www.plctalk.net/qanda/showpost.php?p=254284&postcount=24

 

[steveb1475]

Sorry to bring up a semi dead topic, but how fast of an application are we talking about? Here's a little one that I wrote quite a while ago to kick bottles with no lid off of a conveyor using an encoder (1000 ppr) and old filtec rejector that we had laying around at 600 bottles/min. Just put the time critical stuff in the counters interrupt routine with an immediate output.

 

[Ken Roach]

Reading some of the last posts, I see that the "why use an HSC instruction if the accumulator only updates once per MicroLogix program scan" question arose.

The answer, of course, is that the HSC performs best when your output functions are controlled in the User Interrupt Routine that is triggered when the accumulator reaches the High Preset. The main program doesn't look for the accumulator values, the HSC does.

There is another function of the MicroLogix 1100/1200/1500 high speed inputs, called the Programmble Limit Switch (PLS) function. This is more like an accumulator-driven cam switch.

 

[irondesk40]

That is what I advised on the thread I refered to earlier - the first thing to do immediately before starting calculations is to copy HSC:0.ACC to another register. Please note that the HSC:0.ACC is a long integer, therefore I do not recommend using a 16 bit N file. Use L9 or other Long file and use 32 bit math. See the thread in another forum for how to deal with HSC roll over if needed.


Also keep in mind that a timed interrupt routine can be used to read HSC:0 elements at specific intervals.

Alaric
I looked at your post on the link you posted.
Could you post a example program? I could not quite understand your explanation. The hsc is something that i am looking at using in some applications with the micrologix series and very interested in your explanation.