Sign in to follow this  
Followers 0
blueheeler

Omron vs AB scan time

17 posts in this topic

Is the Omron's CP1H significantly faster than AB's ML1100. I'm working on a project where we are going to pack parts coming off a press every 250 ms (240 parts per minute). And I've been told that the scan time on the ML1100 will be a problem.

Share this post


Link to post
Share on other sites
I'm working on a project where we are going to pack parts coming off a press every 250 ms (240 parts per minute). I've been told that the scan time on the ML1100 will be a problem and that the Omron CP1H is faster. How does the speed of the ML1100 compare to Omron's CP1H?

Share this post


Link to post
Share on other sites
The CP1H is very fast for a shoebox but you would need to download a CP1H manual and work out an approximate scan time. Let us know if you cannot download the manuals and we may be able to post them here for you.

Share this post


Link to post
Share on other sites
It will depend on how fast you need things to happen, or more importantly how repeatably. If you are looking for precision at 1ms timing, Omron will do much better than A-B. A-B treats their timed interrupt with lower priority than they should. If you try to use anything lower than 3-5ms for the interrupt rate, the scan of the timed interrupt will vary some due to overhead issues and varies even more if communication is ongoing - being online with a pc being the worst case. Omron has a better repeatablility for the timed interrupt even at the lowest time intervals. Other than that, you will need to check the specifications for each PLC with regards to what you are trying to accomplish. Just thought I'd mention the TI issue, because A-B has been fudging what they can really do for a timed interrupt from the beginning of time.

Share this post


Link to post
Share on other sites
I'm betting that both PLCs are faster than the actuators they are controlling.

Share this post


Link to post
Share on other sites
Alaric you are almost invariably right about the scan being faster than actuation. However, if precision timing is involved, repeatability is just as, if not more, important than relative speeds. For example, if you are controlling a solenoid that takes 20ms +/-0.5ms to energize, 15ms +/- 0.5ms to de-energize and you are trying to control the total time of actuation to the nearest millisecond, a scan time of 5ms is way faster than the actuator, but completely unacceptable for the control solution. On the other hand, if the actuator has 10ms of variation in its activation or release or 10ms of variation makes no real difference to the application, then a 5ms scan is no worse than 1ms. Based on the machine rate of 4 cycles per second, I was assuming he was looking at precision timing.

Share this post


Link to post
Share on other sites
MicroLogix 1100 specifically traded some raw speed for other parts of the feature set, like user-configured web pages, online editing, and lower power consumption. The MicroLogix 1500 is actually the fastest member of the SLC/MicroLogix family for pure logic-solving. When I do high-speed functions with the MicroLogix 1100, I tend to make more use of the discrete Event Input Interrupt (EII) than the Selectable Timed Interrupt, which I tend to use more to keep longer periodic tasks more accurate than normal Timers can make them. The first 4 inputs of the DC-input MicroLogix can interrupt the controller and run a specified ladder routine immediately. Is your concern for the ladder scan, or a repeatable time-based function, or an interrupt-based function ?

Share this post


Link to post
Share on other sites
Allen Bradley MicroLogix 1500 is actually the fastest member of the SLC/MicroLogix family for pure logic-solving # Approximate scan time for a typical 1K user program (includes timers, counters, etc.): 1 millisecond # Simple bit instruction execution: 0.7 microseconds OMRON Slowest Baby shoe box PLC - CP1L - Simple Instrucions 0.1 micro seconds Latest CJ1-H-R Basic Instructions 16 nano seconds = 0.016 micro secs 30K program 7:3 ratio with basic and special instructions - 870 micro seconds = 0.870 milliseconds Interupt response 40 micro seconds BTW My 2 cents.... I use a number of diff brands of PLC's and Online editing should always be available on todays modern PLC's. No excuses....

Share this post


Link to post
Share on other sites
I never understood dumbing down the PLC just because its a brick, something that AB is notorious for. Dumbing down the brick = re-engineering the logic engine = increased development costs + customer frustration. I've wondered why AB doesn't put a logix5K engine in a brick. Maybe AB doesn't want their bricks encroaching on sales where the customer might use a compact-logix instead because of the brick's limiations. IMO, in doing so to avoid stepping on their own right foot, they step on their left foot.

Share this post


Link to post
Share on other sites
That is fast. That is about 60 MHz. Now the question is what is a basic instruction? The TI DSPs we use are running at the same speed and can do a floating point multiply and add in that time. We use 32 bit wide 0 wait state ram so there are no wait states when fetching memory. So our basic instruction is 16.66 nanoseconds too but it takes a clock cycle to load the data, another clock cycle to test the bit and another to save the result on the stack before getting the next bit. This means it takes a couple basic instruction to make one check if a contact is on or off. Our controllers with Ethernet run a 400 MHz. I can do a basic instruction in 2.5 nano seconds. However, the same steps must be taken to test a contact. The difference is that the 400 MHz CPU has wait states so it is only that fast when the data is already loaded in the cache. In reality it is about 2 as fast at the 60 Mhz DSP, not 6 times as fast. Too many of you guys get hooked on numbers that really don't mean much unless you can write identical programs and do a apples to apples comparison with a scope or something else that is accurate. On line editing takes some overhead. I would bet the Rockwell compiles each rung into machine code to get the best speed results. The rung are downloaded somewhere in memory and a point to the rung is kept. When the PLC does a scan it scans through a list of pointers to rung executing each on in turn. The code for each rung doesn't need to be in any particular order because the rung list keeps track of the execution order. Accepting the edits just changes the pointer pointer in the run list and marks the new rung as active. PLCs without on-line edits are much simpler to implement because all the code can be compile in one block before downloading. No list of pointers or memory management would be required. Small PLCs may not have enough memory to manage. This is ok for small PLC where it doesn't take too much time to download the code. People would protest if they had to download the whole program for a large Control Logix program.

Share this post


Link to post
Share on other sites
Pete, your TI processor would still be faster. From what I undestand, a basic instruction is a LD / OUT etc... the Floting Pt etc would be Special Instructions.... Back in 2004, being there and done that with a real world application.. AB SLC 5/05 15.5 kW memory used. (inc data table words) .. Converted to Omron CJ1G and added a hell of a lot more code to handle HMI and other plant controls (Total Steps 20K plus all of the data tables on top) and yes, the CJ1G was considerably faster then the SLC. Unfortunately, i cannot directly compare to a newer micro logix 1500 with the same program. BTW, I still stick to the fact that "Online editing should always be available on todays modern PLC's. No excuses....". I will put this into perspective.. pretty much all of the Omron PLC's released since 1990's have on line editing, CPM1, CPM2A, CQM1, CQM1H, C200H, C200HS, C200Halpha, CS1, CJ1, CP1L. The only one that does not is the ZEN which is a programmable relay (but hey, i could be wrong on this as well).

Share this post


Link to post
Share on other sites
It's coming...In fact I think it was scheduled to come out soon. They showcased it at Automation Fair I heard

Share this post


Link to post
Share on other sites
Peter, I have been reading this topic with great interest. Excuse my ignorance, but what is a TI DSP?

Share this post


Link to post
Share on other sites
TI = Texas Instruments DSP = Digital Signal Processor, a type of microprocessor that specializes in digital signals

Share this post


Link to post
Share on other sites
http://en.wikipedia.org/wiki/Digital_signal_processor

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