just another PID / SLC question

Mathias

Member
Join Date
Feb 2004
Posts
20
hi all,

can anybody tell me how AB's (slc 5/05) p, i and d-s are set
(i've searched ab.com without any luck)

p is in gain, but there is a .1 and a .01 possibility
does one have to multiply the calculated gain by 10 or divide it??

i is in min/rep (*10 or /10) of (*100 /100)

d i don't really care about,

thanks for the info,

Mathias
 
From SLC Instuctiion Help.

Tuning Parameters

Controller Gain Kc Type in a value. (With 5/02 controllers the valid range is 0.1 to 25.5. With the 5/03, 5/04, and 5/05 controllers the valid range = 0 to 3276.7) This is the proportional gain. Generally set this gain to one half the value needed to cause the output to oscillate when the reset and rate terms are set to zero. Note: RG bit must be set to accept values exceeding 25.5 when using 5/03, 5/04, and 5/05 processors.
Reset Ti Type in a value representing minutes. (With 5/02 controllers the valid range is 0.1 to 25.5. With the 5/03, 5/04, and 5/05 controllers the valid range = 0 to 3276.7) This is the integral gain. Generally set the reset time equal to the natural period measured in the Gain calibration above. Note: RG bit must be set to accept values exceeding 25.5 when using 5/03, 5/04, and 5/05 processors.
Rate Td Type in a value representing minutes. (With 5/02 controllers the valid range is 0.1 to 2.55. With the 5/03, 5/04, and 5/05 controllers the valid range = 0 to 327.67) This is the derivative term. Generally set this value to 1/8 of the integral time above. Note: This word is not affected by the RG (Reset and Gain Range) bit.

Rockwell Software 2000
 
there's no such thing as a "simple" PID question ...

Greetings Mickey,

could you please point me to exactly where you found the quote you posted above? ...

Note: RG bit must be set to accept values exceeding 25.5 when using 5/03, 5/04, and 5/05 processors.

this information is incorrect ... now I know that it’s not your fault because you’re obviously quoting something from the official “book” with a copy-and-paste ... it’s just that I sort of have a hobby of finding obscure “quirks” like this in the official documentation and filing them away in my own little scrapbook ...

anyway ...

[attachment]

as shown above, it is entirely possible to enter a value greater than “25.5” even when the “RG” bit is turned OFF (that is, it is not “set”) ... the example I’ve pictured uses an SLC-5/05 processor ...

and now on to other things ...

Greetings Mathias,

I’m not sure that I understand exactly what you meant by your first post ... are you asking how to “calculate/determine/estimate” the appropriate values for “Kc” and for “Ti”? ... if so, then the rest of this post will be a waste of your time ... but I’m going to assume (gosh I hate that word) that what you ASKED is exactly what you MEANT ... in that case:

first ... since your SLC-5/05 processor supports it, just go ahead and turn the “RG” bit ON (=1) ... this (as shown in the picture above) will give you two decimal places of resolution for your “Kc” and your “Ti” entries ... with the “RG” bit turned OFF (=0) you will only have one decimal place of resolution for the “Kc” and the “Ti” entries ... more resolution is a good thing ... turn the “RG” bit ON and leave it ON ...

now for the “Kc” setting (the one you called “p”) ... suppose that you’ve already decided that the amount of Proportional gain that your application requires is “99.15” ... in that case you should go to the PID Setup Screen and click on the “Kc” entry field ... then type in “9915” ... the software will automatically interpret this as “99.15” and display the value “99.15” in the “Kc” entry field ... note: if you want to dig a little bit deeper, go look at the value in “word 3” (of words 0 through 22) of your PID control block ... there you will see the (integer) value “9915” ... just remember that the PID will use the correct desired value “99.15” when it performs its calculations ...

just to nail this down ... suppose that your PID is set up as in the lower window in the picture above ... knowing how I set up this example might help you understand the units that seem to be confusing you ... notice that the “Smax” setting has been entered as “100” ... notice that the “Smin” setting has been left at “0” ... these settings mean that I want my PID to consider both its “setpoint” and its “input” as numbers which range in value from “0 to 100” ... as in “percent” of full scale ... notice that I’ve entered the Setpoint as “40” (for 40%) ... next I went to the PV (process variable – the PID’s input) location on the data table ... there I manually entered the value “6389” ... I had arrived at that number this way: 16383 times 0.39 equals 6389 ... the value “16383” is the PID’s way of expressing a “full scale” input reading ... the number “6389” is thus a 39% input reading ... specifically, by manually entering the value “6389” into the PV location, I’ve told my PID that the input is 1% below the desired Setpoint ... this results in the PID seeing an Error (a departure from the desired target) of 1% ... you might notice that the “Scaled Error SE” field displays a “rounded-off” value of 2% ... don’t let the “rounding off” confuse you ... now ... given the 1% Error and the “Kc” entry of “99.15”, the PID calculates an output this way: 99.15 times 0.01 equals 0.9915 ... this is an output (for the Proportional component) of 99.15% ... and that value (rounded off to “99”) is what is being displayed in the “Control Output CV (%)” field ...

note that the “Ti” entry of “327.67” results in a VERY small amount of Integral action ... that’s why it doesn’t immediately affect the PID’s output ... the “Control Output CV (%)” field will, however, eventually increase to “100” percent if we wait long enough for the small amount of Integral action to accumulate ...

assuming that I have correctly interpreted your question, the information above should have nailed down the entry of the “Kc” (or as you said, the “p”) setting ... if you need more than this (or something entirely different) then please post again with more details on what’s giving you trouble ...

as for the “Ti” (or as you said, the “i” setting) ... the same “how-to-enter-the-value” approach applies ... note that the units of this setting are in “minutes per repeat” ...

this post and this post should help you understand how the PID interprets those particular units for the Integral term ...

since you said that you’re not worried about the “Kd” (or as you said, the “d”) setting then I’ll skip any discussion of it for now ...

disclaimer: the values used in the examples shown above are NOT realistic ... they are intended for educational purposes only ... please do NOT assume that they might be anywhere near correct for your application ...

I hope this helps ... if it doesn’t, please include some more detail in your next post again and I’ll try again ...

rgbit compared.jpg
 
Last edited:
Ron,
Your right, copy and paste ,I guess you can't always believe what
you read. Or is there something else to this that we both don't
know about.
 
so where did this come from? ...

come on, Mickey, please help me out here ... how about a link or a publication and version number? ... I couldn’t find that business about the “RG” bit in the

SLC-500 Instruction Set Reference Manual

the “RG” bit is described on page 9-10 [Adobe Reader page 176 of 648] but the incorrect “must be set for values over 25.5” text doesn’t appear in that section ...
 
I believe Mickey got that from the RSLogix500 help file. If you check
under PID:Additional Parameters, it says it right there.

shot1.jpg
 
thanks Marty,

I finally found the part that you pictured ... even so ... the text that Mickey posted is “different” so I was just wondering exactly where his information came from ...

thanks again ...
 
I believe it's the same text, just when it gets copied and pasted the formatting is removed and everything is run together.

Marty
 
martyh,

thanks for your help. You got it. Sorry for the delay as I had
more important things to do. A morning at the park with my two
year old grandson.NO GREATER JOY THEN THAT.
 

Similar Topics

Hi all, I posted recently about a HVAC-style process control application I'm working on, and have a related question that probably warrants its...
Replies
12
Views
3,907
Hi all, I have pretty much read all the PID threads on this forum which helped me a lot. I am working with the following system and trying to get...
Replies
21
Views
6,230
Hi Guys, I have been trying to tune a PID for a tank heating system in RSLogix5000. The system is a reasonably large volume of water at about...
Replies
8
Views
2,158
This is my first post here...so HI I own a small injection molding operation (thermosets) and I've had good success replacing the relay logic in...
Replies
9
Views
2,379
Just curious, trying to make set up of 20 PID tags identical. What if on first scan I COPY MyPIDtag01 to MyPIDtag02 Length 1 COPY MyPIDtag01 to...
Replies
4
Views
2,917
Back
Top Bottom