Sequencing Temp transition over time

saultgeorge

Lifetime Supporting Member
Join Date
Jul 2015
Location
Detroit
Posts
535
Hey, folks!! I need professional input on a sequencing I'm creating in RS5K V. 20.04. I have to have this done by the 5th, next week!!

I have a cooling and heating routine in an Air handler I've written logic for. The unit is used for extreme cooling and heating in a test cell for prototype cars.
Their system uses a liquid called Syltherm, which will get to minus 40 and up to 140 Deg. F.
Their spec calls for sequencing from 140 to 30 Deg either raising or lowering not more than 20 Deg/hour. From say, 29 to -40 no more than 10 Deg/hour. I'm assuming this is so the mechanical systems (plumbing, valves, etc.) do not experience extreme thermal shock. Makes sense. So far I have 2 different gain rungs for each PID (Cool and Heat valves) that are conditioned by limit instructions and I have a timer that I can use with logic to calculate how much time it needs to take based on the difference between the temp and the new setpoint. Do y'all think I'm on the right track? Maybe someone can tell me how to do this in one rung, which is usually what happens. Thanks and have a great weekend. I can post snips of some of the logic if one of you folks needs it.
 
I'm not sure what you mean by "gain rungs".

The typical way to handle setpoint changes in a PID is not to 'stairstep' the setpoint (i.e., go directly from 140 to 30, or even from 140 to 130, but to ramp the setpoint down over time. This way the the PID trends the PV as the SP changes.

You can use the RAMP function block to do the ramping, or just roll your own:
PID.SP = SPold + ((SPnew - SPold) * TON.ACC/TON.PRE)

When a new setpoint comes in, one-shot MOV ((SPnew to SPold), calculate the TON.PRE by subtracting the two and dividing by the ramp rate, and RESet the timer so that it starts timing again.

The calculation for TON.PRE get a bit trickier if you've got more than one ramp rate, as you seem to indicate (20° /hr & 10° /hr), but not that difficult (and why I prefer the roll-your-own ramp logic over the canned one).

Not in one rung, unless you get branch-happy. But IN THEORY, an entire PLC program could be done in just one rung; just using lots and lots of branches. It wouldn't be readable, though.
 
Garry
I can't use any other hardware/software. I have to use what they already have installed. They're motor actuated valves, 0 - 10 VDC Out.


Aardwizz--that looks like a good idea. I was joking about 1 rung. I have some of the calculations in already, I'll just tune it up like you show there and give it a try. I have a PLC here in the office so I can test everything, which is good. Thanks, guys!!!
 
Aardwizz
BTW, "gain rungs" I mean I have rungs with PID Parameters (from a UDT) in move blocks so I can tune the PID while it's running.

So I made one set of moves (Kp, Ki, Kd) for the 20 Deg/hour and one for the 10 Deg/hour.
 

Similar Topics

Does anyone have sample logic with conveyors that start and stop according to product on the line?
Replies
1
Views
99
Hello, I am struggling to figure out some logic. I have an analog pressure sensor coming into the PLC and would like to have multiple timing...
Replies
4
Views
150
One of my customers builds equipment with a MicroLogix 1100 using a program I wrote some years ago. Now he wants to transition to a Micro 820...
Replies
1
Views
1,857
Well Gentlemen, once again, I am trying to figure out how to do something in Logix, and I cannot seem to get it to work like all the YouTube...
Replies
13
Views
2,931
I need to make a sequence of lights flash on and off in automation studios using AB PLC. I have to use timers and counters. The required sequence...
Replies
51
Views
13,512
Back
Top Bottom