Calculation Help

[gatfu]

Hi all! I was looking for a little help in determining a better way of accomplishing a current process that is not to efficient with in my program. Here is the set up....

ML1400 running RSlogix 500
1 Thermocouple
1 Ignitor

During our operation we divert Syn gas to a blow off flu. At the point of diversion we are turning on the ignitor to lit off the excess gas. Once we have reached a temperature of 600 deg f we then turn off the ignitor. Here in lies my problem. The gas always lights at different temps, sometimes at 550 sometimes at 300 all depending on the quality of the gas coming through the flu. If it lights at 550 then it only takes a minute or so for the thermocouple to reach 600 and then turn of the ignitor however if it lights at 300 then it could take as long as 5 minutes to turn the ignitor. this is terribly inefficient as well as it dramatically decreases the life of the ignitor being on all that time.

I am trying to figure a way to program it so that once the PLC sees the thermocouple rising at a steady rate then to shut the ignitor instead of waiting to reach a certain temp. I believe this is totally doable however I am not to good with the advanced math instructions. Any help in accomplishing this is greatly appreciated!

 

[bernie_carlton]

If you record the temperature just before firing the ignitor, then continue to monitor the temperature would it be ok to turn off the ignitor when the temperature has reached a certain number of degrees above the pre-ignition temperature?

 

[gatfu]

I must be missing the instruction to freeze a recording because I was thinking that as well but couldnt find the instruction. I was trying to fool around with one shot but i know thats for bit level.

 

[TConnolly]

Add the desired temperature rise to detect ignition to the starting temperature. Then do a compare.

See the attached example. This only shows you how to turn off based on a temperature rise. You still need to program any safety considerations.

 

[gatfu]

Thanks Tconnolly....gonna give that shot now.

 

[Spartan13]

Idea

Add the desired temperature rise to detect ignition to the starting temperature. Then do a compare.

See the attached example. This only shows you how to turn off based on a temperature rise. You still need to program any safety considerations.

I'm assuming that the igniter will raise the temperature also, just at a slower rate, so in your example you'd hit a problem if the temperature went over 400 without igniting the gas...

I'll look into figuring out how to properly attach .zip files so I can post a RSL of it, but here's what I'd do (given what I know now):

Your examples (300-600 in 5 mins or 550-600 in one min) imply a gas heating rate of about 1 degree/second, after ignition. I am assuming that your igniter raises the temperature by less than 0.5 degrees/second.

T4 is the timer used here;
B3:0/1 is the Igniter (I'd use line 0 strait from TConnolly's program, but for simplicity's sake that is all presumed);
I:1/1 is the thermocouple input.
N7:1 is the temporary temperature setpoint (temp temp )
> is used to separate parts of the same instruction.
- is used to separate inputs and outputs.
// is used for notes.
& is used to indicate AND for commands/outputs.

(XIC) S:1/15 - (MOV) 300>N7:1 //Startup temp setpoint

(XIC) B3:0/1 & (XIO) T4/DN - (TON) T4 (4 seconds) //rolling timer

(GRT) I:1/1>N7:1 or (GRT) I:1/1>600 - (OTU) B3:0/1 //Igniter off

(XIC) T4/DN - (MOV) I:1/1>N7:1 & (ADD) 3>N7:1 //Setpoint raised every 4 seconds, based on the current temperature

Basically, every 4 seconds N7:1 is set to 3+I:1/1. If I:1/1 rises by >3 degrees in 4 seconds, there must be ignition and the Igniter is turned off. If not, the process continues until 600 degrees is reached, by which point ignition should have occurred.

This may be a little too rapid for what you're doing (airflow/thermocouple inaccuracy/scan time) but the numbers can all be changed as needed (check every 20 seconds if it's risen more than 15 degrees etc).

Just food for thought, hope it helps; let me know if there's anything wrong and I'll try to fix it.