I need to design a PID controlled tank pressurization system using a PLC5-80C1.5

The tank pressurization will use a series of timed setpoint changes, with adjustable ramp rates, either down or up as the setpoint change requires.

Tank pressurization is handled by a variable position valve (4 - 20 ma), and tank venting is handled by a separate but similar variable position valve. In operation the tank is being drained and filled simultaneously at different rates while maintaining and changing setpoints and ramp rates (psia/sec) at preset times. Starts at 100% (small ullage) and finishes at
<30% full (large ullage). Pressurization utilizes GN at 2500 psi supply pressure.

For the curious, this is a 50,000 gallon Liquid Oxygen tank used to supply oxidizer to a rocket engine during static test. Pressurization to 100 psia max is currently handled by a very old home built electronic controller built by a genuine but now departed Rocket Scientist. It is old and
difficult to keep operating, much less calibrated.

The "ramp (up/dn) to setpoint" problem is straight forward. The pressurize / vent function is my problem. Should I use two separate but coordinated pid loops, or one pid loop biased to 50% output at setpoint with the CV inverted below 50% and direct above 50%? i.e. 2048 counts at 50% converted to 0 to 2048 in each direction. I recognize this approach will cut my resolution in half, giving 0 - 2048 counts for each 4 to 20 ma as opposed to 0 - 4095 counts for the same range.

Any comments or suggestions or experiences would be welcome.

David McConnell
Rocketdyne Propulsion & Power
Stennis Space Center, MS

 

The normal approach when you have two actuators having the opposite effect on the process is to do such a split (not necesary at 50% however)

But in your case (if I got it right) you do not want them to be both to be closed at 50% right? but rather half open or something to have some flow? If yes then you don't loose resolution (you simply reverse the signal for one of the valves)
If no - ie the "normal" way both valves would be closed at 50% (or whatever your split-point is). Below is one opening and above the other.

Anyway trying to solve it using two controllers is probably quite hard. I wouldn't worry too much about the lost resolution, there is probably more resolution than you rave mecanically in your valves anyway.

You might want to put the split at some other place than 50%, the "best" palce to put it is so that the process gain is as much the same above
and below as possible.



/Johan Bengtsson

Do you need education in the area of automation?
----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------

 

I agree with Johan: single PID In fact single PI (and little I ). At sometimes in the past Tufline had a controlled 3way valve. But two paired valves should do same.
Your black box number is chinese to me.

 

I have a somewhat different system, although a similar control. I have two VFD's pulling air from a common plenum. I want to maintain a somewhat constant vacuum within the plenum, but I don't want the fans to "fight" each other. I did separate PID loops for each fan. The loops utilize the same PV. I want to control at about -1.25 in/h2o. So I set one at -1.3 and one at -1.2. If the air into the plenum is less than one fan can pull out, The -1.3 fan controls the in/h2o at -1.3, and the -1.2 PID moves it's fan to minimum speed. If the air into the plenum is greater than one fan can pull out, the -1.3 PID drives it's fan to max speed and the -1.2 PID controls the pressure there.

You might be able to do something similar. Set the inlet PID to 99 PSI with the outlet at 101. The difference in my app and yours is that you need to control in and out and I control only out. I'm not sure if it would work or not, but it may be worth some thought. Both valves may simply go to min, each trying to get to SP. If you reverse the SP's they may go max.

Is there anything past the exhaust valve that should limit how much oxygen is let out?

 

hi
instead of a 50-50 control, we have, one which is slowly controlled and the second regulates quickly. both have individual setpoints. If the quick pid has problems the slower setpoint is changed to help.
In either case it could be complicated.
later
zan

 

John,

I do not know what the original question was, but I would like to say that it may be appropriate to take a close look at the process objectives of the
equipment in conjunction with the hydraulic flexibliity offered by VSD's (see my book - available from ISA). While systems may "work" with staggered setpoints, they usually perform better and are more energy-efficient when
their process and control objectives are closely aligned.

Best Regards,

David W Spitzer, PE
Copperhill and Pointer, Inc.
http://www.icu.com/spitzer845.623.1830 (phone/fax)

Enhancing productivity by:
- solving measurement and control problems
- providing process control engineering guidance
- evaluating instruments and assessing new technologies
- presenting training seminars for people who use and make instruments

 

An easier solution (not that I doubt that your current solution works this is just an easier way to get the results) in your case would actually be to do as we suggest in this case too but slightly different:


Une one PID, split the output like this:

output fan1 fan2
0% 0% 0%
10% 20% 0%
20% 40% 0%
30% 60% 0%
40% 80% 0%
50% 100% 0%
60% 100% 20%
70% 100% 40%
80% 100% 60%
90% 100% 80%
100% 100% 100%

(illustrated with a split point 0f 50%, this is not necesary, but probably quite right since I suppose your fans are the same size)

You could then of cours implement some kind of switch (preferrably logical in the control system) to determine what fan is fan1 (so you can evenly distribute working hours if you like that)



/Johan Bengtsson

Do you need education in the area of automation?
----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------

 

Of course an even easier solution is to just put the PID output to both fans (ie 30% output from the PID make both fans run with 30%) but you have probably already thought about that...


/Johan Bengtsson

----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------