Tuning in a pump PI loop.

sparkie

Lifetime Supporting Member
Join Date
Nov 2014
Location
KS
Posts
1,139
So, I'm being asked to tune in the spare drive in a PI loop for a pumping system. The system has a pump, 6" main line being fed from a cistern (the pump sits at the bottom level of the cistern), and uses a 4-20mA pressure transducer as feedback and an Altivar 21 drive as the controller.



The minimum pressure for the system is 17 psi according to the operator.



I have not actually tuned in a system like this before, and am just doing a bit of home work to spot any potential "gotcha's".


Only one drive will run at a time, and I'm planning to match parameters as a starting point with the working drive. The reason I'm asking is that recently there was a massive pressure drop when a fire hydrant was opened, and I wouldn't mind increasing the response time of the system to prevent this, and I'm wondering if decreasing the integral coefficient wouldn't accomplish this goal.



Are there any "gotchas" that this brings to mind or anyone or any specific info that I can give here to better define the problem?


I'm a bit new to this particular situation, and am just doing my homework before I blindly jump in.
 
Actually, while on this subject, I'm particularly interested in pumping systems, so if anyone has a quick link to an article or a good book that I can gain some more in-depth theory on the subject I would be very grateful. I've often found that experience can point to some pretty succinct resources.
 
I know this is not the drive you are using but the information is universal and should help you.
https://www.yaskawa.com/products/drives/iqpump-drives/iqpump-software-tools/application-simulator

just few things a pump running below about 20hz is not moving much fluid and running above 60hz the motor load will go up very fast if you need to run much above 60hz then you should look at adding a second pump


Thanks. Checking into this when I finish these load calcs. There actually is a second pump in the system, I just have it set up to only run one at a time. Now, they want me to go ahead and update the control loop as well.


Do I just need to fill out the form requesting the software to get a download link? I see they want an email and phone number.
 
Sparkie
Hi, this link takes you to a bunch of water and pump math. Some may be too simple for you but there is good stuff in there. I worked for a large water utility out west as an industrial electrician and a PLC programmer. State law was anyone who could access water supply had to have some certs, so all the electricians and programmers had to get them. I had a distribution operator III, which there was only one level higher, and a Treatment operator I Full Certification. Lots of hours studying and lots of testing. Maybe this can get you started. Hope I've helped.



https://www.dbstephens.com/wp-content/uploads/2018/08/2hr-Operator-Math.pdf
 
The reason I'm asking is that recently there was a massive pressure drop when a fire hydrant was opened, and I wouldn't mind increasing the response time of the system to prevent this, and I'm wondering if decreasing the integral coefficient wouldn't accomplish this goal.

I worked on a system that had a similar situation, though it was an entire water distribution system. The issue was that there was a corp yard where water trucks would fill up for construction and it threw off system pressures and over cycled things. Nothing destructive luckily but yes, wild drops in pressure can definitely affect an otherwise nicely tuned loop.

Few questions for you: first off, what kind of controller are you programming? Is it a PLC, VFD, etc? Second, what kind of loop is it? Proportional gain vs band, what PID topology is it using?

As for your question, yes reducing integral would slow down the response time but it may not be ideal. If you were running a PLC that was programmable, you could tackle this a few ways. First, create code that would look for a big, sudden drop in pressure. This would indicate that a fire hydrant is open. If you noticed this happen, you could program a change in the tuning values (less integral, for example). Or, you could take an average of pressures instead of the instantaneous value, and use the average to control the system so big drops wouldn't immediately disrupt your loop. That's what we did on the system I mentioned earlier, we took an average of system-wide pressures and used that for control, so if one section had a big drop it didn't affect things as much and in the end, that worked out pretty well.
 
Thanks for these updates guys. I'm going to go through the links with a pen and paper this weekend and get some stuff figured out. Given that I'm seeing so much with pumping systems lately, I need to start educating myself a bit more in-depth. At least I have the math / science under my belt, so I have some foundation to work with. My next post will include the specific VFD's, the control loop, the pipe size, setup and pump information so perhaps I can get some specific instruction there.
 
Last edited:

Similar Topics

Is it possible for me to tuning 2 servo at the same time, because this servo is connected by mechanical so I'm afraid that I need to tuning them...
Replies
3
Views
98
Have a logix controller and I'm using the PIDE block for the autotuner. I've got the hang of doing it for direct control things like pressure...
Replies
21
Views
1,669
Hello all, I am using a PowerFlex 700 (Rev 10.001) to maintain tension in a web. We're reading lbs from load cells under one of the rolls...
Replies
85
Views
11,287
Hello, I am attempting to tune a PID loop on a process. The process involves a valve with electronic actuator that has quite a high deadband...
Replies
10
Views
2,065
Hi Guys, During PID Tuning by Ziegler Nichols Closed-loop method, In TIA Portal there is a Trend Tool used to determine Ultimate Period; marking...
Replies
9
Views
1,820
Back
Top Bottom