An example of very poor programming........

I work at a plant that uses 4 vertical turbine pumps to supply non-potable water around our plant.

Every 2 days the lead pump alternates.

There is 1 PID loop controlling speed to all the pumps. The PID is controlled through a pressure set point.

When pressure drops below 75 psi, a lag pump starts. But because pressure is so low the PID is telling the lead pump to run at 100%. The the lag pump starts, and it is getting its command speed from the same PID. System pressure spikes up to the high level and shuts the lag pump off. Pressure drops again and the lag pump starts right back up at 100%.

I'm the maintenance tech at the plant and have limited programming skills.

What would be a good workaround for this situation? I have thought about increasing the ramp time of the VFDs, to slow down how quickly the lag pump gets to 100%.

Could the PID be tuned to help prevent this.

I am taking a programming class this summer and would like to eventually correct the program. Any ideas on the best control philosophy?

I don't know if it possible in your case, but I would add an additional PID. (The PID loop is in a PLC or it a standalone PID controller?)

One for the lead pump and one for the lag pump. Then, you'll be able to fine tune them to the wanted operation.

But there's probably a lot more solutions available!

So I’m not the guy to be recommending ways to improve the system however I will tell you that the guys who are, are going to want to see a lot more detail I.E. which PLC’s, drives, etc…

Firejo said:

So I’m not the guy to be recommending ways to improve the system however I will tell you that the guys who are, are going to want to see a lot more detail I.E. which PLC’s, drives, etc…

Click to expand...

For sure

Instead of waiting for the pressure to drop, turn the lag pump on when the speed command reaches some high level, perhaps 80%. Turn the lag pump off when the speed command drops below some low level, perhaps 30%. Adjust the cut-in/cut-out points as necessary.

Another thing that you can do as well is to fix the lead pump speed to a certain value (when you need both pumps) and then have the PID only control the lag pump. This could work in conjunction with what Steve suggested. The goal being to bring the second pump on and offline smoothly.

when the lag pump starts, prevent it from accelarating when the pressure is above 75 psi. so the lead pump will stay 100%
another way is indeed by using another PID loop with another setpoint and slower response (I factor).

A million ways to skin this cat... I like Keshiks suggestion. I have done similar in concept (4 pumps, 1 VFD and 3 ATL) for booster pumps many times. Client wants all the flexibility of VFD's but not the $$$.

If lead = setpoint then start lag and reset PID (or move value of your choice into PID).

Simplest way I can think of:

- The output of your PID ranges from 0-100%
- From 0-50% PID output, run the lead pump from 0-100%
- From 50-100% PID output, also run the lag pump from 0-100%
- ???
- Profit!

Obviously, you'll need to do a little bit more work around it, because you don't want to start your standby pump at 0.1%, and you don't want a pump going start-stop-start-stop when the load is hovering just around 50%. Perhaps have a condition in there that a pump doesn't run until it's speed command is at least 10%, or something. You'll also potentially need to re-tune your PID a little. But I'm sure you could make it work!

Just gonna pop this in as a side-issue....

Most people believe that rotating duty (lead) and standby (lag) pumps is a good idea, balancing out their "run-hours" and their "wear".

Ultimately though, this means they will require maintenance at the same time, especially if rigid maintenance schedules are involved, sometimes imposed by the manufacturers to maintain warranties.

I have seen successful implementations of the concept of "run-duty" balancing where it is delayed for XXXX hours (and XXXX could be several months). The same pump/motor is always the lead until XXXX hours have accumulated, then the rotating duty/standby is enabled.

I have also seen this technique applied between separate lead/lag1/lag2 sets. One was a HTHW (High Temperature Hot Water) system. This had four pumps to provide flow (demand), and 3 to provide pressure (allowing the water to be at 130 deg.C). All 7 pumps were part of the algorithm to alleviate "simultaneous maintenance".

In any system, however, the lag or standby pumps should "occasionally" be used as lead, as it can give you an "early-warning" before it is really needed, and confidence that the device will perform when requested.

Just some food for thought...

Firejo said:

So I’m not the guy to be recommending ways to improve the system however I will tell you that the guys who are, are going to want to see a lot more detail I.E. which PLC’s, drives, etc…

Click to expand...

Control Logix L-7 Processor.

Altivar 71 Drives. Non-networked. 4-20 mA speed input.

Thank you everyone for your replies. I can definitely add a new PID for the lag pump.

Another option "Inching Control"

The first pump is controlled by a PID. When the output gets to 80 to 85% you start the second pump and slowly ramp it up. This will cause the first pumps PID output to go down. At about 55 to 60% stop ramping the second pump and leave it running at whatever speed its at. If the first pump PID increases again to 80 to 85% ramp up the second pump again. Continue this cycle as needed. If the PID reduces back down to 20 to 25 % then ramp down the second pump, stop ramping down when the PID increases to 55 to 60%. Continue this cycle as needed. You decide at what speed (%) to shut off the second pump.

Click to expand...

Have to ask, was the system ever working correctly?
Did some code just get modified?

ganutenator said:

Have to ask, was the system ever working correctly?
Did some code just get modified?

Click to expand...

It's a fairly new installation and as demand for water around the plant grows the problem is becoming worse.

Saturn_Europa1 said:

It's a fairly new installation and as demand for water around the plant grows the problem is becoming worse.

Click to expand...

I would push back and have them fix it... if its not paid for thats easy or if you are ever planning on using them again they should and want to make you happy, most of the time someone that made the program can fix it faster and easier then someone on the outside looking it