Single Input Multiple Outputs control system - how to?

Yashka

Member
Join Date
Nov 2019
Location
WV
Posts
91
I have the control system that should control two blowers by one common input flow set point. I painted the simple picture, see the attachment.
What is the best practice for such a control loop? Is it one PID? Is it two separate PIDs that have the same SP? Maybe have I do one main PID that controls two cascaded PIDs? Or something else?

2v2HzPFVex3t3c4.png


Could anyone let me links, references, articles, etc. that could clarify how to design such control systems?
 
I have the control system that should control two blowers by one common input flow set point. I painted the simple picture, see the attachment.
What is the best practice for such a control loop? Is it one PID? Is it two separate PIDs that have the same SP? Maybe have I do one main PID that controls two cascaded PIDs? Or something else?

2v2HzPFVex3t3c4.png


Could anyone let me links, references, articles, etc. that could clarify how to design such control systems?

Do the flows join back together?

Are the fans variable speed (VFD)?
 
Single PID loop.

Exactly how it's implemented depends on how you intend to run it. If it's a duty/standby type arrangement, it's pretty straightforward; one PID which feeds the speed setpoint into both fans, but only one runs at a time.

Or, you could be running both fans simultaneously, which would look the same, except your tuning parameters would be different because your system response is different with two fans vs one. In this case you'd also have to consider what might happen if one fan fails - does the other immediately speed up to compensate using some sort of bias calculation, or do you just allow the PID to gradually ramp up the speed? And then how do you handle it when the second fan comes back online? If you're going this way there are a lot of variables to consider, but I'd still do it with one PID loop.

Or is it a case that you'll usually run one fan, but if the load gets too high, you'll bring on the other fan to give you a "boost"? In this case, you could still use one PID loop, but then split the output so that 0-50% PID output corresponds to 0-100% speed on the first fan, and 50-100% corresponds to 0-100% speed on the second. So a PID output of 70% would have the first fan running at 100%, and the second fan rinning at 20%. Using this method you have to consider whether the process will behave the way you expect when one fan is running at 100% and the one beside it is running quite slowly. You also have to consider how to start and stop the second fan - if the setpoint is hovering around 49-51%, you don't want the second fan doing a rapid start-stop-start-stop cycle. You've also got to consider that if your PID output is 55%, you might not be able to run the second fan down as low as 5% long term. You might have to work something out with load sharing so that if your minimum speed is say 25%, when your PID loop is at 50% you get one fan running at 75% and the other running at 25%. Or you could implement logic that says once your PID output rises above 50%, you start both fans, and run them at the same speed, varying with the PID output.

Lots of options, depends on what you need to achieve. But in all cases I'd be using a single PID loop. Source: have done exactly this with two parallel fans and one common input airflow sensor, and a single PID loop.
 
Single PID loop.

Exactly how it's implemented depends on how you intend to run it. If it's a duty/standby type arrangement, it's pretty straightforward; one PID which feeds the speed setpoint into both fans, but only one runs at a time.

...skipped...

Lots of options, depends on what you need to achieve. But in all cases I'd be using a single PID loop. Source: have done exactly this with two parallel fans and one common input airflow sensor, and a single PID loop.
I have to use both fans simultaneously all time due to overall flow isn't be reached by one fan.
Do you mean that I have to use only one PID controller and use the output for parallel control of both drives? I mean that I have to use one the same PID output for simultaneously setting a running frequency of two separate VFDs that are controlling the fan's motors. Am I right?
 
Last edited:
I have to use both fans simultaneously all time due to overall flow isn't be reached by one fan.
Do you mean that I have to use only one PID controller and use the output for parallel control of both drives? I mean that I have to use one the same PID output for simultaneously setting a running frequency of two separate VFDs that are controlling the fan's motors. Am I right?


Use one loop, and run both fans at the same speed (1 analog output for each, but set by the PID output)


If you only run one fan and the other is not running, the air can 'circulate' from the output of fan #1 backward through fan #2 and to the input of fan #1. This does not delivery the air you want.


If both fans are running, that won't happen. It's simplest to run the fans at the same speed.
 
I have to use both fans simultaneously all time due to overall flow isn't be reached by one fan.
Do you mean that I have to use only one PID controller and use the output for parallel control of both drives? I mean that I have to use one the same PID output for simultaneously setting a running frequency of two separate VFDs that are controlling the fan's motors. Am I right?
Yes, that's correct. This will be a generally straightforward operation.

You will still have to consider what to do if one fan fails. For example, do you shut down because you can't maintain the required airflow anyway? That's the simplest option, but it's also a bit of a "nuclear" option.

Or perhaps you let one fan keep going as best it can? If you let one fan keep going, the PID loop will ramp up to 100%, then when the other fan restarts (at 100%) you may experience a high flow alarm until the PID ramps back down and gets the load under control. One way to control this might be to set the PID loop to manual if only one fan is available, and on the remaining fan, bypass the PID output and force it to run at 100%. Then when your second fan restarts, you switch back to PID control and set the loop back to automatic, and it will pick up right where it left off. You should only take this approach if you are 100% confident that it is not possible for one fan to exceed your required airflow setpoint under any circumstances.

Of course, if you choose the latter, you also have to ensure that the fans can't short-circuit, as others have mentioned (i.e. the working fan just blows air backward through the non-working fan, causing a "loop" and reducing the airflow that gets sent forward to your process. This can be achieved by non-return dampers or automated dampers - I have used both.
 

Similar Topics

hello.. I use S-315 to run plastic Extruder. There is heating zone with cooling fan to stabilize the temperature I have not found PID SIMO...
Replies
5
Views
5,698
Hello, friends, I am trying to upgrade a system that uses an Onrom incremental encoder (E6B2-CWZ6C) connected to a Danfoss VFD (FC360), but now...
Replies
4
Views
271
Hi there! There are some inverters or plc's with diferential voltage inputs. In manual says, of used as single just short - in to 0V. And it...
Replies
0
Views
1,292
My program is initiated by a N.O. input and once it is closed, it stays closed until the end of the program cycle (program is still running in PLC...
Replies
15
Views
4,129
hi, i have some question about the single phase line-to-line for the kinetix 2000. the manual only show that contractor using for the three phase...
Replies
0
Views
1,253
Back
Top Bottom