PLC Analog Speed Control

[eight_bools]

Hey Guys,

I have a pair of machines that we are converting from potentiometer speed control to PLC control with an analog output card. PLC is a SLC 5/05 and and VFD is a PowerFlex 525.

I took some reference speeds to get the VFD frequency and parts per minute to come up with a formula. Its working okay but depending on the desired run speed in parts per minute the operator enters I am a little bit off. Not a huge deal but it would be nice if it was more accurate.

I am using the compute instruction and the formula below. Just wondering if there is a better way of doing this. I also attached a printout of the logic we used, its pretty basic. I tried searching for some sample logic online but didn't have much luck.

Max Analog Value/Max Frequency=Analog/Hz
32768/90 = 364.089

Average Parts/Hz = 0.119

(Parts/Min x 0.119) x 364.089

 

[Puddle]

Your frequencies have no decimal places. If you take more accurate readings you'll probably find that 0.119 isn't quite right. Most readings are at 0.12 except for 550 which is at 0.115.

 

[BinderNut]

Curious as to why you're not using an SCP instruction to scale the input-->output.
This will allow floating point math and should get your calculation closer to the desired result by using more than 3 decimal places (or the two places in your scalar value).

Set up your min/max input in the SCP input. You can use the parts per min for you input min/max instead of having to calculate out the parts/Hz. Direct ppm input from the HMI scaled to your VFD. Verify what max Hz is required to get your desired max line speed and set that as your max VFD speed reference.

 

[I_Automation]

If you have feedback for what the SPM actually is you could add a 1 second, or 1/2 second, timer and compare the reading to the setpoint.


If it's going a little slow add a little to the output, if too fast subtract a little.



Don't do this constantly, don't do this the first few seconds after startup or speed setpoint change, and if it's too far off you could add and subtract a larger value.

 

[eight_bools]

Curious as to why you're not using an SCP instruction to scale the input-->output.
This will allow floating point math and should get your calculation closer to the desired result by using more than 3 decimal places (or the two places in your scalar value).

Set up your min/max input in the SCP input. You can use the parts per min for you input min/max instead of having to calculate out the parts/Hz. Direct ppm input from the HMI scaled to your VFD. Verify what max Hz is required to get your desired max line speed and set that as your max VFD speed reference.

I thought about using the SCP, but didn't know enough about it. I guess I could try it out and see if it works. The speed feedback is using the analog output from the drive that someone else set up using the SCP instruction. (wish they would have just set up the HMI speed whenever this was done)

 

[I_Automation]

I thought about using the SCP, but didn't know enough about it. I guess I could try it out and see if it works. The speed feedback is using the analog output from the drive that someone else set up using the SCP instruction. (wish they would have just set up the HMI speed whenever this was done)

Write your routine using SCP and run it, just don't have it control the output yet. Every [Pick_A_Time] log the 2 values to an array.


When you get around to it look at the compare array and see if your routine has a better result, or gives wacky results because of a typo.



When your routine is running better than the original THEN replace the old with your new.

 

[Ken Moore]

90 Hz is the default setting in the sample logic for control with a SLC, is your drive actually configured for 90Hz?