0
\$\begingroup\$

I am designing a control strategy for an AC(induction) 3-phase motor fan in an HVAC air handling unit for pressure control inside the ductwork.

The plan is to use a PID algorithm for the pressure control, by varying the fan speed, however I am quite confused if it'll work on start-up.

If the (Setpoint-Measured pressure) is high enough on start-up, which should always be, the output fan speed will be at 100% at start. This means that the system will start the motor with a 100% motor speed(high frequency). I was wondering if starting the motor with a high frequency AC would provide enough torque in order for the motor to start moving.

So my question is if the PID algorithm is sufficient or do I need to provide a slow acceleration at start-up(starting from 50Hz) and then switching to PID when the fan is accelerated enough.

\$\endgroup\$
1
\$\begingroup\$

VFDs have various parameters of interest in your application.

  • Acceleration and deceleration parameters will give a gentle soft start and stop. Several seconds seems appropriate for you application unless you expect sudden changes in airflow. Your PID response time will have to cater for the lag this will cause.
  • Minimum and maximum frequency are programmable. You might set a minimum of 10 Hz, for example, so that you don't stall the motor.

I was wondering if starting the motor with a high frequency AC would provide enough torque in order for the motor to start moving.

Most fans have zero load (apart from friction, etc.) at startup and the power goes up, if I remember correctly, with the cube of the fan speed. The result is that there is normally no trouble starting a fan. At half speed it will draw 1/8 power but airflow will be reduced by a square law. (Check these figures out. It's been a while since I had to do any calculations.)

\$\endgroup\$
  • \$\begingroup\$ I was thinking of starting the fan with with 50Hz for 10 seconds and then switch immediately to the PID control. I guess a sudden change from 50Hz to 101Hz (which is max allowed frequency for the motor) should not present a problem according to your answer. Thank you \$\endgroup\$ – Phill Donn Aug 1 '18 at 9:17
1
\$\begingroup\$

101Hz to a motor driving a centrifugal fan is something I would not try unless the entire system is designed around that fan running at twice the rated motor speed. Centrifugal loads like fans require power at the cube of the speed change. So if the fan was designed to run at 50Hz and you run it at 100Hz, the power that the fan will want to draw from the motor will be the CUBE of the design power, i.e. if the fan requires 10kW at 50Hz, it will require 1,000kW at 100Hz!

VFDs always ramp the voltage and frequency together. Even if you give a VFD an immediate command of full speed (let's go back to 50Hz here), the VFD will always ramp into that. You can set the ramp time to be very low, as in 0.1 second, and it will ATTEMPT that, but most likely the current surge will cause the VFD to trip off line to protect itself. They are not made to do that.

What you might want to do instead is to set a "Current Limit" into the VFD so that it doesn't try even if the PID loop tells it to. Current Limit in a VFD will override the commanded speed and ramp settings in favor of maintaining the current under the limit settings.

\$\endgroup\$
  • \$\begingroup\$ The acceleration a VFD can do fine, if it's not 1:1 with the motor. It's the deceleration that it can't do without braking resistors. \$\endgroup\$ – Jeroen3 Aug 10 '18 at 19:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.