Re: Vfd Question
Dear Dave,
So that we understand each other, I assume the inverter is driving the motor and that the electric grid provides a line frequency only to the inverter. Switching to the electric grid and back is tricky.
If this is so then:
1. some inverters have a line synch. and transfer when the operator wishes to connect the motor to the electric grid from the inverter after accelerating or before decelerating. This process requires the inverter logic to sense the grid phase and frequency and match it so that the inverter can decouple from or recouple to the motor and allow the grid to power the motor when decoupling or inverter to power when recoupling. Special inverter logic is required. Not all inverters can do it. This can take many seconds.
The VFD can electronically be activate in nanoseconds. You do not want to do this. You can snap the motor shaft. THere are special PID loops to control torque or current rate of rise, proportional regulation based on the error signal about the target speed, or torque, and error bandwidth regulators that can be adjusted.
2. Regardless of a sych circuits you will have a problem with the decel cycle if you have a voltage source inverter (VVVF) and:
a) do not have a regenerative dissipation bank with the VVVF, or
b) you do not have 4 quadrant control with the VVVF but
c) you will not have the problem with a current source inverter matched to the motor design.
Synch circuits can take many seconds to phase lock depending on the driven equipment dynamics such as a conveyor belt with shockwave memory response.
Having said this, I don't understand your need for a synch circuit. Why do you wish to decouple the inverter?
3. Maybe a flux-vector inverter is the key to your needs. This type of inverter has a very accurate rotor position - maybe divided into 1000 divisions more or less ( dual quadrant). Talk to the manufacturer about their follower circuits and PID specifications on rates of change and signal clamping to your needs. You need to be as accurate as the inverter circuit if you wish to set the reguation with angular rates of change unless you allow them to smooth the signal. You have not stated the degree of accuracy, rates of accel and decel, forces involved wrt the nameplate, et al.
The problem is not familiar to me. We design belt conveyors with inverter drives. I may misunderstand your problem and I have already said too much.
Good Luck. ■
Re: Vfd Question
A second comment:
When the inverter is connected to the motor, the rotor position is controlled by the degree of torque generated as the rotor poles become more misaligned to the stator. Assuming a 4 pole drive running at 1800 rpm (30 rps ignoring slip) then the motor will develop maximum snap back when the rotor pole is (+/- 45 degrees) out of phase to the stator magnets. ■
Thanks For The Info.
Hi Lawrence,
Thanks very much for the reply. This was my first posting and as such I managed to place my question in the incorrect category - sorry about that. However, your response educated me on the topic of VFD.
My application is for a prototype machine that I am developing and it requires that I be able to control the motor as I described. However, based on your response I do not think I have to worry about a VFD with sync capability as I will not be sync'ing to/from the grid. In stead of coasting the rotor I could use the VFD to apply constant speed - so the cycle would be accelerate-constant-speed, and decelerate.
If I understand your comment correctly, I will need to dissipate the energy while decelerating into a load bank or whatever the manufacturer recommends - otherwise things will get hot inside I suppose.
The VFDs I've browsed on the www seem to allow a rudimentry speed curve to be programmed into the units but are limited to only a few acceleration/decel points. My application requires that I program in a specific non-linear curve or alternatively (just scheming) control the VFD externally via its analogue input port where I was hoping (perhaps dreaming) that it may be possible to apply the instantaneous control speed/torque nonlinear voltage - or this is what I was considering if I can't find a VFD to store the non-linear curve. I was planning to use a pulsed azimuth encoder of a few hundred pulse per rev to determine the angular position - perhaps the Flux Vector unit you mention would do the trick - I'll check the www.
Anyways, I've rambled too long.
Again, thanks very much for the info.
Sincerely,
David ■
Re: Vfd Question
Note, you may not wish to use a VFD on a conveyor application, however, you may wish to use an electronic softstart. The benefits of a softstart is the ability to slowly ramp up speed at startup to lessen the torque on the belt and increase the load at startup (full belt). And the softstarts are about a third the cost of VFD's and when used on motors 40HP and above they greatly lessen the motor start amps. Typically a cross the line starter uses 600% of the full load amps for startup. Example, a 100 HP motor with 125 full load amps rating will exert 6 times that or 750 amps for several seconds at startup that will either cost you if you have peak demand charges or will greatly affect the power company if you have several of these motors starting at the same time. The electronic softstarts typically only use 150% of the motor full load amps during startup and the ramp up can be 60 seconds or more for a slow increase in speed from zero. 150% of 125amps is only 187 amps at startup versus the 750 amps. My company standardly uses softstarts for all motors 50 HP or greater. You need to purchase softstarts with bypass contactor and I recommend the Allen Bradley prowerflex softstart..Have a great day
Tom ■
Re: Vfd Question
Dear Tom,
You cannot regulate the speed to the desired accuracy with a gated voltage verses the inverter'a control of frequency.
Dave,
You will find that most inverter designs have a speed/hertz follower circuit that sets the speed-time ramp to quite accurate positioning within the inverter per your external non-linear multi-part speed-time selection. This is PID or derivative type controller.
We do this all the time on large conveyors. There are typically 5 parts to our optimized (tension minimization) acceleration ramp for overland conveyors. There are linear, pause, non-linear phases in the same speed-time ramp. Often the inverter is fed the signal, we have designed, through a PLC output. ■
Re: Vfd Question
Dear Nordell,
I never said you could "regulate the speed to the desired accuracy with a gated voltage" verses the inverter'a control of frequency....
If you look again, what I said was that you may wish to use a electronic softstart inlieu of a VFD to gain the benefits of ramped startups, allowing for low starting currents, mechanical ease of low torque on belt converyor at startup, increase ability to start loaded belts, and lower cost versus VFD. What you loose if you use a softstart versus a VFD is obviously you do not have the ability to control or change the speed of the conveyor once started up. A softstart will ramp up the the line frequency of 60 HZ and run the conveyor at the same speed as a cross the line starter. I was just trying to show that you have other options of a VFD or a cross the line starter. We use these electronic softstarts on almost all are new converyor applications and hardly ever use a VFD unless we are trying to use the conveyor to control feed rate to a piece of equipment. We hardly ever use cross the line starters on conveyor unless they are very small horse power...
Have a great day ■
Re: Vfd Question
Dear T Zacharias;
Respectfully, I believe the thread starter was looking for very accurate speed regualtion as per his questions and comments of his first posting. ■
Re: Vfd Question
David,
If I uderstand your application correctly, I believe that you will need a VFD with what is called 'flying start' capability as well as dynamic braking to allow the VFD to control the speed during deceleration. You will also need an encoder for speed feedback.
Savage Services Corp.
Director--Engr. & Project Development
email: lylep@savageservices.com ■
VFD question
I am considering to use a VFD to control the accleration and deceleration of a flywheel using a 1/3 HP 3-phase motor. My application needs to accelerate, coast, and decelerate the flywheel at specific angular regions over the course of 360 degrees - and the cycle repeats.
Its my understanding that if the frequency of rotation following the coast region is different than the frequency at the start of the coast region then the VFD and motor are 'out-of-sync'.
I have three questions:
1. I understand that some VFDs have a "sync" feature that will monitor (some how ?) the change in frequency and adjust the frequency accordingly. How quickly can the VFD respond ? For example, if it takes the VFD 1 second to figure out that it's out of sync with the motor then perhaps a VFD is not the proper solution for my application.
2. If I use a VFD that does not have "sync" capability what sort of problems/dynamics should I expect ?
3. I need to vary the motor torque as a function of angular position. If I apply a control voltage from an external system via a VFD analogue input port what sort of time lag is typical for the VFD/motor to respond.
Insight into these will allow me to determine if I need to use a DC motor or AC motor.
thanks for any info. on this ...
David ■