Motor Regeneration with VFD

[Ranch]

080717-1205 EST
The post by Ranch did not clarify anything with the use of VSI since we really do not know what this means.

If the Ranch wasn’t so frickin busy Ranch would be all over this post. I have not read anything except for portions of Gar, Jraef, and Minnie …. As I should.

Keep it going though – the typical VSI all to familiar with name types such as VFD, Freak drive and so on and that that VSI predominantly has taken on two different familiar forms, the variable speed drive and the UPS

Theory understood in the forties, tried in the fifties, marketable in the sixties, hated by the seventies, honed in the eighties, and finally achieved the reliability required to become ubiquitous in motor control in the nineties once the voltage controlled Mosfet embedded gate IGBT replaced the terrible hogging current controlled darlingtin pairs …. my ears still ring.

So on that note, the IGBT won't survive inductive "kick back" and we need a free wheel. That's what makes your OV when the front side of the VFD is uni-directional.

I’m tired but still want the OP to tell me what his or her application is.

Did anyone bite on my analogy between induction motor and induction generator comment?

 

[mull982]

I apoligize I've been out of town and havent been able to comment or answer questions.

My application is a tripper car which has two 5hp drives to run the tripper car up and down a set of rails. Right now we are just using a reversable starter but are thinking about putting VFD's to control these drive motors and the issue of the tripper car presenting a high inertia load and thus regeneration has come up.

So basically the two drive motors will be started in one direction, it will get to the end of the track, and then reverse to head back in the other direction after pausing for about 5sec or so. Right now This constant on/off and reversing of direction is hammering on our starter. I'm thinking we would probalby need some sort of decelleration.

 

[Jraef]

If it is a big tripper car without a lot of friction from the belt (depends on the design), then a regenerative VFD may be the better way to go in the long run. The use of DB resistors will be problematic because I can imagine this is a very dusty dirty environment and the resistors would load up with dirt quickly, accelerating their demise. Just make sure you have a cooling system for the VFD or locate it far enough away to prevent the same fate for it.

 

[jim dungar]

You may also want to consider oversizing the drives as much as possible (check with your manufacturer first, some drives require a minimum size motor). Make sure that your VFD is for a constant torque device rather than one of the more common, and less expensive, variable torque units.

 

[Loffgren]

RE

RE

Increase the DECEL TIME.

 

[mull982]

If it is a big tripper car without a lot of friction from the belt (depends on the design), then a regenerative VFD may be the better way to go in the long run. The use of DB resistors will be problematic because I can imagine this is a very dusty dirty environment and the resistors would load up with dirt quickly, accelerating their demise. Just make sure you have a cooling system for the VFD or locate it far enough away to prevent the same fate for it.

I'm confused to exactly what a regenerative VFD is and how this differs from a regular VFD with a DB resistor. The person in charge of this project was talking about mounting the VFD right on the tripper car itself saying that this would cut down on regen by being closer, and was the reason for my origonal post. So you are saying to keep the VFD far away from this enviornment such as in an air conditioned electrical room?

You may also want to consider oversizing the drives as much as possible (check with your manufacturer first, some drives require a minimum size motor). Make sure that your VFD is for a constant torque device rather than one of the more common, and less expensive, variable torque units.

What is the difference between a constant torque drive and a variable torque drive?

 

[petersonra]

I'm confused to exactly what a regenerative VFD is and how this differs from a regular VFD with a DB resistor. The person in charge of this project was talking about mounting the VFD right on the tripper car itself saying that this would cut down on regen by being closer, and was the reason for my origonal post. So you are saying to keep the VFD far away from this enviornment such as in an air conditioned electrical room?

When you try to stop the motors, the kinetic energy in the system the motor is driving has to go somewhere. The motor will typically turn into a generator and back feed the energy to the DC bus on the VFD. A VFD with a DB resistor will take that energy and use it up as heat dissipated across the resistor. A regen type VFD will take the surplus energy and put it back on the AC line for use elsewhere.

He is mistaken about the reason for mounting the VFD close. It has no effect whatsoever on regen (except the tiny amount of energy burned up as I^2R losses in the extra wire). Some VFDs exhibit harmonic reflection on longer runs of wire so they suggest shorter runs if possible. There are other ways to handle this problem if shorter wiring is not practical.

I would always suggest putting a VFD somewhere cooler over somewhere warmer. It not only extends the life of the drive it makes it a lot more convenient to work on.

 

[jim dungar]

What is the difference between a constant torque drive and a variable torque drive?

Keeping it very simple.

Constant torque devices need to perform the same amount of "work" regardless of their speed.

Variable torque devices expect to due less work as their speed is reduced (this is why we can get energy savings from pumps and fans that don't need to turn at full speed).

Component wise, for a given HP, a constant torque device is usually "beefier" than an equivalent variable torque.

 

[weressl]

I have had issues with several of my VFD applications in the plant where they were tripping due to regenerative voltage feedback from the motor with some high inertia loads. I have looked into several solutions including braking resistors, dc braking etc... Someone has recently told me that by putting the VFD closer to the actual motor this would help alleviate some of the regeneration. I'm not sure that I buy this because I cannot see why this is true. I would tend to think just the opposite that with the dirve further from the motor the regen voltage would be less simply due to the resistance in the cable and associate voltage drop. Can anyone comment on this?

The same person also mentioned that using VFD cable would help alleviate the regen voltage. I always understood that the VFD cable was a cable simply to help with heating due to the harmonics and irregular waveforms from the dirve. Can VFD cable help eliminate regen voltage?

Besides of the breaking resistor or DC injection breaking parameter change the changing of the stopping ramp time will help. Some of the ASD's - the better ones - no longer care about regenerative voltage if you just configure stopping without ramping. (It can only regenerate voltage if you keep supplying voltage to it.)

 

[petersonra]

Keeping it very simple.

Constant torque devices need to perform the same amount of "work" regardless of their speed.

Variable torque devices expect to due less work as their speed is reduced (this is why we can get energy savings from pumps and fans that don't need to turn at full speed).

Component wise, for a given HP, a constant torque device is usually "beefier" than an equivalent variable torque.

some lines actually have two ratings for each model. one for variable torque and one for constant torque.

 

[weressl]

I have had issues with several of my VFD applications in the plant where they were tripping due to regenerative voltage feedback from the motor with some high inertia loads. I have looked into several solutions including braking resistors, dc braking etc... Someone has recently told me that by putting the VFD closer to the actual motor this would help alleviate some of the regeneration. I'm not sure that I buy this because I cannot see why this is true. I would tend to think just the opposite that with the dirve further from the motor the regen voltage would be less simply due to the resistance in the cable and associate voltage drop. Can anyone comment on this?

The same person also mentioned that using VFD cable would help alleviate the regen voltage. I always understood that the VFD cable was a cable simply to help with heating due to the harmonics and irregular waveforms from the dirve. Can VFD cable help eliminate regen voltage?

Another way of avoiding the issue is to put an electromechanical break on the motor shaft that is actuated by the stop signal. This is useful if you want a short and positive stop of the rotating machinery.

 

[jcormack]

Tripper car

Tripper car

I have put VFD's on five applications with tripper cars. They work very well. You have to provide units with either full regen drive capability or large "dynamic" resistor banks to shunt energy from the DC Buss. Also, as others have stated you need "Constant Torque - Flux Vector" units. As the tripper moves in one direction it is against the tension of the belt & is motoring - in the opposite direction the belt tension is pulling it - so it is in constant regen in that direction. Braking resistor would need to be sized from 75% to 100 % capacity.

On a normal VSI PWM VFD (gotem all in there) you have a basic retifier on the front end that powers a DC buss (capapitors + link choke inductor) This DC is then "modulated" by MOSFET IGBTs switching on/off at a high frequency (carrier freq or "tone" in drive parameters) to derive a synthesized waveform - rotating the motor at different speeds as required. The normal rectifer front-end does not have the capabilty to send energy "back" to the incoming source. Hence the need for a resistor to dump energy off the DC Buss or an "active" front end that can put energy back into the power grid. An active front end may be SCR based or IGBT based - most simply use another IGBT power section as that is their base product. I have had very good sucess with ABB's regen VFD's in tripper car applications. I personally do not like to have the VFD on the tripper - why expose it to the vibration ? That being said, I do have two installations where the drives are on the trippers, it will work that way. You can make a case that having the imp of the cable between the VFD and the motor will help limit the regen current's dI/dT. You also have to decide how you want to set the drives up to loadshare - two motors on one drive, two drives not connected, or one drive speed master the other a torque slave - you hve to be carful here or you may have the tripper "racking" if the two motors are on seperate drive chains or rack & pinion - if they drive a common shaft then some of these issues go away.

Give Ranch a break - he used to build VFDs for a major Manuf. (and he's Canadian !)

 

[jcormack]

That won't work in his application. When the Tripper is moving in one direction it will be in Regen, as the feed belt tension is "pulling" it. In the other direction it is motoring.

 

[cschmid]

I have put VFD's on five applications with tripper cars. They work very well. You have to provide units with either full regen drive capability or large "dynamic" resistor banks to shunt energy from the DC Buss. Also, as others have stated you need "Constant Torque - Flux Vector" units. As the tripper moves in one direction it is against the tension of the belt & is motoring - in the opposite direction the belt tension is pulling it - so it is in constant regen in that direction. Braking resistor would need to be sized from 75% to 100 % capacity.

On a normal VSI PWM VFD (gotem all in there) you have a basic retifier on the front end that powers a DC buss (capapitors + link choke inductor) This DC is then "modulated" by MOSFET IGBTs switching on/off at a high frequency (carrier freq or "tone" in drive parameters) to derive a synthesized waveform - rotating the motor at different speeds as required. The normal rectifer front-end does not have the capabilty to send energy "back" to the incoming source. Hence the need for a resistor to dump energy off the DC Buss or an "active" front end that can put energy back into the power grid. An active front end may be SCR based or IGBT based - most simply use another IGBT power section as that is their base product. I have had very good sucess with ABB's regen VFD's in tripper car applications. I personally do not like to have the VFD on the tripper - why expose it to the vibration ? That being said, I do have two installations where the drives are on the trippers, it will work that way. You can make a case that having the imp of the cable between the VFD and the motor will help limit the regen current's dI/dT. You also have to decide how you want to set the drives up to loadshare - two motors on one drive, two drives not connected, or one drive speed master the other a torque slave - you hve to be carful here or you may have the tripper "racking" if the two motors are on seperate drive chains or rack & pinion - if they drive a common shaft then some of these issues go away.

Give Ranch a break - he used to build VFDs for a major Manuf. (and he's Canadian !)

So what does being canadian have to do with it..Nice to know more about ranch thank you..

 

[Ranch]

Still Alive?

Still Alive?

Gosh – guess summer is over, where did it go?

Hope this last thread I left is still alive – love the old school ”increase the decel comments” – just coast already or turn your braking torque / regen limits down to zero. Perhaps turn your slip comp off ……

May I curve this thread to a power quality theme?

Yesterday, I really got to put the screws to a system I really despise, 18 pulse VFD (ouch – an odd times six) to test if an XFMR manufacturer met my spec – initially +/- 3% balance on nine legs throughout the load range, then I relaxed to +/-5%. Not even close*

Perhaps not in the general intent of Mr. Holt’s web site but I am compelled to share with all just what an abomination the commonly specified 18 pulse municipal market (our money) VFD systems really look like.

I will attempt to attach an image. If successful, note all the numbers are percentages or amperages. This is the reality of 18 pulse ….. utter garbage. Quit specifying it and quit supplying it. I am thoroughly exhausted when as a tax payer we shell out for such technical stupidity. We don’t seem able to control the price at the pump but we can control this.

Any takers, I’ll explain …. Tomorrow perhaps - *bottom line is this data shows a transformer unbalance from +115% / -22% balance from no to full load.

 

[Rockyd]

bookmarked