Backdriving a VFD-controlled induction motor

Is there any issue with running a VFD-controlled 3-phase induction motor in torque control mode in the braking region continuously (back-driven by another motor)? The driven motor would simply be acting as a variable load on another motor. Would heat build-up in the stator and rotor be an issue? It shouldn't be re-generating, so don't see an issue with that causing VFD damage, but could be missing something here.


This application wouldn't be for a production machine or anything, just for a test device.


Thanks for any input.

I assume you want to run the motor as a controlled load for the for the driving motor
I short Yes
But the vfd would have to be setup in regeneration mode the only limit would be how are you going to get rid o the energy
I would suggest a line regeneration vfd that would give you 100% torque 100% of the time
For short term you could used a DB Resister / Buss Loader but you would be limited to a short time limit
As heating in the motor it would be n different then running a motor a full load

As long as the motor is not drawing more than rated amps and is actually turning at sufficient speed to allow cooling via the motor fan, i don't see an issue.

If it is braking then as GaryS says the VFD will need to be able to dump that energy somewhere.

For some of the dewatering centrifuges we use, the backdrive motor is actually supplied from the main VFDs DC bus terminals via fast acting semiconductor fuses. The back drive motor is maybe 15-20% the size of the main motor and only runs when the main motor is running, so this works well for feeding energy back to the main motor.

If this is something you have design control of then I would go with Saffa's suggestion. Tie the drive DC busses together through a couple of fuses and let the load motor feed the drive motor. I had this exact arrangement for a reversing gearbox life test many moons ago. It worked out pretty well.

Keith

I agree, if possible, share the DC bus with other drives and use the regenerated power as efficiently as you can.

As far as DB Resistor dumping is concerned, some drives (and some braking modules) are rated for continuous duty, so sizing the resistor becomes the issue. We have used electric duct heaters for some large loads on dynamometers where the utility would not allow line regeneration.

Thanks all for the responses.


There has been mention of regenerating in this case, so correct me if I'm wrong on the following points, as I understand them:


1) If one asynch induction motor is driving another, and both are on VFDs, if there is no stator current in the driven motor, it can be driven at any speed and will not regenerate because there is no magnetizing current. So having the driven-motor VFD simply connected to the driven motor without the IGBTs enabled/switching causes no regeneration and makes no difference.


2) Regeneration happens when the rotor is driven faster than the stator field. This could be the case when a VFD is ramping down motor speed (stator "sync" speed is reduced but the rotor speed is still faster), or this could happen in the Motor-drives-Motor case I've described if both motors are driven in the same direction by their respective VFDs, but the "driving" motor is driven faster than the driven motor's stator field. In this case, the driven motor will act as a generator and send excess current to the DC bus capacitors of its own VFD (assuming the DC buses of both VFDs are not tied together).


3) In the case I've been describing, the driven motor VFD is set to torque control, and the stator field rotation direction of the driven motor is in the opposite direction to its rotor rotation (it's being driven by another motor in speed control). I understand the driven motor to be the "braking region", where slip is > 1. I was under the impression that no regeneration takes place in this region. Is this not correct?


Finally, is the purpose of tying the DC buses of the VFDs together to send the regenerated current somewhere where it can be used immediately by the driven motor, since in VFDs without line-regeneration it would otherwise have to be dissipated elsewhere?

Jieve said:

Finally, is the purpose of tying the DC buses of the VFDs together to send the regenerated current somewhere where it can be used immediately by the driven motor, since in VFDs without line-regeneration it would otherwise have to be dissipated elsewhere?

Click to expand...

I believe the answer is yes. I once worked on some machines that had large material "let-off" drives where there were two motors, one driving the web take up roller that pulled stock from the supply roller which held the stock wound up together with a fabric web. The letoff drives were all tied to a common DC bus and the supply roller drives were set up in torque mode so that they could produce a constant tension on the material while allowing the take up roll to pull on them. They generated power back to the common DC bus that was immediately useable by the other drives.

There were 4 drives on this system, two pairs as described above, one online, and one standby.

There was also a large dynamic brake set up on that system so that any excess DC bus voltage could be dissipated. I noticed that the only time there was any activity on the braking module was when the standby rollers were being manually operated and decelerating while the online feeder was also running.

I don't know how much that arrangement reduced the power consumption of the system as opposed to one with no common bus and 4 braking modules, but I expect that it was quite significant.

If you don't have a Line Regenerative VFD for the "Load" motor, but it has DC Bus terminals, you can probably just toss on a line regen Bonitron unit.



I've been using them on things from 5HP up to 300HP with zero issues.

Wow where to start
Nobody here can give you any real information or help on this project there is just not enough information provided.
As for connecting the vfd dc buses together that referred to a common buss configuration it may be the way to go on but as I said we just don’t have enough information. This is not a new thing it’d been around for about 30 years or more. I have actually done quite a few common buss systems using Powertec drives before you could even consider it a vfd.
But understand this when using common buss configuration only one vfd or drive can supply the power to the full buss. In some cases that master vfd must be larger than in necessary for the motor it’s powering. The reason for this is that the buss of different vfd’s will charge at a different rate even if they are the same size on the same power supply, so you could see buss fuses blow, power sections blowout. There are many components in the buss power section that all play a part in the charging and discharging of that buss then you still need either a line regeneration module, line regeneration vfd or a buss loader module ( Braking module ) to safely dump any excess energy in the buss it gets more complicated then you what most think.
On some vfd’s you can actually connect them to a externally source DC buss example, a 500V DC buss in an old mill ( If memory serves me I think I may have done that years ago ).
Also keep in mind, not all vfd’s will allow you to do common buss configuration the Yaskawa U1000 Matrix drive can’t be connected common buss ( it does not have a DC Buss )

As for how the system works I will try and offer some explanation but you must first understand how a motor works all motor can be used a generator if you drive the motor and pull power from it.
Well forget for the moment about how the master motor is powered and controlled. The regen motor is coupled directly to the master motor it will be considered a generator you can control how much load in on the system by controlling the amount of current you draw from the generator the moe current you draw from the regen motor the more load on the master motor. You can do this with a vfd capable of regeneration. The key is dumping that energy you can dump the energy as heat usinf buss loader resisters or you can line regenerate it back into the ac power line both have advantages and disadvantages. Most vfd’s dump the energy through a resistor on the dc buss but to dump 100 % you will need an external buss loader module and resistor. Then best I have seen on a internal buss loader is 10% braking power without the external buss loaded so you must consider your hardware needs carefully or you will not get the results you hope for.
As for motor heating both motor would generate about the same amount of heat and it should be easily dissipated by the motors as they rotate. That is if everything is sized correctly.
I will leave you with something to think about ( it was passed to me many years ago by a very good engineer when I just starting out )
“ Man does not have the power to either create or destroy energy we can only change its form “
I have found this to be absolutely true and before you start nay saying think about it all the energy we use was already here long before man was on this earth and it will still be here long after man is gone from earth and we are only just beginning to understand it and use it.

Thanks all for the responses.


Glad to see that this has been done for ages and it's a common practice for certain applications.


GaryS, I appreciate the input. I hear you on not being able to provide a specific answer without more specifics. This post was simply to find out if it is doable in general, which it seems to be. I will look more into the common bus configuration.



The VFDs that will likely be used are Rockwell Powerflex 525 and ABB ACS355, both at least have DC bus connections. The motors aren't large, 2hp. But will need to dig deeper to see if it will work here specifically.

I just have to add one thing here
I would never ever mix manufactures / brands of vfd’s in a common buss configuration
In this case you said mixing Powerflex and ABB while the Powerflex may well be an ABB I know they outsource their vfd’s.
There may be a slight a difference in the buss voltage profiles, that difference could lead to major problems, blown fuses, blown power modules, blown buss loader module as well as others

As I stated you may not even need to do the common buss configuration or even need second motor-vfd to get the results you want one vfd may be able to do it all with a much simpler control system.
If you have a choice I would recommend a Yaskawa U1000 Matrix drive as your load vfd, they don’t have a dc buss they are line regenerate units so they can give you 100% torque / load in both forward powering and reverse regeneration.
When you do figure out what you need and get it working please follow up here and let us know how you made out so we can all learn something from it.

Revisiting this post, as this project is once again current.



Basically I've got two similar test machines; one is using 2x AB Powerflex 525 drives, and one is using 2x ABB ACS355 drives. In both cases, one induction motor acts as a load on another. I'll stick to the Rockwell setup as that is the one I'm on currently. All motors and drives are 2hp.


The rockwell drives are set up in a Shared AC/DC bus configuration. The configuration follows the recommendations in Rockwell's document on Common DC Bus applications (DRIVES-AT002). Both mains inputs are fused w/ class CC fuses, and both use line reactors. The DC buses are connected via short cable fused w/ class J fuses. The system interfaces with a MicroLogix 850 as control.


So now I have a couple questions:
1) Rockwell recommends putting both VFDs on a single disconnect, as in case mains fuses blow to either drive, the DC bus of the powered VFD is still powering the other drive/motor. In this specific equipment, both VFDs are the same size, and are not oversized for multiple motors. What would be some recommended ways to accomplish shutting down the "fuse-blown" VFD? Things that come to mind are: a) place a contactor between driven connections in the DC bus, b) place a contactor or shunt trip breaker in the main line supplying both VFDs. Pros and Cons?



2) Then the next question becomes, since CC fuses are used in the mains and there is no feedback to indicate anything is blown, how should the contactor/breaker signal be triggered? It seems the powerflex 525 drives fault on phase loss, so maybe a digital output could be used to signal the PLC, and the PLC could be used to switch the breaker/contactor. However, haven't dug deeper (yet) into how this works in a Common (Shared) DC setup, but I'm not sure exactly how it works, since in the shared DC bus case the drive wouldn't normally even have an AC input at all, but still needs to run without faulting.


3) Finally, currently there is no braking module set up. The idea is to only enable the driven motor (generator) when the driving motor is running, and the assumption is that any current generated should more or less be used by the driving motor immediately. My gut says this may not be a great assumption, but I figured it would be good to test out first.


Any input on these questions would be appreciated as always.

First let me say this I have done Common Buss configuration many times going back to Powertec in the early 90’s and doing a common buss configuration on a vfd was not even possible then.

Let’s clear up the terms there is no AC/DC common Buss. Common Buss by definition is DC.
I would not recommend powering both vfd’s, there should only be 1 power source. The reason for this is they there in no way to ensure the charge rate to the buss will be the same on both vfd’s even if they are the same size the components may vary a little and cause an over load or fault at any time.
It is possible to connect multiple vfd’s on a common buss but there still should only have 1 power source.
You need to install a fuse from the buss to each vfd buss terminal to protect each vfd.
You should also consider a buss loader module and resister to dump the excess buss energy if necessary
Otherwise you may get a buss overvoltage fault.
I would not recommend using a breaker anywhere on the buss connections, remember the buss voltage will be about 700VDC most breakers are rated for 500VAC or 600VAC max. While the breaker on an AC voltage will be able to break the fault current but DC will arc across and could destroy either or both vfd’s. I have seen this to many times where it was tried and failed destroying things. You have to keep in mind that when the breaker needs to trip and open the breaker load it is already be above the load rating of the breaker. I have seen many breakers where the contacts weld closed.
Do not place a contractor at any point on the buss if you try to open it under load it could weld the contact closed and here again burn up components.
A single breaker feeding both vfd’s should work fine that way when one switch disconnects everything
If the only load on the system is from the motoring vfd then a buss loader may not be needed. But you should still consider installing a Buss loader resister (DB Brake Resistor) on each vfd to prevent buss overvoltage trips.


I would recommend a Yaskawa vfd for this project they have an advantage here the power to controls board comes from the buss. The ac line powers the buss the buss powers the vfd controls so as long as there is power on the buss the control will have power and control the vfd
Depending on your application the micro 850 plc may not be needed I would need more details on that.

I hope this helps

Thanks for this response.

LetÂ’s clear up the terms there is no AC/DC common Buss. Common Buss by definition is DC.

Click to expand...

In their manual on common bus applications, Rockwell calls it a "Shared AC/DC Bus", which is why I called it that.

I would not recommend powering both vfdÂ’s, there should only be 1 power source. The reason for this is they there in no way to ensure the charge rate to the buss will be the same on both vfdÂ’s even if they are the same size the components may vary a little and cause an over load or fault at any time.
It is possible to connect multiple vfdÂ’s on a common buss but there still should only have 1 power source.

Click to expand...

Rockwell makes the point you're making here, but stops short of saying it shouldn't be done, and instead shows their recommended way to do it. For the 700 series power flex drives they recommend back-back diodes between drives (from Bonitron) in the DC bus to add an additional voltage drop, to overcome the issue of having different impedances at the VFD input causing overloading of the rectifier section of one of the drives. For the 500 series drives, they recommend line reactors on both drives instead of the diodes; I'm guessing to reduce the input impedence difference between drives, and/or reduce the reactive current at the drive input.

I would not recommend using a breaker anywhere on the buss connections, remember the buss voltage will be about 700VDC most breakers are rated for 500VAC or 600VAC max. While the breaker on an AC voltage will be able to break the fault current but DC will arc across and could destroy either or both vfdÂ’s. I have seen this to many times where it was tried and failed destroying things. You have to keep in mind that when the breaker needs to trip and open the breaker load it is already be above the load rating of the breaker. I have seen many breakers where the contacts weld closed.
Do not place a contractor at any point on the buss if you try to open it under load it could weld the contact closed and here again burn up components.

Click to expand...

This is a good point; in this case though the VFDs are on 208V, making the DC Bus voltage around 300V. The braking chopper switches on at 395VDC.

A single breaker feeding both vfdÂ’s should work fine that way when one switch disconnects everything

Click to expand...

This is what Rockwell recommends also (actually they just recommend a common disconnect), and works for shut down during maintenance, but it doesn't exactly solve the problem where the mains fuses on one drive blow and the other doesn't, unless this is somehow communicated to the breaker (via a shunt trip). Is that what you meant here? Or how did you envision this working if one of the drive fuses blow? This is sorta the problem I'm trying to figure out, how to shut down the other VFD if fuses on one blow.

But you should still consider installing a Buss loader resister (DB Brake Resistor) on each vfd to prevent buss overvoltage trips.

Click to expand...

I think you're probably right on this.

I would recommend a Yaskawa vfd for this project they have an advantage here the power to controls board comes from the buss. The ac line powers the buss the buss powers the vfd controls so as long as there is power on the buss the control will have power and control the vfd

Click to expand...

Although I haven't tested it yet (I will), I would imagine this is the case for powerflex 52X drives as well, as Rockwell does give examples of how to set up multiple drives with a single rectifier source. If the common bus didn't power the other drive, then controlling the additional motors wouldn't be possible. Additionally, we already have the drives, reactors, etc., so would prefer not to switch models at this point. But will keep that in mind for next time.

Depending on your application the micro 850 plc may not be needed I would need more details on that.

Click to expand...

Most likely this isn't needed for VFD control, but I wanted to interface a PanelView with it to read out DC bus trends, and there will be other features of this system that will eventually require PLC control, assuming everything works correctly. Also the drives will be enclosure mounted, so easier to control everything from a single panel.



So how would you recommend handling the situation where fuses blow to one of the drives, and it's desired to shut down the other?

Wow
I really didn’t think I would have to reply again I just want to offer some help if I can.
So here goes
I think before you get in way over your head you take a course in basic electricity
Particularly DC current
Again I say there is no AC Common Buss unless you think the AC supply line is a common buss. Here all of the equipment on the supply line a on a common system. Think about it.
DC Common Buss has been around since electricity was introduced although it may not have been recognized.

I think in that manual the person that wrote it just didn’t understand it and made a simple mistake. But the people reading it need to fully understand what they are doing. That’s the only reason I brought it up.

As for the Back Back Diodes you really need to read and understand how they are using them, and they are not used in the same application as yours. In their application they are using them with a Regen unit to dump the buss energy back in the ac line. In your application you want to reuse the energy in the motoring vfd so is not going back into the ac line.

As for the breaker on the dc buss. It’s not the voltage that hurts you is the arc current when it has to open. The larger the arc current the harder it is to break the arc.
On an ac system the current naturally goes to zero every 8 ms ( Base on 60hz system) so even closed the current through the breaker contacts will go to zero every 8ms and that works to naturally limit the arc
On pure dc ( As in a DC Common Buss ) that doesn’t happen the only thing that clears the arch is distance, even at 350vdc the distance opening arc can be several inches depending on the amount of current passing through it at the time you open the contacts. You wouldn’t believe the number of service calls I have gotten because the original engineer didn’t understand the difference between pure dc as in a dc buss and rectified dc as used in a dc drive system. The rectified dc system still passes through zero every 8ms naturally opening the current while with the pure dc the current will maintain until the arc distance is to long to sustain the arc if it can. Sometimes it will destroy the breaker, a fuse will always open at some point.

As for the blown fuse indicator I really don’t see the problem there either
I would use a breaker on the line side of both vfd’s so you will never have a single incoming line open when the breaker trips all 3 lines open, the vfd will have no power, the vfd just shuts down and the motor coast to a stop. No problem there
If the line breaker on the regen vfd opens ( a very unlikely given the only load on that breaker is the control power). There will be no load on the system / motoring vfd so that motoring motor will just run at speed with no load.
If the dc bus buss opens them the there is no way to transfer the energy from the regen vfd to get to the motoring vfd so the buss voltage on the regen vfd will go up until it trips the vfd fault on over voltage.
Or if you did is I as I said before and installed a buss loader resister on the vfd then it will just dump the energy to loader resister and into the air as heat. Here again in a short time the vfd will fault on an overvoltage fault ( Most internal vfd breaking transistors are only rated at 10% drive capacity )

If you still think you need to check for open fuses then I think the only fuses you need to monitor are the buss fuss. Therein lies the challenge as the voltage on both sides of the fuse when open will still be relatively close. So I think you would need to monitor the buss voltage on both sided of the fuse and if that voltage is greater the about 20 to 25 volts then that would indicate an open fuse. As the voltage on the motoring buss would go down and the generating side would go up to the brake threshold.


Here again when you consider all of the hardware you need and the wiring to connect it together I would look at the new matrix drive from Yaskawa there is no buss at all to deal with, they are 100% line regeneration as they come from the factory.
So just power up both drives and set them up done no add on breaker, resistors, contactors, fuses and no additional labor to put it all together.

As for the micro 850 if you just want to monitor the parameters of the system I would consider a standard pc, that way you can monitor, display and log the data as you want. You can also control the system through it as well. With direct connection to the vfd’s.

I hope this helps you to understand thing a little better

If you want into get into more details with you project PM me and I will see what I can do to help