"Braking" with GE VFD

This is where I am now confused. We have used VFD's on exhaust fans that often take several minutes to coast to stop, as in Tom's case. These are used for PID control of pressure (inches of water). If the PID calls for the VFD to slow down the fan to get to setpoint, are you saying that it is impossible to slow the fan down at any rate faster than what normal coast would be, without braking? I've seen many applications like this with VFD's, and no braking, work just fine..I just really never thought about how it could be doing it. Maybe we should all individually call GE support and see what different answers we get.

Marty

Edit: I posted this before reading DickDV's last post..that was very informative, thanks!
 
Last edited:
I am still confused.

I just came from taking a test in German so my head is crammed with what it could hold on that subject.

Anyway I am not sure I understand all the issues therefore I am going to provide information as I understand it and then see if it is correct. If it is correct and the GE drive in question can do dc injection/compound braking then it seems that this drive doesnt have the capabilities of other drives Tom may have used.

AC drive using single quadrant:
Coast to Stop
To stop an AC motor in single-quadrant operation voltage and frequency can simply be removed and the motor allowed to coast to a stop.


Controlled Deceleration
Another way is to use a controlled deceleration. Voltage and frequency are reduced gradually until the motor is at stop. The amount of time required to stop a motor depends on the inertia of the motor and connected load. The more inertia the longer it will take to stop.

DC Injection Braking
The DC injection braking mode stops the rotating magnetic field and applies a constant DC voltage to the motor windings, helping stop the motor. Up to 250% of the motor’s rated current can be applied.
Compound Braking
Compound braking uses a combination of the controlled deceleration ramp and DC injection braking. The drive monitors bus voltage during operation and triggers compound braking when the bus exceeds a set threshold point. As the motor decelerates to a stop a DC voltage is periodically applied to the motor windings. The excess energy on the bus is dissipated in the motor windings.



Its my understanding the above should NOT be used in applications that are required to stop/start repetitively in short time periods.​

Its my understanding that if you are using a drive in 2 or 4 quandrant mode that the motor regenerated voltage can add to the DC bus link and cause bus overvoltage trips. In these modes you need to use braking resistors to remove the excess voltage.
Ok what did I get wrong state improperly or leave out?

 
The four quadrants are defined as:

Forward direction, forward torque (drive)
Forward direction, reverse torque (braking)
Reverse direction, reverse torque (drive)
Reverse direction, forward torque (braking)

In a DC SCR system, one needed two bridges to accomplish the four quadrants, if regeneration were to be the braking method. Contactors or reversing field supplies could be used to "switch" the leads to make one bridge behave like the two, but that's just engineering.

One can easily see that with three phase AC motor drive systems, the non-braking forward and reverse directions are easily accomplished with switching changes in the inverter. There is really no difference between these two quadrants other than "rolling the motor leads" electronically.

It is in the braking quadrants that we have problems (either direction). We draw energy from the line and provide it to the motor to inject energy into the motor when in the "drive" mode. Where do we put the energy from the motor when in the "brake" mode? Without a front end to invert the power back onto the line, the only answer is "dissipate it as heat."

Where we dissipate it as heat is the question. Your examples all dissipate the energy as heat in the rotor, or allow the inherent losses in the system to achieve the deceleration as in the coast stop.

Controlled deceleration just has the drive keep the motor fluxed during deceleration, allowing the drive to monitor the motor without the need for an encoder. Without maintaining rotor flux, the drive would have no other way of determining the motor's direction or speed. With no rotor flux, the drive would see zero motor Volts and zero Hertz, i.e., a stopped motor, even if it were actually still rolling. The drive could then conceivably plug the motor with very low frequencies trying to "start" the already rolling motor. This can cause not just heat in the motor, but can actually damage the motor or shear its coupling while fluxing the rotor. Reverse torque can be applied while trying to "start" a rolling, defluxed motor, causing severe torsional stresses. I once saw a gas injection fan (900 Hp, 4160V) shear its coupling in such a condition.

What we actually see as voltage on the motor is the back/counter EMF (CEMF) from the motor's regeneration. A motor has essentially zero resistance in its windings. Without CEMF, a motor would pull tens of thousands of amps at 460V excitation. The motor's windings actually see the difference between the applied voltage and the regenerated voltage (CEMF), and this is what mitigates the current draw.

Therefore, a motor must have rotor flux and stator flux to generate torque, and this rotating rotor flux generates a voltage in the stator that is opposite to the applied stator voltage. If you apply a torque opposite to the motor's direction, then the motor will stop more quickly (brake), but the energy that has been removed must be present in some form still.

So, to answer your question, you got nothing wrong and left nothing out, that I can see.

The old SCR switching AC drives had no way to convert the motor energy (without additional hardware). They typically used controlled deceleration to keep an "eye" on the motor during deceleration. The controlled deceleration essentially just provided magnetizing current, but no torque producing current.

If an LC filter were hung off the old-style drive, between the inverter and the motor, then the capacitors in the filter were prone to pump up during deceleration, and had to be monitored and protected. Without an output filter and without controlled deceleration, then the motor would just de-flux, and coast stop. This was because the motor had no path for its regenerated EMF and therefore no current would flow. Because no current was flowing, the magnetic fields would collapse.

Injection braking causes rotor heat. The amount of inertia will determine how much heat the rotor absorbs, and eventually dissipates, during deceleration. One can easily see that injection braking at a rate faster or more often than the heat can be dissipated will eventually overheat the motor.

I feel like I've taken over this thread. I'll stop now. Sorry for the long posts. I'm trying to illustrate a concept using words, without pictures or other illustrations, and that always takes more verbage than normal. I fear I'm causing more confusion than illumination, so will desist.

Regards,
Don
 
It's been fascinating following this Thread from the sidelines. I'm a process control engineer, not an electrician and while I know my way around electronics pretty well I have never studied and have only a basic knowledge of what I regard as "heavy" electrical stuff.

I've learned a lot in this Thread and who knows, one of these days, some of it may come in useful on some far-flung factory floor!
 
Oh, boy! This has gotten to be quite the discussion. Don's post above has lots of good points.

I think that DC Injection braking should be separate from this subject because it is completely different from the other types. DC injection braking simply shoots a DC current into two windings of an otherwise de-energized spinning motor. The rotor magnetizes itself and resists rotation until rotation stops. Note that there is nothing synchronous about this braking. It simply causes resistance to rotation. Note also that DC braking cannot be used for going from a high speed to a slow speed. It can only be used for stopping--the motor has to be otherwise de-energized. The braking energy accumulates as heat in the motor and is passed to the surrounding air by the motor's cooling fan.

What Ron is calling Compound Braking sounds a lot like Flux Braking. This type of braking is Synchronous and can be used for decel between two speeds (motor energized) or for stopping. The braking energy is also wasted in the motor.

As for snubber or dynamic braking, (regen braking is just snubber braking with the energy returned to the incoming supply lines instead of being wasted as heat in a resistor), the drive continues to magnetize the stator field at a specific frequency. If the motor is connected to a load, then the rotor slips down in speed (negative slip if you will) in proportion to the magnitude of the load torque and the motor requires current from the drive to accomplish this. On the other hand, if the motor rotor is connected to something that is forcing it to turn faster than the stator field as in braking (positive slip) then the motor produces electrical energy that is sent back up the supply lines. This returning energy passes thru the bypass diodes on the IGBT's and charges up the DC link voltage. Unless there is some provision to get rid of this extra energy (a standard drive normally has none), the bus voltage will rise until its upper limit is reached and the drive faults. When faulted, the drive removes the excitation from the stator and the motor freewheels along with the load to a stop.

As mentioned before, there are only two common ways of getting this excess energy off the DC bus and those are a braking resistor or regenerative braking.

As for duty cycle, basically you get what you pay for. If you only stop once in 4 hours, you can pour up to 10 times a resistors rated wattage into it for a few seconds without harm. On the other hand, if the braking is to be continuous (that usually means more than 10 minutes), then the resistor wattage has to be much higher.

When you buy a regenerative or four-quadrant drive, you will almost always get a drive with equal motoring and braking horsepower. But it doesn't have to be that way. You could, for example, buy a conventional two-quadrant 100hp drive and add a 10hp Bonitron regen braking package to it for 100hp motoring and 10hp continuous regen braking. But the principal is the same regardless of sizing.

As for the motor, motoring at a certain level and regen or snubber braking at that same level results in about the same amount of motor heat either way. The braking energy is sent away so it doesn't appear at the motor.
 
Tom, I've been doing some heavy thinking on your faster decel rates on fans and may have a possible explanation.

Since braking hp is equal to torque times rpm divided by 5250 just the same as motoring hp, and since the horsepower on a fan is a function of the cube of the speed, it follows that a fan will decellerate quickly from top speed down to maybe 25% speed and then take forever to coast the rest of the way down because there is no air load and the bearings have very little drag.

Now, if the internal drive losses are, let's say 5% of the rated hp, this 5% is available as braking. From max speed to slow speed, the fan comes down very quickly because of air load and the 5% motor braking is not a significant part of the total. But, as the fan turns slower and air load goes away, the 5% losses become significant and keep the fan from coasting on forever at low speed.

I've seen this happen when using ABB's flux braking. It doesn't have any noticeable effect until you get down to slow speed and then it really pulls hard to a stop. That makes sense to me because, as speed goes down, per the formula, the torque goes up for a constant hp.

With a non-linear hp curve as in a fan, it wouldn't take much braking hp to make a large difference in coast time even tho it all happens on the slow speed end.

Am I making any sense here?
 
Man, I had no idea what I was starting. The only reassurance is that I am not alone in being at least moderately confused! I'm glad I'm not alone in being interested in the answer, too.

You are definitely making sense Dick. I am concluding that the deceleration I see (I think it may be adding to the confusion to call it "braking") is part losses in the drive, part windage or friction in the blower, and part flux braking. As I said some time ago, to decelerate the blower in the time frame I was talking about would require less than 10% of motor rated torque, maybe 8% to 5% depending on motor WK2. That seems in line with what you are saying.

To also back up your explanation, we do usually kick in DC injection braking at 2 or 3 HZ. The only clarification to your pretty good educated guesses is that this is a high static pressure air system, so we run out of flow at about 45 Hz.

I've sent a link to this thread to my GE rep and my C-H rep. I'll add A-B to the mix, and any responses I'll pass on. I will be on site the week of June 27, and I'll post my field test results too.
 
Last edited:
As it happens, I just got a reply from Cutler Hammer. My questions are black, the response in blue.

I have some questions I thought you could help with:

1) What percent of nominal motor torque is available for occasional deceleration of the motor without using a chopper and external resistor. We normally only have one or two start/stop cycles per day. Drive offers max 10% breaking torque w/out chopper and external resistor.

2) The specs for the SV9000 show DC braking at 30% nominal torque, but I am assuming this only comes into play for the final stop below 1 or 2 Hz. Is this correct? If freq > nominal motor freq, parameter 4.9 determines DC Braking time. See manual for more details.

3) In the normal deceleration, without a braking resistor, is the braking method flux braking? yes. If so is the inertial energy dissipated as heat in the motor? motor & stored energy in the DC bus-that's why if decel is set to quick, drive trips.

4) What is the difference between the normal braking method and regenerative braking? In normal braking, energy is dump in resistor while in regen, energy is dump back into the power line.
 
There! That's the key---flux braking! One thing not mentioned is that the amount of braking available is highly dependent on motor efficiency. ABB reports about 15% on a standard efficient motor and less than 10% on a premium efficient motor.

But those are good answers from your C-H contact. Good for them.
 
Tom , is there more than one blower here ? if so , and you wish to actually achieve the stop times you require without the expense of an external brake resistor , you may consider coupling the DC buses together and using regen braking , in that case the excess energy will be dumped ino the DC bus and used by the other drive . Saves money on two counts - used very frequently on stretch units and bridle control
 
And here is the response from Allen Bradley. Both of the replies are consistent with each other and with some of the responses here. I'll add the GE reply when I get it.

1) What percent of nominal motor torque is available for occasional deceleration of the motor without using a chopper and external resistor. We normally only have one or two start/stop cycles per day. Approx. 5%-15%....... "time" is one of several important things that others are missing. Capacitors, DC bus chokes can store energy.
Safetronics is a GE/Fugi brand label. Many GE drives do not have DC Link chokes. Also they use the minimum amout of capacitors....to help lower cost. The Bus regulation gains are fully adjustable in the AB PowerFlex, etc. GE's is likely not.

2) I am assuming DC braking only comes into play for the final stop below 1 or 2 Hz. Is this correct? Comes into play much earlier........but is most effective below 20% of rated speed. Effectiveness increases with decrease in speed.

3) In the normal deceleration, without a braking resistor, is the braking method flux braking? If can be if programmed so in the AB drives. If so is the inertial energy dissipated as heat in the motor? Yes. We are now introducing a "Fast Brake" option as well. This is different than the DC Injection or the Flux braking....and can be even more effective.

4) What is the difference between the normal braking method and regenerative braking? Regen goes back to the line, other methods disapate heat in motor, resistors, etc. If you regen a lot.....and at high power levels......relative to the incoming power from the utility.........they will require a co-gen license.......you spin the electric meter backwards........ . If you do it not often.......but with very large amount all at one time (fast e-stop) with all machines......you can pop the incoming substation circuit breakers. Regen drives basicly have another drive on the front end (IGBTS) along with sonme other things to allow the energy to flow back to the line and be used elseware.
 
GE/Fuji Drives (AF300 G11 & P11 ) below 15 HP come with a built in DB resistor. For your HP, you will need to add the control unit and the DB Resistor. You showed us page 5-16, you need to read page 2-9. You will not "control" the decel on this application unless you add a resistor or a Bonitron Regen device to the drive. I cannot believe your GE rep did not inform you of this. You simply cannot control the decel unless you can dissipate the energy when you reach maximum DC bus Voltage. Add the resistor and control (Make GE Pay for it) and your customer will have a very reliable drive system.
 
POint 4 is very ambiguous , generally it is not worth the effort to supply power back to the utility . It is however feasable to use regen braking with a common DC bus , where the drives themselves are not fed AC , but DC only and have no rectifiers installed . This is often worthwhile . In your case , if you have more than one drive , consider connecting the buses together
 
For those who don't have the manual, here is the page "guest" is referring to.

Note that the manual doesn't say "to provide any deceleration" but rather "To improve braking performance..." I have come to the conclusion, confirmed by my own experiences, by the other suplier's responses and by some of the responses to the thread here, that deceleration torque is available from most drives. This deceleration is at a low duty cycle and at a small percentage of motor full load torque. This torque dissipates the energy by charging the DC bus and as heat in the motor through flux braking.

Not surprisingly, the percentage of full load torque available for deceleration varies with manufacturer. If this torque is 5% to 10% of full load torque it is adequate for my application, and I don't need braking resistors.
 

Similar Topics

I need some clarifications from VFD experts. I have a project with a vibratory conveyor that must be stopped relatively quickly. Stopping is once...
Replies
2
Views
760
Hi. I have a FR-V520 3.7Kw VFD that seems to have lost it braking. I used to run the machine it is driving and when the motor stopped the...
Replies
9
Views
3,510
Howdy, I was called onsite this afternoon to take a look at a Lenze SMV drive with a braking unit and resistor. Whoever had installed the braking...
Replies
4
Views
2,624
Hello! What would be the proper value for the DC Brake time on VFDs for the DC injection to the motor? Is there a way to calculate? Having it...
Replies
7
Views
4,391
Am teaching a class Oct 18 and 25 Basic VFD and 3 phase motor applied to electric vehicle. Have VFD and motor setup made by Fuji for GE field...
Replies
12
Views
6,784
Back
Top Bottom