ABB ACS550 Braking Torque

I'm popping in here kind of late but first I need to clarify that the ACS550 is not a DTC drive so all this about variably pulse rates doesn't apply. The 550 is simply a generic type of sensorless vector control and does not have excellent motor shaft control at and near zero speed just like any other sensorless vector system.

On the other hand, the ACS600 and now the 800 and 850 are DTC drives. You can expect them to develop full nameplate torque at zero speed.

In web applications, I would be very hesitant about using the ACS550. The ACS800 is much preferred. However, as kamenges has found, the 550 can do a very good job if an encoder is the speed feedback device. With an encoder I would expect zero speed control to be quite good.

And, just a note for kamenges, the tension control you describe IS a braking application except it doesn't involve stopping. All quadrant 2 and 4 operation is braking or, the motor developing torque in the opposite direction from the rotation.

Glad you got it solved. These are nice drives within their limitations, of course. If you are going to do much braking, you will not be able to depend upon flux braking alone. Flux braking is constant kw braking so at higher speeds there is very little braking torque. As you slow down, the braking torque increases noticeably. I would think that you would want to use the built-in brake chopper and add a braking resistor for web tensioning apps like yours.

Sorry I wasn't available earlier but I spent my whole day nursing an old 250hp DC motor back into operation on a hydraulic pump test cell with 40% field-weakened overspeed. And it was a Bardac 4Q drive which I am starting to like.
 
Thanks for the reply, DickDV. I'm glad you dropped in on this.

I didn't realise that the ACS550 was not one of the DTC offerings. The way it was presented to us the ACS550 was like an ACS800 without some of the application bells and whistles. Kind of like the difference between a PowerFlex 700 and 700S. That is obviously not correct. Also, from what you stated it sounds like the ASC550 is always operating in sensorless vector mode. the encoder is just used for velocity loop closure. Is that correct?

As you said, getting the encoder feedback active really made a huge difference at zero speed. In hindsight how the motor reacted to speed error makes alot of sense now.

I also understand this is a braking application. My response on this was geared toward this from dmargineau:
So you are actually trying to use the ACS as a "brake" by running it in Vector:Speed control with "zero" speed reference...
.

I just wanted to clarify that I wasn't trying to use the drive and motor as a glorified friction brake. I want to run it at a speed setpoint but there is the possibility that the web will overhaul the roll, generating (in my case) quadrant 2 operation. We do have braking resistors connected to the drives to get rid of the bus energy.

Thanks again, everyone.
 
kamenges, The ACS550 is software selectable (Para 99.04) between open loop V/Hz and Vector control. The vector control is either Sensorless Vector with no encoder or Flux Vector with encoder.

The sensorless vector control is pretty much a commodity grade system and is good enough for many applications but it is not as good as DTC (Direct Torque Control) by an order of magnitude especially near zero speed. The addition of an encoder turns the 550 into a very tight accurate system right down to zero speed as you found.

If you had chosen an ACS800 with DTC, you probably would not have needed the encoder to get the zero speed performance as DTC is quite good even at stall speed. We do use encoders on DTC systems for extreme accuracy and zero speed control where the shaft has to stand perfectly still. With DTC alone, the shaft may "jiggle" back and forth a little and, on a few applications, that is unacceptable so the encoder is added.
 
Thanks for clarifying the sensorless vector thing. I misunderstood what you meant. That control style is what most of the other players do. Like you said, had we gone with the ACS800 we could have gotten by without encoders. We really don't need to do much at zero speed on this line. The reason we ended up with the ACS550 is that we were told the ACS800 Ethernet/IP interface was in beta testing when we had to order the drives. We didn't feel comfortable being that close to the bleeding edge on this project. Although, in hindsite, having better than 1RPM speed command resolution would have been nice.

Keith
 
I guess at the time of your original inquiry they had a warehouse full of 550s...:)
The ACS800 RETA-01 Ethernet/IP communications module has been available for at least four years or so...:nodi:
Speaking of salesmanship...o_O
 
Uhmpf!! That's a kick in the privates.

If you tell me the RETA-01 on the ACS800 can take a speed reference with better than 1RPM resolution that will be a stab in the heart.
 
On one of our productions lines, we are RPIing fifteen RETA-01 modules (installed on fifteen synchronized 100KW ACS800) at 5ms each via unmanaged switching bridged by a 1756-EN2T; the speed reference is digital.
So let's say that the 1756-L62 RSL5K ACS800 scheduled task runs every 5ms with a 5ms scantime; this would give you a total of 7.5ms delay of establishing the system's Speed/Torque loop error.
What would that be when talking speed reference resolution at 1700 RPM base speed?:unsure:
 
Originally posted by dmargineau:

What would that be when talking speed reference resolution at 1700 RPM base speed?

Your post tells me speed command update rate. I'm talking about command resolution, or granularity.

For example, I can command 1700 RPM or 1699 RPM. But I CAN'T command 1699.25 RPM or 1700.34 RPM. The speed command resolution to the drive is RPM, not tenths or hundredths of RPM. That's not a big deal if I have a fan or a free-standing conveyor. But when I have 5 drive points tied together by a web it would be nice to have a speed command resolution at least as good as the drive velocity accuracy.

Keith
 
As I have stated in my previous post, since the RETA-01 modules are fully EtherNET/IP integrated within the Logix system, the reference is digital (REAL data Type-32 bit resolution).
If the controller decides that the drive needs to go from 1129.9999 RPM to 1130.0001 RPM it will accomplish this within 7.5ms.
 
Just because the reference is digital doesn't infer a specific resolution. We are communicating to the ACS550 via Ethernet/IP, which makes the reference digital. However, we are limited to a 16-bit integer and a speed reference resolution of RPM. Granted, I can command it to go 32,767 RPM but I don't need that.
 
ABB ACS550 series has a vast selection of drives. They range from simple wall-mount fractional hp drives to large free standing ones up to 500hp.

I believe you could have used a different series like ACS310 or ACS355.

ABB came out with a new micro drive this year, ACS250. I haven't had a chance to play with one yet.

http://www.clrwtr.com/ABB-ACS250-Drives.html
 

Similar Topics

I just replaced this drive as the old one went out when the plant had a power surge/flicker. 1-Installed the drive 2-Powered up 3- Got an...
Replies
4
Views
1,668
Hi Team, I am working on one Paper Coating machine, My Winder, infeed, Coater and Unwinder is runnning through ABB's ACS550 VFD, My winder is...
Replies
0
Views
1,323
Hey all. I tried to startup a drive last thing before leaving for the weekend. On power up, the drive constantly gies earth fault. This is with...
Replies
8
Views
2,227
Iam getting Fault F021 ("System Fault Current Measurement is out of range"),Any help ??
Replies
1
Views
3,028
Please i need help i lost all setup parameters to control 3 pumps pressure i put again some of them but i get stuck in ref 1,2 choice and...
Replies
2
Views
1,451
Back
Top Bottom