Determining maximum available braking torque on DC Drive

Total_newbie

Member
Join Date
Aug 2020
Location
Detroit
Posts
24
We have a DC Drive (ABB DCS800) which is running a load (initially) at a fixed speed. For testing purposes, we want to simulate a different inertial load on the machine at the time that we apply a mechanical brake (simulated inertia). We achieve this by switching (after we achieve a given speed) to torque control - and then adding or reducing torque in order to achieve a target acceleration/deceleration while brake is applied.

I would like to determine the maximum simulated inertia which we can achieve with drive. If we are adding inertia (= positive torque), I believe that this calculation is relatively simple. I know max torque that drive can put out (at base speed), and I can use this value to determine max torque which I can add (assuming we are at or below base speed) to simulate a greater inertia. My question is how much reverse (braking) torque the drive can put out when it is rotating in forward direction. My impression is that this should be independent of present (forward) speed. Can I assume that I can get max torque in reverse direction in this case? (I think that this is quadrant 2 - forward braking). And - there must be some switching time which is required to go from positive torque to negative torque. I cannot find this though in specs.

Any clarification of these issues would be greatly appreciated!
 
We have a DC Drive (ABB DCS800) which is running a load (initially) at a fixed speed. For testing purposes, we want to simulate a different inertial load on the machine at the time that we apply a mechanical brake (simulated inertia). We achieve this by switching (after we achieve a given speed) to torque control - and then adding or reducing torque in order to achieve a target acceleration/deceleration while brake is applied.

I would like to determine the maximum simulated inertia which we can achieve with drive. If we are adding inertia (= positive torque), I believe that this calculation is relatively simple. I know max torque that drive can put out (at base speed), and I can use this value to determine max torque which I can add (assuming we are at or below base speed) to simulate a greater inertia. My question is how much reverse (braking) torque the drive can put out when it is rotating in forward direction. My impression is that this should be independent of present (forward) speed. Can I assume that I can get max torque in reverse direction in this case? (I think that this is quadrant 2 - forward braking). And - there must be some switching time which is required to go from positive torque to negative torque. I cannot find this though in specs.

Any clarification of these issues would be greatly appreciated!

The energy that you generate while braking has to go somewhere. The way that the DC drive implements 'negative torque' is by adjusting the field current, changing the DC motor to a DC generator (but quickly, so as the speed falls it can change it back to a motor again).

The DC bus voltage will rise and trip the drive on over-voltage if there is no load resistor to switch on and drive the current through, or some method to regen back into your supply grid.

There is no 'switching time' from positive to negative torque. That is wrapped up in your maximum acceleration and deceleration time. If you have a 5 second accel time and a 10 second decel time, it will take 5 seconds to ramp up to rated speed, and 10 seconds to ramp back down to 0. Going reverse counts as decel, so another 10 seconds to get to rated speed in reverse.

Hope that helps
 
I assume you have a regenerative drive. I believe that drive is available in either flavor.

The available torque should be the same in any quadrant, unlike dynamic braking, the SCR's can control the current to (almost) zero speed. Unlike an AC drive, there is no worry with a DC bus, since the back-to-back SCR's will push the current into the motor, or back onto the line, depending on the toque (current) direction.

There is a slight delay between forward and reverse current, to protect the drive and motor from passing AC current to the motor by turning on both directions.

The lockout time is probably in the 20ms world or so. Not a big deal on an Inertial Dyne. It really only comes onto play on an indexing type of application.

Sounds like you have a pretty good handle on how the drive and motor work. You should be OK to play with and perfect the simulation you're after.
 
Thank you for the responses. What is confusing me now is that during a forward braking stop (initially running at ~700RPM), I am applying enough negative torque (I am calling for constant torque) to (in theory) come to a stop in about 6 seconds. What I actually see is that during first half of decel (~3 seconds, going down to around 350RPM) - I get steady decel at the rate which I expect. Torque output shows that it is around 85% of max available torque. Then - when I get down to 350RPM, torque output now maxes out at 100% - but my decel rate drops, so 2nd half of stop ends up taking about 25% longer than first half. I know that drive is running in regen mode. I am wondering if at lower forward speed (= less regen power available) - there would be some reason why drive would not be able to supply enough additional current to keep torque steady? Thanks for any ideas...
 
Asking what is the maximum available braking torque of a DC motor or any motor for that matter is a loaded question. Without context and more information itÂ’s really canÂ’t be answered. Normally it would be 100% of motor torque
But the truth is the torque it takes for something start to break apart would be the maximum for both DC and AC motors. I have seen gears strip off, belts burn off, motor shafts break, amateur spin on the shaft all kinds of bad things. The best answer is how much energy can you safely dissipate. Assuming that you can dissipate all of the energy then you would be safe at 100% motor rated torque for 100% of the time or 150% torque 25% of the time.
With most DC drives non regen the energy dissipation is through the Dynamic Brake Resistor (DB) the DBÂ’s are switched in across the armature and the drive is disconnected when you want to stop the motor. The energy to slow down and stop the motor is sent to the DB and dumped to the atmosphere as heat. A good example of this is a locomotive while they are diesel electric on the older ones the traction motor is DC the newer one the traction motors are AC both use DB resistor banks for dynamic braking. They are located on top of the cab Keep in mind that not all locomotiveÂ’s are capable of dynamic braking. If the drive is a regen drive then the energy is sent back into the power line and. The limitations are how much energy can you pass from the motor through the drive back into power line.
Keep in mind that the torque goes down as the motor speed goes down as the motor slows down the torque developed is less and energy produced is less.

From you last post it looks like you are not really braking the motor but just ramping the speed down to 0. With the speed going down the torque developed in the motor will also go down. The torque developed by the motor will be what is necessary to maintain the set speed. Again without the ability to see everything it is hard to advise you.

Sometimes I am not very good at explaining things I hope this helps. It is more complicated then this but is a start.
 
Thanks for the explanation - this certainly does clarify some of the general ideas which I am trying to understand. But … in my case, I am getting drive up to specific speed (speed control), and then switching to torque control (where I send an analog signal corresponding to percentage of max drive torque which I am requesting). (My control routine actually is PID loop which controls drive torque to yield constant decel rate.) I get enough (negative) torque I am looking for when I start my decel, and decel rate is exactly what I am looking for. But, as I decelerate to lower (forward) RPM - I can no longer generate enough (reverse) torque to maintain decel rate, and decel rate drops.
As I understand - when drive is going fast, I am taking drive energy and feeding it back through regen to produce the (reverse) torque I need. Obviously, as speed drops - my available regen power also goes down. So - at some point shouldn't my drive have to switch over from (completely) regen braking to actually drawing (outside) power to provide torque? It seems to me that it is when this switch occurs that I can no longer get the power which I need to provide same torque...
 
Some Drives have setting to Extend ramp to prevent OC trip

Some Drives have setting to Extend ramp to prevent Over Current trip. Used this on Yaskawa AC drives. Else the drive can trip on ramp up or down if accel or decel is calling for more current than drive can produce before a over current fault. Just a thought on your comment
 
It sounds to me like you have the drive set to limit the regen torque to 100%, and the load (for some reason) is requiring more:
Then - when I get down to 350RPM, torque output now maxes out at 100% - but my decel rate drops, so 2nd half of stop ends up taking about 25% longer than first half.

To decel faster, you'll need more torque. If it's over 100% for a few seconds, shouldn't be a big deal. Check your settings and commands. The drive is likely doing the best it can with the instructions you're giving it :confused:

IMO, you should not have to screw around with torque settings to follow a linear decel ramp. The speed loop and ramp generator should take care of that, assuming current limit is not going active.
 
Let me see if I can clarify things a little
First when you refer to reverse torque I think this is where the confusion is
When I hear reverse torque command I think that you are trying to reverse the direction of the motor but what you are really trying to do is reduce the torque limit on the motor
Try this set the max output of the vfd to 60 Hz and leave it there you want to control the torque limit not speed. If on start up ramp up the torque command ( torque Limit) until the motor is running at the desired speed. Then on stop ramp down the torque command (torque limit) until the motor stops or is reduced to the new desired speed.
Torque control is a bit tricky when you are first start working with it.
Think of it this way if the load requires 50% torque to move it and your max set speed in the vfd is set at 60 hz you start out with the torque limit is set at 0 and slowly increase it. The motor speed / hz will be 0 until the torque limit exceeds the required torque to move the load 50%
At that point as the torque limit increases the motor speed will also increase but as the torque limit increases above the level required to move the load the speed will also increase very quickly to the max speed. It’s a balance torque control with speed limit. As I said before I like to keep my max speed limit to just above the line speed I want, that prevents the over speed you are seeing.
Old servo drives were torque control ( torque limit) with speed feedback to limit the torque so as the load demand goes up the torque limit in increased to allow the speed feedback to match the speed set point at that point the torque limit controls the speed.
In your case at the end of the cycle leave the speed command alone and reduce the torque command ( torque limit) I think you will find that the output speed will go down just like you want. You may have to decrease the torque limit quickly as the load is removed to prevent the over spee. If you want to limit the accel rate when you come out of torque control then set the ramp up rate. When you are in torque control the acc and dcc ramps have very limited value
In your case it looks like the load / torque requirements drop away quickly and the torque limit is set above the requirement so the vfd quickly accelerates up to the max limit set in the vfd.
 
Yes thank you I rechecked and I see where I referred to the speed as hertz I am so use to working with vfd’s it just slipped in I will have to be more careful.
It doesn’t change the control system, in this case I am sure the original design was a torque controlled motor with speed feedback to control the speed by adjusting the available torque.
As was stated the motor speed is feed back into the controller and is used in the torque PID control.
In the old days the DC motors didn’t run very smooth at low speed under speed control so the designers used torque control with speed limiting. I have seen many servo’s on CNC machines controlled that way.
I see you are thinking that at some point the DB draws line power for braking. Well that just can’t happen DB only works if the energy is coming from the motor. The dc drive is disconnected in DB and will work at full power with the supply line turned off. That’s where the name comes from “Dynamic Braking” if the drive is dumping the energy back in the line then it is Regeneration but the drive must be able to do that. On Dc drives if it has 3 or 6 SCR’s it not regen if it has 12 SCR’s it may be regen or it could be just reversing.
I would look to be sure that the motor speed signal is getting back into the control even if it didn’t get back you would still have torque control until you remove the load then the speed would spike to full speed.
This one of those I would really need to be there and see it
 
All this information is very helpful. Although I was trying out this inertial sim on DC drive, we are also using it on AC drive - so my understanding of AC drive is also important.

One thing which you mentioned which confuses me is "DB only works if the energy is coming from the motor." Yes - DB only works if it can get energy from motor, but if motor is not spinning (or not spinning fast enough to provide enough energy for DB) - and I nevertheless send a command to drive for (e.g.) negative 50% torque - won't drive pull needed power now from supply grid (rather than regen, which is not sufficient) - in order to reach target torque? (Just the same way that if I start drive from zero RPM, that energy has to come from grid...)

What I am seeing (now trying same thing on AC drive (ACS800), controlling torque to achieve a target decel rate) is that I get exactly the torque which I am requesting (e.g. negative 50%) as drive decels from a high RPM down to a certain cutoff RPM, but then below that RPM, torque readout from drive shows that it is maxing out at (negative) 100% - yet decel rate slows down. I do not see any errors or limit flags set in controller.

I figured that at some RPM, drive must switch from regen power to supply grid - but I don't understand why (if this is case) - I cannot get required power from grid to continue my required torque. (I am thinking of regen brakes on car. Regen braking power will continue until some cutoff speed, and then mechanical braking would have to take over...).
Thanks for the help. I see that I need to assimilate a lot of different concepts in order to get big picture...
 
Yes, Regen and DB are different animals, for sure. AC and DC are also different, since the AC drives require a conversion to the DC bus, then conversion back to AC. The DC drives simply pulse portions of the AC line to make the DC current.

Zero Speed creates a dilemma for either drive, since the motor basically has no energy stored. There's different ways to look at it. For a torque controlled scenario, obviously you would need to send current in one direction according to the command. With AC, the frequency of the current follows the rotor speed (plus or minus slip, depending on rotation/torque direction). For DC, it's much simpler, since the current = torque directly.

So, for DC, you can easily supply torque either direction (with a regen drive) just by which SCR bridge is being used at the time. The current flows to/from the motor/line either way. The only issue is the changeover from one bridge to the other. There has to be interlocks and delays in the drive to prevent both bridges from conducting at the same time (This would look like a short-circuit).

With AC, the regenerative drives are considerably more complex, since you need a synchronous converter to get the current back on the line. This is basically 2 inverters back-to-back. But, since there is only 1 bridge connected to the line, the current can be switched much quicker. When motoring, the current comes from the line, through the flyback diodes into the DC bus. When regenerating, the transistors conduct back onto the line from the DC bus.

For DB, the output transistors must take energy out of the motor and return it to the DC bus. When there is too much energy on the bus, the voltage goes up and the DB transistor comes on to dump the bus into the DB Resistor, thus clamping the voltage by wasting the energy into heat.

Without rotation, the AC drive an quickly pulse the motor one direction then the other faster than the rotor can move, thus creating 'full torque at zero speed'. An SCR-based DC drive can not do this, since the SCR's can't switch off fast enough due to line frequency. For this reason, DC servos have traditionally been more like AC drives with DC bus and PWM transistors, only putting out DC instead of AC. They still use the same DB transistor and DB resistor as AC drives, unless there's a synchronous rectifier (AC Regen unit) feeding/regulating the DC bus.

Hope this sheds some light.
 
The 100% torque required at lower speed to (try to) maintain the ramp rate is confounding.

Do you see a similar torque profile when accelerating? If the load is 99% inertia, I would think the torque would be equal throughout the ram, assuming its a linear ramp.

I would try a longer ramp time, watching the speed and torque as a senility test. Adjust the ramp time so that the torque never goes over say 75%...
 
one quick question
what is happening with the load on the motor dose it drop to 0, stay the same or dos it go up
remember you are not commanding a torque you are only setting and changing the limit the drive is allowed to deliver to the motor if you allow the motor to have more torque then it needs then max speed limit controls the speed but if you allow the motor to have less torque then it needs then torque will control the motor speed
with torque the load on the motor is important
 

Similar Topics

We are in the process of installing a new DC drive to replace a 1373 series Speedpak drive. There are 8 leads coming out of the motor -- Two for...
Replies
5
Views
810
Good morning! I am looking for a clever method to determine the number of occurrences in a series using ladder logic. What I need to determine is...
Replies
6
Views
1,236
I have my PLC and HMI connected to my local network over DHCP just like my laptop. I can upload and download including VNC to the HMI over wifi...
Replies
2
Views
1,129
ive been tasked, due to a restructure of our inventory, to find out how big our SLC programs are. specifically, we have several dozen 5/05s at...
Replies
4
Views
1,589
I'm programming a AB 1769-L18ER processor with Studio 5000. I'm wondering if anyone knows away to determine how many inputs are on thru logic...
Replies
6
Views
2,351
Back
Top Bottom