VFD and linear motion

turner15429

Member
Join Date
Apr 2018
Location
Landrum, SC
Posts
13
I am currently trying to figure out how to program a machine that has a rack and pinion driven by a 7.5 hp gear motor and a Mitsubishi D700 series VFD. I have a AB Ethernet IP encoder hooked up directly to the pinion drive shaft for location feedback. The drive has 3 inputs controlling it forward and reverse and a 4-20ma for speed. The sled is driven back and fourth at a distance and speed set by the user. The acceleration time and deceleration time are pertinently set in the VFD. The 4-20 just sets the max speed. The pinion only rotates 20 times max and I only need to +or- an inch so this doesn't have to be extremely accurate. My plan was to just to switch my output from forward to reverse when the counter gets to a certain set point. The main problem with this is I don't know how to calculate how many revolutions I'm going to get on the deceleration so I can subtract that from the set point. Any guidance would be great!
 
Picture a graph of velocity vs time. Your sled goes from Y inches per second to zero inches per second in t seconds. The distance traveled during deceleration is equal to the area of the triangle described by the point where you begin to decelerate, the point where you get to zero speed, and the horizontal axis, or Y/(2*t).
 
Picture a graph of velocity vs time. Your sled goes from Y inches per second to zero inches per second in t seconds. The distance traveled during deceleration is equal to the area of the triangle described by the point where you begin to decelerate, the point where you get to zero speed, and the horizontal axis, or Y/(2*t).
So if I'm going 40 inches a second and I stop in 1 second I will travel roughly 20 inches
40/(2*1)=20
Also does the way I want to control my machine seem fairly feasible?
Is my drive going to have a ramp as linear as I expect?
 
If you are using a rack and pinion that would mean the motor has to start and stop for each direction?? I doubt that a 7.5 hp gear motor is going to last long doing that.
 
This has been covered before on this forum. A fellow posted a calculation that can be done in the plc to drive a VFD to a position.

If it doesn’t need anything fancy, then don’t make it fancy.

The data points you will need:
1) Current Position
2) Position Setpoint
3) Position Error
4) Minimum Speed
5) Maximum Speed
6) Position Window

I can’t remember the exact formula, and can’t seem to find the post.
Basically, the position error with a gain is going to drive the motor to the position setpoint.
The minimum speed keeps the motor turning until it reaches setpoint. The max speed keeps it from shooting to the moon. A position window is nice to have so you can tell when the motor has positioned itself to the position setpoint.

If you don’t find the old post by Monday, I can give you a sample.
 
Just remembered this thread, sorry. Lol

The calculation I was talking about is this:

((Square Root ( Target Position - Current Position )) x Velocity Gain ) + Minimum Velocity

This will output a speed reference that you can then send to the VFD.

You will also need a max speed limit as well. Have another rung that checks if the speed reference is greater than X, if so then MOV your max speed into your speed reference.

I can’t remember or find where I originally found this but it was here on this forum. I don’t take credit for it at all.
 
Last edited:
Just remembered this thread, sorry. Lol

The calculation I was talking about is this:

((Square Root ( Target Position - Current Position )) x Velocity Gain ) + Minimum Velocity

This will output a speed reference that you can then send to the VFD.

You will also need a max speed limit as well. Have another rung that checks if the speed reference is greater than X, if so then MOV your max speed into your speed reference.

I can’t remember or find where I originally found this but it was here on this forum. I don’t take credit for it at all.
That doesn't look quite right. A 2 is missing.

The basic formula is v^2=2*decel*dist
http://zonalandeducation.com/mstm/p...ntroductions/TimeIndependent/Introduction.htm
but when transitioning from one speed to another it is v^2-vmin^2 = 2 * decel*dist.
Therefore the equation is v = sqrt(vmin^2+2*decel*dist)
dist is the command position - actual position
decel is the deceleration rate
 
Thank you Peter.
If the calculation is time independent, what would the units be for the deceleration rate?
Would it be the distance from target to decel?

I am currently testing your revised calculation against the one I mentioned.
 
Thank you Peter.
If the calculation is time independent, what would the units be for the deceleration rate?
Would it be the distance from target to decel?

I am currently testing your revised calculation against the one I mentioned.
Acceleration rates have units if distance/time^2.
It can be meters/second^2, ft/second^2, inches/second^2, mm/second^2.
Some crude motion controllers use counts/second^2.

The equation you found is wrong. It is missing the 2.
 
Peter, that equation isn't intended to do profiled motion or determine time ir distance to decel. It is basically a self contained position proportional controller. The square root will provide a quasi-linear decel rate in time but in reality it is just position reflexive. The axis operates at a speed proportional to how far it is away from where it should be.

We do this here on occasion with low bandwidth, low accuracy positioning axes just because it is so simple and it is fairly hard for people to mess up.

Keith
 
I'm not sure about the D700 but most drive decel time settings are relative to max speed. Until told otherwise we will assume this is the case. Divide the drive max speed in RPM by 60 to get revs/sec max speed then divide that by the decel time to get revs/sec^2.

At that point distance to decel from a known velocity to zero speed in motor revs is:

V^2/(2*a)

where V is the motor velocity in revs/sec and a is the decel rate in revs/sec^2. divide that number of revs by your gear ratio and you end up with pinion revs to stop. Since the encoder is on the pinion shaft this is probably the number you want to work with. You could also convert to inches of pinion pitch diameter travel if you want bu that isn't necessary.

Keith
 
Acceleration rates have units if distance/time^2.
It can be meters/second^2, ft/second^2, inches/second^2, mm/second^2.
Some crude motion controllers use counts/second^2.

The equation you found is wrong. It is missing the 2.

Thanks Peter for clarifying.
I was not able to test for position accuracy improvement. I did place a compute instruction in parallel to my other to compare how the two reacted to the same inputs.

The revised calculation seemed to offer more control of the speed or would seemingly be able to slow down to a specific position with more repeatability.

The first calc is essentially very near sighted. It can either be setup to creep slowly towards a target and stop at target or to hit the gas and speed toward the target then hit the brakes when it finally can see the brick wall (target).

This is probably just a trade off between accuracy and speed, but it does work. Our machine will hold +/- .100” with a max travel speed of 250IPM and minimum at 1IPM.

I think Peters calc will essentially put some glasses on the previous calc so that it can see the brick wall sooner and prepare to stop.

If that is the wrong way to look at it, please correct me.

PS: I don’t understand how the calculation can be time independent but the decel rate be in a measurement of distance over time?
I don’t know what other unit it could be? Most if not all rates I have seen are distance over time squared.
 

Similar Topics

Hello, I'm trying to delve a little into rs-485 communications for a couple projects of mine. Until now I've been using a delta vfd and a delta...
Replies
1
Views
13
I have an issue with Power Flex 525 during running processing, the VFD stopped suddenly while the PLC and VFD connection ok, VFD does not have any...
Replies
1
Views
85
Has anyone setup communications with Red Lion 3.0 MODBUS to Baker Hughes Centrilift GCS VFD? Thanks
Replies
0
Views
70
Hello, I've been trying to learn this a while now and still have not found out how this works. I have an Omron CJ2M PLC and an ABB ACS 355 VFD...
Replies
1
Views
190
Looking for some assistance. I am very familiar with Yaskawa VFDs, but not so much with AB VFDs. I am working on some hardwired AB PowerFlex 700...
Replies
2
Views
86
Back
Top Bottom