What's new
What's new

VFD torque boost for stall prevention.

viper

Titanium
Joined
May 18, 2007
Location
nowhereville
I am fighting with an older machine that seems to use an open loop type VFD (I think) in which it does not know the actual rpm, therefore cannot adjust for additional load. Is this type of function called vectoring? More or less I am looking at the drive to see if there is a way to get a tach signal to it so it can work better to add more power when needed.

As I understand it, this VFD will put out a certain frequency and amperage for a max load and has no idea if the load is stalling or not. I am not sure if bring a tach input back will help this or not.
 
See if you have a PWG or encoder option that can be added to the VFD.
This acts as a digital tach.
You would also need to have some way of fitting it to the motor shaft.
I have done this in the past and it has made considerable difference.
Many of the more sophisticated drives take an electronic snap shot of the motor and tries to maintain the correct/optimum vector angle between phase and rotor, based on certain current voltage feedback, this is accurate to a degree, but usually loses effectiveness as rpm drops.
Minder.
 
This machine already has an encoder on the motor coupled before the transmission. It is used with the control to sync the motor and the Z axis for rigid tapping. However, it would seem that the actual drive does not know much of what is going on. It gets a reference voltage going back to it to half ass determine load. I am trying to find settings right now that will allow me to better boost the torque or "stall prevention". Seems anything over 50% on the load meter is destine to stall. I guess the way this one is setup is it drops frequency in a stall to reduce slip. I need to to boost everything and put put what it was made for...


Any idea if an inductive amp meter will read accurately on the motor leads from the VFD? Just real curious to see how much is actually getting to the motor. I have to account for some loss in the trans and such but 60%??!! I am hoping I can do something creative with this guy to just get a little more go out of it.
 
If you look at any high end CNC spindle, they are going to have feedback from motor to drive, the spindle encoder is just to sync the Z axis (mill) to the spindle revolutions, it is important to also have tight control of the spindle motor itself via motor to drive feedback.
Minder.
 
I realize the setup on this machine is not ideal so just trying to improve what is has a touch. I know it does not have a true closed loop and it shows. Performance is drastically lacking.
 
As I understand it, this VFD will put out a certain frequency and amperage for a max load and has no idea if the load is stalling or not. I am not sure if bring a tach input back will help this or not.
A basic VFD (Scalar Drive in V/Hz mode) puts out a controlled voltage and frequency, but amperage is a function of load. But you're right, it has no idea whether or not the motor is keeping up with the load, other than to monitor it for overload conditions.

This machine already has an encoder on the motor coupled before the transmission. It is used with the control to sync the motor and the Z axis for rigid tapping. However, it would seem that the actual drive does not know much of what is going on. It gets a reference voltage going back to it to half ass determine load. I am trying to find settings right now that will allow me to better boost the torque or "stall prevention". Seems anything over 50% on the load meter is destine to stall. I guess the way this one is setup is it drops frequency in a stall to reduce slip. I need to to boost everything and put put what it was made for...
What you need is indeed a "Vector Drive". Vector Drives are a higher tech variant of the original Scalar drives in that they have a MUCH more powerful microprocessor on them that can do the complex math of calculating voltage and current vectors to either maintain precise speed control or precise torque control (or both) within the limits of the motor regardless of the load applied to it. So it can essentially correct the error so to speak between what you want the motor to do and what the load is allowing it to do.

If your VFD is not currently a Vector Drive, then it cannot do anything with information from a tachometer or encoder etc. If it is, Closed Loop Vector Control (encoder feedback) is not something for the faint of heart, it takes a lot of tweaking capability to make use of that kind of functionality. But there are a lot of "Open Loop" or what are called Sensorless Vector Control (SVC) drives that use motor data and current measurements to accomplish nearly the same performance, probably all you need. Most SVC drives have what is called "autotuning" or "teach mode" where the VFD tests the motor it is connected to and creates a model in software so you don't have to. that makes them very simple to implement. The only difference now between Closed Loop Vector and SVC drive performance tends to be at Zero Speed meaning when you need full torque from a motor that is not moving, i.e. a hoist application.

Torque Boost is something entirely different an its purpose is to attempt to make up for a V/Hz drive's poor low end torque production, i.e. at roughly 25% speed and below. All it does it allow you to program a manual tweak to the V/Hz ratio at the selected speed and below, which would technically cause the motor to overheat, but by limiting it to low speeds, the motor power is already lower than it is rated for, so you can get away with it for a little while. Running in Torque Boost continuously however can be bad for your motor.

Stall prevention is something all together different again. Its purpose is to allow the VFD to keep the motor energized even if the motor stalls. It is exactly the OPPOSITE of what it sounds like you want. It drops the frequency output as a way to keep the motor from overloading on a large step change in load, essentially sacrificing HP and speed for continuity, assuming that it can re-accelerate it over time to recover.


Any idea if an inductive amp meter will read accurately on the motor leads from the VFD? Just real curious to see how much is actually getting to the motor. I have to account for some loss in the trans and such but 60%??!! I am hoping I can do something creative with this guy to just get a little more go out of it.
No, typical meters will not accurately read the output of a VFD, the signal is way to "dirty" with harmonics for most meters to accurately interpret it. The best way of seeing the output current is to read it from the VFD display, but if you have a transformer in between the VFD and the motor (why?) then that is not going to account for losses in the transformer.
 
There is not any electrical devices between the drive and motor, I was indicating a transmission from motor to final output. I am seeing pathetic outputs from the spindle and need to address the issue.

My goal is not to swap out the drive, but rather take a look at the operating parameters to see if there is a way to improve performance. As I understand it, decreasing the stall parameters would force the drive into fault earlier when the "quazi closed loop" voltage says it is bogging down. I can bring it to a complete stop pretty easy. Pretty pathetic for a machine tool...
 
Viper

Complete documentation of the drive unit would be a start toward possibilities.

I've been pouring of the Toshiba Tosvert Documentation today as I experiment with "over voltage supply" ;-) (now what the heck is that?). This G3- VFD drive will do most everything but butter your toast!

If I could be so smart......

CalG
 
Something does NOT add up here....

The VFD doesn't NEED to know the RPM normally, because with any reasonable load, the motor will not have much slip. If it DOES have a lot of slip, then typically the motor current will be headed for the sky, which the drive will notice.....and supply, if the current limit is not reached first.

A vector drive is good, especially at lower speeds. But reasonably near normal motor RPM, the motor cannot stall without drawing lots of current. If the VFD can supply it, it will, and if it cannot, then feedback won't help anyway.

If the motor stalls, then even connecting it "to the wall" won't help.

Now if, as is likely, you are operating at very low speed, THEN feedback or vector drive may be effective, because there is voltage to spare, and the VFD actually has something it can do.

But even there, the ultimate limit is current, since even a constant torque is a falling power as you go to lower speed. At 1/10 speed, you must have 10X torque, which is 10X current, to maintain power.

The fix is to over-size the VFD, and maybe the motor, so that at least short periods of high torque can be had.

A 10x larger motor would supply the original nominal HP at 1/10 speed.....

A 10X larger current limit in the VFD will supply lots of current at low speed..... if it isn't for very long, a smaller motor can take that, you don't need one of the full 10X nominal power.
 
I have a hunch that he has an olde VF-1 with a 5HP spindle and Yaskawa GPD-503 drive. My machine has the 515 because it was replaced. Open loop vector. I don't know why he is having issues, it's got a gearbox for low speed torque. My VF-0 has 5HP and I learn to deal with it. If I need any major hogging I'll toss it on the Shizuoka with a "4 HP" motor and vari-speed, with hi/lo range.

Luckily I have a 15/20HP machine coming soon and it shouldn't have any troubles.
 
Seeing what your symptoms are, I think you should check out a different issue. It could very well be that your motor is improperly strapped; i.e. you have a 240V supply and the motor is strapped for 460V operation. When that happens, your torque is reduced to 1/4 of normal and in a lot of cases you will be able to stall it very easily. Eliminate that possibility before wasting more time on this, it's a very common problem.

If that's not the issue, then please post the make / model of the drive so we know what your dealing with.
 








 
Back
Top