Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Tricky setup of VFD on machine tool spindle. Mits A500 1

Status
Not open for further replies.

fastline12

Aerospace
Jan 27, 2011
306
Due to a drive failure today, we are installing a replacement A500 VFD. We planned to review this application more before attempting but hand is being forced due to down time.

The installation is very straight forward but this machine orients the spindle for a tool change by commanding a low rpm (approx 40rpm)while pushing a shot pin at a collar with a slot in it to lock the spindle in place. The way it was designed, when the pin falls in, there is an immediate signal to the drive to halt operation. However, the previous drive was a V/Hz drive with no knowledge of rpm. The new drive will be operated in sensorless vector mode and the concern is if the drive can out power this locking mechanism and damage it if signal to stop is not received very promtly.

Obviously there is too much being left out here to really say but I am more or less asking if other machines are doing something similar successfully with a sensorless system? Or even encoder feeback system. maybe the sensorless system in the A500 is slow enough to react to the stop? The motor would be spinning fast enough to develop full torque at these speeds.
 
Replies continue below

Recommended for you

Can you energize a digital input to put the drive into a torque limiting mode.
 
Itsmoked - I firmly agree that encoder feedback here would be best for everyone but the issue is with the control side. When TC is commanded, the control simply outputs a low speed and does everything else internally so a way has not yet been found to send an alt signal for TC which could be used to orient the spindle as needed. Obviously this would be best route and one in which we are still looking into. One possibility is the use of an extra Mcode which would alter the programming slightly from the typical M6 TC to say an M26, M6 but that is not ideal.

Lione - torque limiting is something we really need that this point but again does not look like we are going to have any type of extra signal to the drive at tool change from the control. The old v/hz drive would not have power at these speeds anyway so it was a non issue. I will double check though.
 
I'm no expert on Mitsi drives, but they are not known for skimping on features. If you have an SVC version, chances are very good that you have a torque limit capability. You would not need any external input, it's something you should be able to program into the operation of the drive to be initialized by a drop in RPM below a programmed threshold. In other words you don't necessarilly want torque limit when you are running fully loaded to the task at hand, but say that as it rotates to the locking position at 40RPM, you program it to be in reduced torque limit at or below 41RPM.

In addition, you should also be able to program a torque trip, meaning what could be desctibed as an "electronic shear pin" where if it sees an immediate increate in torque, as in the locking pin dropping in and locking the shaft, it will immediately shut down. I could do this for you on several other brands of drives (Rockwell comes to mind, but that's because they sign my paycheck...), but I would be highly surprised if the Mitsi is incapable of this. You need a good Mitsubishi drives expert.

Or a different brand of drive...

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)

For the best use of Eng-Tips, please click here -> faq731-376
 
I agree with the above, if the controller won't give an output during tool change then just limit torque when you're below a certain speed.
 
I should have also posted, I'd think the new drive could run in V/Hz mode so maybe you should just use that mode.
 
The main reason for switching to this drive is the advantage of SVC for a machine tool application. As load increases, the drive needs to react better to the load to keep it from a stall condition. V/hz has proven one of the worst ideas on the planet for machine tool spindles.

Jraef, my concern with a torque limit is how the drive would otherwise perform in acceleration? IE, if we programmed a "less than 40" rpm reduced torque, every time the spindle accelerates, it would push from 0-? rpm. Would this affect acceleration or am I misunderstanding how the torque limits work?
 
Again, I'm no expert on Mitsi drives, but certainly that is a workable with others. You would simply say something like "XXX torque limit during acceleration, X torque limit during deceleration". Many of the higher end drives have the equivalency of a PLC built into them, one that would by nature be devoted to the motor control applications of course, but no less functional. So anythning you could conceive of in PLC logic can be executed in the VFD. Mitsubishi makes PLCs as well as drives, I would again be surprised if it is not capable of that.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)

For the best use of Eng-Tips, please click here -> faq731-376
 
Yes, there is actually a section of parameters dedicated to PID functions. That being said, I don't want to make this any harder than it really needs to be. I felt my previous question was pretty general to apply to all capable drives. Call it torque limiting, current limiting, etc, I am just curious how a drive typically would handle a setup like this where you are basically wanting to accelerate right through the limited section.

There is a section that will allow limits during accel, constant, and decel but I am inclined to think that if I only use the constant speed limiting, we really don't know when the shop pin would fall in and might be during spindle accel which could prove problematic.

So would it be reasonable to consider the drive would look directing at the commanded speed to validate the torque limit? IE, if limit is set to 40rpm and below and I command 50rpm, it would drive right through the limits to get to 50rpm?

Again, I realize not a Mits expert but I tend to think most drives would respond similar.
 
It would be limited only in that range. So if you just did a tool change and were now accelerating to 3000rpm it would be limited during accel(under 40rpm) by the inertia of the spindle. I'd imagine this would increase the time to Setpoint by perhaps 50ms? I don't see how that would affect your machine tool, since they're always going to wait until "speed reached" before moving to the next G code. You'll probably even get that 50ms back by harder accel available in the Vector mode.

Keith Cress
kcress -
 
A reduced torque should still be able to accelerate the spindle without any problem. I doubt you are tapping with an open loop VFD, but are there any machining operations done below the tool changing speed? If yes, then the reduced torque at the tool changing speed likely won't work.

There must be some kind of signal to control the alignment pin. Why can't that signal be used to kick-in a reduced torque mode.
 
Lion, the machine indeed does tapping. I hear what you are saying about the shot pin though. There is a signal going to the pin when doing a spindle orient. However, that signal will need conditioned for the drive. Possibly just a resistor will do it, I will just have to see but I am pretty sure it is 120VAC.


I would prefer not to create problems when tapping. As it is now, a 1/2x13 tap struggles when coming out of material due to the very low torque.
 
I was able to verify that indeed there is a contact or limit switch associated with the shot pin that goes directly to the drive. Because the OEM is clueless, I have to assume the other side of this switch (N.O.) is being used as state change in the control to allow it to proceed with toolchange. Sooo....the whole mess and pony show is dependent on ONE limit switch.. That switch fails, there is a line of things that could happen including tool changer crash, destroy shot pin from over torque, etc.

But, assuming we go for broke, signal to the drive to halt (or coast stop) is given only with the delay imposed by the switch which I will assume is deadly fast being a limit switch. So let's assume 10-20ms, Could the drive really even react in that amount of time to a stalled spindle? Speaking generally of course.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor