two motors to run at same rate/different gear ratios

[RandyO]

I need to have two pieces of equipment to run in sync. although one motor runs with a 5:1 gearbox and another has a 15:1 gearbox.How can I make these two machines run in sync??
I'm needing to use an encoder to track revs, With ppm, how do I wire the encoder to the PLC AB 1200 w/2 in/2 out analog I/O add on module.
I only need z pulses, then I believe I need to write program and make a numeric value(ratio) to send the 0-12 v signal to the vfd on other motor. I understand ladder logic but this something I haven't had to do yet.

This appears to be easier::::
Or can I use the 0-12v signal from the first vfd and reduce/increase that value incrementally through the PLC(program) to make a signal that would go out to the other VFD. these two machines need to start(ramp up and down in sync)
Compensatimg VFD the 0-12v signal does not make this happen, the two pieces of equipment are too far different in the gear ratios
Thanks
RandyO

 

[DickDV]

In my opinion, I would leave the PLC out of this and simply run the same speed reference signal to both drives (0-10v or 4-20ma typically). Then set the drive with the 5/1 gearbox for the minimum and maximum motor speed you need to get the minimum and maximum machine speed you want.

Once you have that set how you want it, simply program the minimum and maximum motor speeds in the second drive (with 15/1 gearbox) for exactly three times the values you put in the first drive.

Now, with the same speed signal, the second motor will run three times the speed of the first motor at all speeds.

There are other more precise ways to synchronize speed but this method is usually good enough especially if the machine loads are similar. In fact, using basic V/Hz drives with nothing more than slip compensation, if one machine is running at nearly no load and the other machine is running at nearly maximum rated load, you should see no more than 1-2% speed error assuming the motors are designed for 2-3% slip.

 

[RandyO]

I've tried that

Dave, the plot thickens
There are other factors: the 5:1 gearbox drive runs an 10 inch rubber wheel that is perched atop a roller array, that is used to pull a continous polyester strip(3 mil thick x 3" x 16,000 yards) around a multi roller(2" diameter roller), head with varing diameters (an eight roller array with a bolt circle of 8.00", thus making the 10" diameter). This machine is used to make flexible air conditioning duct outer jacket you see in attics it looks like metalized jacket.
During the machines revolutions it has a 2.750" linear payout per rev.
The machine I designed and built winds up this jacket as it is revolving around at the same time/rate as the jacket is being made. The normal way of doing this is by means of a mechanical drive shaft.
The gearbox ratios are a factor, as would be the circumference of the drive rubber wheel, and the circumference of the jacket being made, thus making it more than a 3:1 ratio.
When tring to use the VFD compensation factor, when my machine runs at 60 hz, the other is at aprox. 10%+ thus derating the drives parameters, not good for production numbers. I will do the calcu's for abosolute ratio, but it far exceeds the VFD's ratio capability.
Can't I use scale to data in logix ladder???
Thanks,
Randy

 

[DickDV]

Randy, if I'm understanding you correctly, when the lead machine is running at 60hz, then the following machine must be running 10% or so faster than 60 or somewhere around 66hz.

As long as the motor and the machinery can stand an extra 10% speed, you can program the follower drive to 66hz and you will have the speed match you need. Just be sure to do exactly as I've described in the earlier post setting the minimum and maximum speed of the follower drive to the same ratio of overspeed compared to the lead drive.

Modern motors under 250hp can almost always tolerate 25% overspeed and often more.

I'm a bit baffled at your comment about processing the speed signals some other way and getting around the supposed drive limitations. If doing it in the drive won't work, how will doing it in the PLC make the drive run any faster?

 

[LadderLogic]

Randy:

Are you trying to synchronize shaft speeds or shaft positions ?
Since you have mentioned a mechanical drive shaft as a traditional means of achieving the same result, I suspect that just matching speeds would not do for your application. You need a true following, a master/slave setup. I am not sure this can be done just with a Micrologix...

 

[Xian]

Can you try this?

Set up a CMP for the speed refs for one drive (probably the faster turning motor would be best). Then you can throw on a tach to find out how much slower the other drive needs to go. Use the CMP to adjust/offset the speed ref to the slower drive. SpeedX *.857 for example. That way you can vary it on the fly to match speeds and verify it with the tach. This would also provide a means of tuning the speed relationship if the rubber roll wears over time.

Christian

 

[Xian]

my bad

sorry randy that should have been CPT not CMP

 

[Lancie1]

RandyO,

Certainly you may, in you PLC ladder logic, take an analog speed signal value from one of your motor drives, then multiply or divide it by another number, and send it out to the second motor drive.

However, if your speeds are too far apart, one drive may be forced to operate at a very low speed, which is not good for the motor over a long time period. If the speed is less than 25% of the motor full speed RPM, then I would consider swapping the gearbox, or swapping the motor for a lower speed motor.

 

[Dartrunner]

Can you make the wind up section VFD work as constant torqe to maintain just a slight tension on the material?

 

[Platootod]

My question is my answer

Do we need inverter? use with PLC

platootod

 

[Thomas]

other possibility

Perhaps a tension feedback device (e.g. a small takeup between the two driven sections coupled to a potentiometer providing a 0 - 10vdc out) would work better. Run the winder as master and the feed slaved. Too loose slow the slave, too tight speed up...

 

[rdrast]

/shrug

Frankly, if in normal production, one section is to be run at 100%, and the next section has to be run at 10% speed in order to achieve a match, then you need to change your gearing.

It doesn't matter WHERE you scale the speed references; in the drives, in a PLC, in a micro-controller, with a pot... With the current gearing, you will always have one VFD running very slow (or very fast, I'm not clear exactly on which).

 

[rwhite1157]

how is the motor coupled to the drive? belt, chain, direct? unless its direct it seems you could change the sprocket or sheeve and get the same speed. Just a thought... R.W.

 

[BarryP]

Hi Randy

I have experienced similar problems with this sort of installation and have cured it by running a rotary encoder on the material at the points where the synchronisation is required and where each motor is driving. That is to say I have put REs at the master point and at the follower points. I have done up to 3 followers or slaves off the master with the relationships being master to slave 1, master to slave 2 and master to slave 3. The REs are geared so as a single rotation of the encoder will give the same amount of distance of the material at each of the points being monitored.

The REs are all fed into high speed counters on the PLC which also needs to have analogue outputs for each motor, along with an analogue input for the master.

The philosophy is that the counters will count each pulse of the encoders, so the higher the pulse per revolution of the encoders, the better the resolution. The counting may start at zero and continue up to the max that the PLC can handle then start over again. So long as you dont lose any counts at the time of the zero transition it will be fine.

The master drive has it's speed sent as an anlogue input to the PLC, say 0-10v = 0-100% of full speed. This speed value is used as the base speed for the slaves.

As the master runs so the Master RE will start rotating and the counts will commence. Using comparison networks, you can determine the difference between the master and the slave counts. Using the speed value from the master you can then drive the slave at this value and increase it or decrease it (using maths functions or better still PID loops) to suit until the difference between the counters is zero.

Through the acceleration and deceleration phases you will find rather large differences if you are using straight variable speed drives. As I wanted to synchronise through both of these phases I used closed loop vector drives. These give me 250% torque at zero speed. The master was set with a long acceleration and deceleration characteristic and had each of the slaves had their accel/decel setup for maximum response. In this way the slaves are made to be very responsive to the speed signals. If this creates a problem with jerking, the setting of the S curve parameter in each of the drives will see this disappear.

I have successfully built and installed 3 of these systems 2 here in Oz and one in Italy. I use the same philosphy and system for several other scenarios. I have built them using Siemens S5, S7 PLCs and also on the GE Fanuc Versamax units.

Hope this helps

 

[danbee]

Check out Fenner M-trak for drive sync.