PLC 5 and MOTION CONTROL

Hello all....
First off, this place looks very Kewl and I look forward to spending some time here. I have been around many boards like this with many topics but this is te first time I every hit one for PLC's! Kewl!

Ok...

I am looking to do a little motion control unitizing PLC 5 (5/80) and a laser.

All safety aspects aside (although strongly considered), I am looking for some example code and general thoughts and ideas.

·This machine travels smoothly about 1000 feet both directions on
rails
·Machine is DC driven with collector rails to power motor
·Communicate to the Drive 1203 Remote I/O Communications Module
·I believe the feed back from the laser is 4-20 but I might be
mistaken. (Still on the drawing board)
·I need to position this machine in many places with in the 1000’
with a tolerance of about ½” or less
·Naturally I will need to accelerate the machine, and slow it
gradually to the stopping point.
·I really wish I was using Rslogix 5000 but I am not

I appriciate any feed back and look forward to helping out anyone that I can as well.

Thanks all

You are joking, I hope.

The laser feedback for 1000 ft of travel is a poor idea. The 4-20ma is also a poor idea.

I will leave the why as an exercise.

Stick to your guns and stick with the Control Logix solution.

Hmmmm

Ok.. the Control logix is not an option at this point. Even it is cheaper, faster and has far better motion control blocks "built in"

"The Why"

Well as I see it... the resolution for 4-20ma over 1000 ft is poor. Even if I went to a 0 to 5 vdc feed back it still won't give me a ton of resolution per foot let alone a 1/2 inch. I understand this and agree.

An encoder over this distance can prove to not be repeatable.

I even thought that maybe every ten feet, reset the encoder, or laser and reduce the resolution to a 10 foot span.

Many points to ponder....

But.. how would this motion logic look in PLC 5? I mean you can set up preset points to travel to but I am a bit confused as to the logic in PLC 5. Continue to move with a
"If actual input is less then preset continue to travel at a preset speed?"
Seems a bit primitive...

Thanks for the input Peter!

I have a solution. Maybe solutions.

Do a google search for "SSI Ethernet motion control" at least one of these will interface nicely to the PLC5 using the MSG blocks. At least one will work nicely with RSView over Ethernet too. In fact at least one can work with RSView without the PLC5! At least one will show up on the RSLinx.

What is SSI?

SSI is a feeback interface that can provide 24+ bits of absolute position data. That can easily handle your 1000 ft with better than .001 inch resolution.

See Stegmann Absoluite Encoders with SSI interface

SSI is much better than the analog input for position feedback.

The answers are out there. I am watching the X files now.

Ruling out the Control Logix is a BIG mistake.

Ruling out the Control Logix is a BIG mistake.

Click to expand...

Some products already bought prior to beginning of actually engineering project...

Ok... truth be known... My dang hands are tied...

I have concidered absolute encoders... I have not seen SSI for some reason. Intersting... I am going to be reading... Alot more.

The Motion Coordinator looks like it will take all the guess work out of PLC 5 logic...

Thanks for the direction Peter!

RSVIEWRULZ

First of all, I have now experience of PLC 5's (I only know Siemens and GE logicmaster 6), but I do have experience of positioning large rail mounted gantry cranes.

we use Stegmenn Absolute encoders (C-Pomux Linear type) for positioning our cross travel (24 metres total travel) and I can vouch for their reliability and ease of use. We did look into using these type of encoders on our long travel (550 metres) but what put us off was the price, it became too expensive. At the moment we use an incremental encoder for the long travel and we position to within 10mm accuracy. We do have a home position and we do position check about every 10 metres, this would not be necessary with an absolute encoder. If using an absolute encoder proves not to be right for you then using an incremental encoder would be the way to go, providing you do some error checking for the position, then you can get some very satisfactory results this way.

If you are interested in how we do position our cranes, including acceleration and deceleration, let me know and I will give you a description of how this is done. It might give you a few ideas on how to go about your project,

Paul

Hello Paul,

Yes by all means.. Any new ideas are read into for new possibilities!

[email protected]

Here is my delema with using an encoder.

The Machine, Car or even called a Train carries a bin with about 1700 pounds of power in it. It is not a very large machine but the foot print is about 4'x4' about 6' high. compacted as small as possible. Not much on board room.

When it was designed (1980's) the drive and control were mounted remotely and the power to the Car is through 2 collector rails. (DC)

If I go with an encoder or anything "on board" I have no easy way to communicate to the PLC with it.

The cranes I have worked on that were "automatic" via PLC control had laser guidence to operate. This is what has lead me in the path I have been on. I really like working with Absolute encoders as they are not only programable, they are smart.

Looking at an upgrade currently happening at a friends plant in Tenn. he told me about a laser they are using.

http://www.trimble.com/ics5000.html

Seems like it might solve my problem. Still tons of research to do! Including cost of this handy little unit.

I really like the input I received here. Fantastic to have a network of thinkers.

Any other input is greatly appriciated!

Thanks!

One supplier of laser distance measurement is SICK.

The model that seems to suit you best is DME3000-1
Range: 0.1-500 meter
Accuracy: +- 5mm.
Outputs: SSI, Serial, Profibus (maybe not all in the same unit).
Price: approx €/$ 3000.

Why is a laser range finder not a good idea?

The trimble unit will not do as a sole source of position feed back to a servo controller. The feedback rate is too slow for a servo controller.

Refraction due to thermal gradients in air will also cause the feedback position to jump around making the cart jump.

These laser range finders can't be measuring the speed of light because at 3.2 x 10^8 m/s light will travel .320 m/ns. This leaves trianglization which is difficult to do unless the source and detector are far apart and as I indicated above the thermal gradients will be a problem.

The laser would work as a coarse position indicator for a servo system using a quadrature encoder for fine position control.

Many years ago we made laser alignment devices for another company according to their specifications. We beat the specifications in the lab. However when the laser alignment equipment was used out doors there were all sorts of problems due to thermal gradients. It turned out that the alignment device could only be used when there was no wind and no sun. The product was not successful.

Points are noted Peter... Thanks!

Not sure on the enviroment that you have or the nature of the area but I have heard of using wireless ethernet or radio type communication for data collection.
As long as enough other safegaurds are meet this may be they way to go to use an encoder. So that eliminates the communication problem.

Assuming that is the only reason for not wanting to go that way.

Drewcrew6

RSVIEWRULZ,

To keep the information within this forum, I decided to post my reply here rather than send you an email.

Below is a brief and simplified method of how we position our cranes on long travel.

The incremental encoders used are the 2500ppr type. These feed into a high speed counter module, via profibus this gives us our encoder feedback to the S5 PLC for crane positioning. The encoder counts are scaled in the PLC program and converted to millimetres.

A the ‘home’ end of the long travel we have what is known as the home flag, this flag (a 5mm round steel bar) is mounted on the rail sleepers and is about 12” high. On the side of the crane there are mounted 2 sensors, one at about 11” from sleeper height and one at 8”. The higher sensor is the home flag sensor, when the crane crosses the home flag and this sensor is activated the encoder counts are reset within the logic.

Spaced at roughly 10 metres apart, the whole length of the crane travel are more ‘flags’, these are about 10” high and are designed to trip the lower sensor, which we call the recalibration sensor.

The position of these flags is measured and stored within a data block in the PLC.

E.g. Flag 1 = 10.15metres
Flag 2 = 19.80metres
………….
Flag 54 = 541.20metres

As the crane travels the length of the track it will regularly cross these flags and activate the ‘recalibration’ sensor. When the recalibration sensor is crossed, the encoder reading is compared to the position stored in the data block. If the position error is within our acceptable limits (the ‘window’ of error is + or – 15mm) then the crane is allowed to carry on running, if the error falls outside this ‘window’, then the crane is slowed to a stop and has to ‘rehome’ itself, by travelling back up to and crossing the home flag. We have also programmed in other safe guards, like loss of the encoder feedback or if the flag sensor does not get activated within a certain time (distance travelled).

The crane is sent a ‘target’ position (by radio link) and using this target position we then set the speed reference that will be sent to the drive, the acceleration and deceleration ramp times are set within the drive. If the target position is more than 20metres away from the actual position, then the drive is sent a 100% speed reference, this speed reference is incrementally decreased as the crane gets nearer the target position, until the crane is positioned within our ‘in position’ window (10mm).

If anybody would like anymore information on this method then let me know.

We are currently using Siemens radio remote I/O links called 'profinet' and have had excellent results with this technology, as drewcrew6 states, this might be something you could look into.

Hope this has helped in some way.

Paul

Personally, I also like the SICK DME-3000 Lasers to control my automated hoists. I have tried quite a few systems and have come to like the DME-3000 the best.

I do a lot of work in the plating and metal finishing industry and due to the corrosive atmosphere and equipment abuse, I try to eliminate any type of prox or limit switch where I can.

With the SSI signal, as was pointed out before, I have set my accuracy to 1 mm which is more than enough for a big crane. The signal has always been repeatable,(I have some lasers that have been running for over 5 years without a hitch) and no limit switches to worry about homing and constant replacement.

The programming in it's short form is once the laser is set to your scaling requirements is: 1.) Get signal 2.)Use SCP with laser signal and desired location 3.)With Analog output to VFD, run machine at 100% till 4.)Another SCP to scale from 100% to 0% with desired ramp down distance to target location.

I have used this type of system for quite a long time now with excellent results though my longest line has been 400' the laser should work just the same at 1000'.

Good Luck,
Chris

Hey Paul, did you ever come up with a solution for your "missed flags" problem we were discussing over at Ron's site?

beerchug

-Eric

Eric,

That little project is still 'on going', we have 'in principal' decided to go with the passive tag idea. I have the majority of the 'paper' work for this project completed (code and schematics done), but those here who run the engineering department are forever shifting priorities and at the moment I have been 'lumbered' with 2 ship to shore driving cab refurbishments (2 - 3 months work on each one) and I am still working on the removal of the 2 multiplexors that are on our ship to shore spreaders, plus all the other stuff that seems to find its way into my in-tray. Maybe one day the 'tag' system will end up back on the top, but as this will only have a minor effect on the 'down-time' of our cranes it is always going to be put onto the back burner, until such time as I only have that left to do.

But, hey I am not complaining (much!) it keeps me employed!

I just wish that management would let me completed one project before insisting that I start another!

Paul