AC Dual Motor Sync, Ramp Up/Down, Dual Limits... 4

[ozdemir23]

Hello. I am in the minority in our company regarding the design of a system that I believe is not as complicated as everyone keeps saying or believing to be true.

First off, I am not an electrical engineer but a structural engineer by education and practical engineer by life.

We have a large mobile frame that moves laterally along a set of parallel tracks. The tracks are separated by about 80 feet. The motion of the frames can be done manually, but the momentum and forces required to stop the frame are so large that we need to control the motion with motorized system. Here in lies the challenge.

Two motors need to operate at the same time (on/off), same speed (synchronized) and have the same inputs for ramping up and ramping down along the tracks. If one motor is not in sync then there must be a fail-safe (shutoff, slowdown, reverse, etc.) to make sure the frame is not twisted.

We had someone say they could design the system to do exactly what we need, but then 8 weeks and several "I'm sorry about the delays" later, we have a system that doesn't do what we need. Here is the current design:

(2) 90VDC motors Bodine 42A-CG Model 4786 (Rpm = 42rpm (need to adjust to max 30rpm); Torque is 580 in-lbs)
(2) Dart Encoders PU-2EQUAD-Z2552 at motor end (extended shaft)
(2) Minarik Drive Units RG60U-PCM (control panel)
(1) EZTOUCH PLC EZP-S6W-RS-PLC-E which includes
.High Speed 2-Counter Module EZIO-4HSCM1 in slot M1
.8 DC Input Module EZIO-8DCI in slot M2
.4 Analog Input, 4 Analog output (Voltage) Module EZIO-4ANI4ANOV in slot M3
.4 Isolated Relay output Module EZIO-4RLO in slot M4
.24 Volt Power Supply
Source is 120v

This design does the following:
.Syncs motors to prevent frame from twisting by counting pulses and comparing

We asked for the design to do the following:
.Sync motors to prevent frame twist
.Ramp up @ start, ramp down midway through travel distance, then ramp back up, then ramp down 2 feet prior to end limit;
.Program a home/start and an end limit
.Make reversible - Turn switch to open position for travel in one direction (from home to end); Turn switch to close position for travel in reverse direction (end to home);
.Release spring loaded switch to position in neutral to stop both motors at any point in between home and end
.No external limit switches or other components outside of the motors, control panel/box, and spring loaded and key operated switch;
.Include failsafes against obstruction (detect increase in amp draw to shutdown)
.In case of power failure, the system should be able to maintain all settings without having to reprogram limits or ends

We were not told it couldn't be done or that it was going to be very difficult or anything like that. But, it turns out they couldn't get the job done as promised.

Is there anyone here that can help solve this challenge?

 

[dpc]

You might look at how center pivot irrigation systems work. I believe they use strain gages that senses strain on the connecting pipe to control the speed of each hydraulic motor. Just a thought.

 

[jraef]

This is common in bridge crane systems, referred to as an "anti-skew" control system. It is NOT for the feint of heart, it's more complicated than it seems on the surface, involving the use of PID loops to avoid over and under shoot of the response from the motors and slippage/wear of the drive wheels. I tackled one once on a project for Boeing, I had to absorb a huge learning curve cost... I strongly suggest contacting someone with experience (not me, I'm too expensive now).

LMGTFY


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[ozdemir23]

I am not opposed to paying for an electrical engineer to design the system with all of the variables or expectations laid out up front. I'm not sure what you suggested for the crane control system is what will work for us. The application involves an aluminum frame that is at grade in a swimming pool environment. So the frame system and the components that would be part of the anti-skew controls would be at ground level and susceptible to kids or otherwise vandalism. I cannot have that as part of the system compromising the design. Even so, I have to assume that a communicating system (feedback) along with a brain could handle this but like I said before, I'm in the minority. Maybe it hasn't been done or can't be done.

This is not for a small DIY. This is for complicated and very expensive systems. I am looking for someone to say that what I want is possible and not asking for the world. If possible and just needs someone to design it with confidence and reliability within the system, then let's talk.

I just don't think this should be that difficult. AND, if it hasn't been done before, isn't this the type of challenge that electrical engineers like to solve?

Anyway, thanks for taking the time to read and I appreciate all of your feedback.

 

[davidbeach]

Reread jraef's post. It's done for large bridge cranes. Regularly done but far from trivial.

 

[mikekilroy]

Antiskew control is hard more difficult at high speeds but not too bad to handle at slow speeds like you aeem to mention.

On large machine tool gantries we often need to run upto 1200 IPM with motors 30-40-50' apart, and do controlled synchronized stop if the skew exceeds about .o20".

So yes, it can be done and is done daily. No external switches all over the place. You just need the proper servo products and people as you have determined and Jraef has said.

We do this with Kollmorgen.com servo motors and drives in your sizes, and many other brands are available to do so also.

http://www.KilroyWasHere<dot>com

 

[sreid]

String encoders both sides, Vector drives both sides, Master Slave configured.

http://www.te.com/usa-en/products/s...paignId=8421&gclid=CPbo9PSQ9dACFd26wAodTz8Cfw

 

[ozdemir23]

Sreid. Can you expand? What or how do string encoders work? Vector drives? What about ramping up and down? Obstacle detection? Master guides slave? Sounds like you know what we need. Plc? I need someone to design, supply and deliver.

 

[mikekilroy]

As long as you have non-slip rotary to linear translation (ex: rack/pinion), then there is absolutely no need for any external position feedback devices like string pots. If your motion system allows slip (ex: steel wheels on concrete) then you will require position feedback at the end location (ex: string pot, laser,etc).

Servo motors often today include absolute encoders inside their housing. This position feedback device keeps track of position - even if power is off. Move gantry with power off, power back up, and it knows exactly where it is.

As for obstacle detection, you need to know what you want. If this means you must detect BEFORE hitting something, you need a camera with detection capability, or some sort of non touch ultrasonic etc sensors. If you just want to detect AFTER hitting something, then your current sense may be sufficient. For design, you will need to discuss this in more detail with the designer engineers.

Where are you located? Determines who may be close to you to do the work you want.

http://www.KilroyWasHere<dot>com

 

[ozdemir23]

I am located in the Chicago land area.

I don't need obstacle detection beforehand so the current sense is all i will need, but it should not allow the skew to be too great. I am not looking for a skew tolerance within thousands of an inch like our heavy machinery in the shop, but more liberal - within 2" let's say.

The motion is controlled with a non-slip drive - sprocket and chain drive.

There was an option that was considered but never materialized to eliminate the need for feedback. We were going to use a single motor at the center of the frame and then a hollow shaft allowing for a shaft to span the entire width of the building. But, that only takes care of one issue with regards to feedback and it looks ugly and it doesn't eliminate the need for the other variables. Plus, our last project just included a situation where each side was not at the same elevation so there would need to have been pulleys and other components that could potentially be a service issue down the road.

What are the drawbacks to servo motors? What is the difference between the motors we are using and the servo motors?

In the past, this might sound crazy, but we used tube motors from Somfy designed for shade controls. These motors have a built-in encoder and allow you to program upper and lower limits with a low RPM of about 20-38. The rated torque was not limiting us but the feedback/syncing of the motors was an issue because they were just hardwired with one switch. The wiring for the switch was not reliable and was the main reason for changing to a more robust system, but otherwise EVERYTHING else was taken care of. It had obstacle detection, heat fail-safes, limits for upper/lower and intermediate, power loss did not result in loss of limits or codes, and everything was contained in one single unit. If something was wrong with the motor or limits or anything, we just swapped one device for a new one. Easy setup of new limits for new device and off you go! But, again, reliability was not there. The switch was nothing fancy and since we are selling a luxury moving building that will be for public use, we need our system to exhibit the confidence to their electrical engineers and staff that the system is reliable and will work 10-20 years. Confidence...

Let me know your thoughts. I think this is very manageable and I'm not asking for the world.

I understand my topic subject says AC motor but we are using DC motors here. I'm sure you all caught on to that, but wanted to clarify in case it wasn't obvious.

 

[ozdemir23]

I also wanted to mention the duty involved in terms of cycles. The motors won't see hi speed or hi use at all through out its life cycle. On a daily basis, if the building is opened and closed 5 times, that means the motors were ran about 3 minutes each time for a total of 30 minutes a day spread out over the day. I'm pretty sure that means these motors won't be the bottleneck or considered to be a potential for failure in the overall system.

I hope this helps...

 

[mikekilroy]

"What are the drawbacks to servo motors? What is the difference between the motors we are using and the servo motors?"

Only "drawback" to servo motor is more cost. But if the stuff you have now is not doing the required job, is a bit more cost a drawback?

Your Bodine Dc GEAR motors with Dart encoders are very similar to today's servo motor. The Bodine has brushes, today's AC synchronous servo motor does not. Your brushes have a life of probably 2000 hours.

Servo compared to sreid's VFD/plain-induction-motor/string pot - certainly a doable solution too has many advantages in controllability, simplicity, and much more precision control. The servo will be perfectly happy holding your skew to .02" without effort. The servo will detect extra load many times more accurately. The servo will be much more robust without the breakable string pots. The servo will likely cost the same since it will not require the position option card in the VFD or the expensive string pots.

If it were MY design I certainly would NOT use a GEARMOTOR like you have! Think about it: your 100+:1 gear ratio makes any load change nearly unmeasurable! If it takes 200#-in@2amps to move, and an obstruction causes this to go to 300#-in(3.3amps), your 100:1 gb makes this down in the mud to detect accurately!

Anyway, you are asking for help. Your redesign cannot be done over the net on a forum. There are a zillion design detail questions that are unknown for someone to pick the proper hardware for you. You should enlist someone from the Chicago area. There are folks that can do this for you around there. Perhaps if it is not against rules of the forum you should post your phone or contact info so one of those may offer you their solutions? I am sure I am not alone in being happy to suggest some people near you that I would think could offer good solutions.

http://www.KilroyWasHere<dot>com

 

[ozdemir23]

Mike-

I completely understand regarding having this designed over forums. It was not my intent. I wanted to make sure I wasn't asking for something only NASA would do, and it looks like I'm not. It appears as though this has been done before and we just haven't talked with the right people.

Cost is not the driving variable, but is always a consideration.

I don't want/need the tolerance to be within the 0.020" range because the system is flexible and can handle a skew, just not 5"-10" skew.

I do find it interesting that you're talking about the gear motor not being your choice for design. Our designer suggested that the upper limit or obstacle detection limit of 3.5amps was going to be sufficient to create the shutoff, but I could sense his answer was not delivered with confidence. Since it is a field set condition, he was allowing himself an "out" if it didn't work like designed or expected.

I am open to having someone call me or I could call them. Instead of publishing it in the public forum, perhaps contacting me via private message would be the more appropriate method.

 

[mikekilroy]

Very good. I mentioned posting contact info as I am not aware of any PM capability on this forum.

http://www.KilroyWasHere<dot>com

 

[ScottyUK]

ozdemir23,

Rather than putting your contact details out in the wild, maybe an email to the contact address on Mike's website would be worthwhile? ;-)

 

[mikekilroy]

Thank you Scotty. He did. After talking to him I told him I would summarize and pass on his info and need to a local Chicago area engineer I know who can help, as well as a PLC wizard here who travels. I did not want to just be the only one offering names of folks who might help him. As Jraef said, I am way to expensive for his commercial need, but I, and others, certainly can offer alternative contacts to him!

In summary, I see he needed to know this is perfectly do-able, then he needs to find someone to do it.

I believe he understands now his requirement is VERY doable (not even requiring exacting anti-skew super algorithms), now he just needs to find someone to do it. I suspect he can reuse all the present hardware and just needs a PLC wizard to make it work in a couple days.

http://www.KilroyWasHere<dot>com

 

[ozdemir23]

Thank you everyone who helped me understand the things I don't understand. Educating and listening and asking the right questions yielded the results I was looking for.

Thanks Mike and Scotty and jraef.

 

[donbaughan]

There are many ways to solve this problems, but as you have found for yourself, very few effective methods at the drive level only. There is a method available today using standard, off the shelf products, Elmo Gantry solution,

http://www.elmomc.com/Solution/Elmo-Gantry-Solution.htm.

This will meet all of your requirements without any additional hardware, software or the need for an upper level or master controller. Feel free to contact Elmo directly for any further details or information, or if there are any questions.