High speed linear actuator required

[khonfahm]

Hi all,

I'm basically after a linear actuator on steroids.

Here are the required specs:

0-80mm stroke, 1040N Force, 0-10m/s speed.

I have found a few products that would almost be suitable:

http://www.tacrockford.com/product/...motors/oswald-lins-cylindrical-linear-motors/

But they do no meet the requirements of opening at 10m/s...

Background of use of product:

I work for an electrical switchgear company. We are basically looking at developing some new linear switches, and require a mechanism that can be used to test some new concepts.

I've been searching and contacting companies for the last two weeks have pretty much just gone around in circles.

Do you have any advice on how I can approach this problem?

Thank you in advance.

 

[MikeHalloran]

Your spec is confusing.
You say 0-10 m/s, then suggest that you need a minimum speed of 10 m/s.
You say 0-80 mm stroke. Do you also mean a minimum of 80 mm stroke?

The context suggests that you do.
I.e., you want to propel some mechanical doohickey by 80mm in 8ms or less.

If it only has to work once, explosives become a possibility.

If repeated operations are a requirement, then you need to look at mechanisms that store energy at some slow rate and release it at a much faster rate.
I think air rifles would be a good place to start.


Mike Halloran
Pembroke Pines, FL, USA

 

[khonfahm]

Apologies for the confusing spec - and thank you for bearing with me in trying to suggest a solution.

It will need to be used mulitiple times and is not a once off solution.

10 m/s is the opening speed for an 80mm distance.

If it is a 70 mm distance, the opening speed required is : 8.75 m/s.
If it is 60 mm; opening speed is: 7.5 m/s

Basically it needs to complete the opening motion for a specified travel length varying between 2-80mm, within 8milliseconds.

As the gap reduces - the required opening speed decreases.

In the worst case scenario: it needs to open 80mm in 8 milliseconds .. which means the mechanism needs to travel at 10m/s.

 

[blacksea]

Except speed 10m/s, you need calculate acceleration/deceleration also. And ones maybe very high also - your speed profile even can't reach such speed! It seems that the only possible solution will be linear motor based actuator.

 

[ScottyUK]

Given that this is a switchgear application, is there a reason why the tried and trusted stored energy system using springs and a charging motor can't be employed? As a switchgear end-user, spring tripping mechanisms are simple and have low external demands for the tripping supply - high power motors sound like a headache and a liability.

Sorry if that sounds negative, it isn't intended to be, I'm just trying to understand what your objectives are in this exercise.

 

[LionelHutz]

If this is just for testing then have you looked at air or hydraulic cylinders? You'd likely need to create a fixed cylinder travel operating a level with an adjustable motion ratio to allow your varying travel requirements. Either that, or a very rugged adjustable stop.

 

[mikekilroy]

hmmmm... ok, buy a linear motor to do it.

you just need 40,000# (140,000N) thrust for the 221,500 in/sec/sec (5,624m/sec/sec) accel & decel rate to make that 8msec trapazoidal move of 80mm. oh, ya, your top speed is 590.5 in/sec (15m/sec).

but good news! at least you won't need hearing protection against a sonic boom - speed of sound is 13,397 in/sec!

 

[mikekilroy]

I forgot to list the accel in G: 573.4G

 

[mikekilroy]

(fun_on, with physics 101)

Just one last spec: at the peak of the 2msec accel ramp, putting out this force and speed, will take approximately 2,110,000 watts (given a typical linear motor).

that would be 220,000 vac 3ph @ 294 amps assuming the linear motor can handle these values.

(/fun_off)

You won't find a linear motor that can do what you want.

 

[mikekilroy]

Since the application is not a single one shot move (no pun intended), maybe you should check out how much energy and speed is in a moving bullet; seriously. a .45 bullet is still under 50 cents each, and I can tell you a plinking target gets rotated real nice and fast when hit. maybe you can just put a 3/8" thick hardened steel 'handle' on the device to move? It could even be fun to shoot at! Mike Halloran is probably leading you down the right path!

 

[nornrich]

khonfam,

If you're looking for an actuator with this level of Force, Speed and Stroke and you need a range, you might want to take a look at LinMot. They are an excellent servo controlled, linear actuator.

Regards,

Rich.....

Richard Nornhold, PE

http://www.energizer.com

 

[khonfahm]

Thanks for the suggestion ScottyUK - for the actual device - we plan to use the trusted stored energy system.

Right now while we develop the interrupting device we need to develop a mechanism to do the testing for various configurations. With the stored energy system, there is not much room for varying the mechanisms specifications once it has been made.

Hahaha - "won't need hearing protection against a sonic boom"

- thank you all for the input. Honestly, thank you so much for the suggestions.


I'm glad I put this question up. It definitely helped me clear up that it's not a very simple solution. I put it up, incase I missed something very obvious. I've been looking into servo controlled linear actuators in more detail. I think the concept of rifles/bullets will require too much work to from an OH&S(occupational health and safety) perspective.


I may have to adjust my requirements to see if I can still 'test' the new products we plan to develop.. anyway, I'll hopefully be able to adjust my requirements once I bring this up with the switchgear engineers.

Thanks guys.


Right now this is what I'm leaning more towards: A reciprocating engine with a clutch type mechanism. The motor will be able to be started up and will be spinning the output shaft at the required speeds. As soon as I need it to engage and open the mechanism; It's a matter of 'dumping' the clutch, and hopefully with that the acceleration will be achievable.

 

[ScottyUK]

Hydraulics can achieve some pretty rapid movements, especially if you use oil one direction and a spring in the other. When the spring moves the actuator the speed is largely governed by how fast oil can be vented, and that can be pretty quick. Anecdotally, governor valves on a turbine typically use this sort of system: when the hydraulic oil is dumped in a trip situation the valves close very rapidly considering their size. That's from a system designed principally for precision control rather than pure speed - I'm sure it could be optimised for the latter.

There are some very knowledgeable guys in forum1083 who could tell you more than I can about where the limits are.

 

[mikekilroy]

As an EE I admit I don't recall the name of it, but you might also consider the old railroad wheel drive mechanism - a wheel with a slot in it that moves a linear rod.... If I did the math correctly, a 10" dia wheel with a slot in it or hole to attach the rod to, rotating at 7,500rpm could make that 80mm move in 8msec also. Not sure how you would make it live long but maybe that is not an issue to a good ME designer for the short time you need to run each test?

A free motion sizing program such as Motioneering by Kollmorgen can help you try different types of actuators such as linear motors, rotating wheels, etc:

http://www.kollmorgen.com/en-us/service-and-support/technical/motioneering/

 

[waross]

Some hydraulic systems use compounding for a fast return. When a retracted cylinder is to be extended rapidly, pressure is applied to both ports. The effective cylinder area becomes the area of the rod, rather than the area of cylinder. The travel speed is increased by the ratio of the cylinder area to the rod area.
If you are considering clutches, look at the flywheels and air operated clutches used to drive alligator shears such as are used to cut re-bar. If you go that route, look for a high slip motor to drive it.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[khonfahm]

Thanks again for your response.

I will look into the Motioneering Program today.

In regards to designing the hydraulic system... I really don't want to have to go into the specifics too much myself. Ideally I would like a company with the experience in making them to specify what I need for my requirements.

For Example, it's like buying a car: I want a car that does 0-60mph in under 6 seconds. I am not going to learn everything about forced induction or v8's (etc)... I just want a manufacturer to tell me; "here, our VW Golf R meets your requirements, it uses a turbo charger on 2.0L motor".

Does that sort of make sense? I want to have a general idea of how the mechanics work... but not in the full detail that would require me to design the entire mechanism myself.

 

[MikeHalloran]

How many millions will you be needing?


Mike Halloran
Pembroke Pines, FL, USA

 

[khonfahm]

You mean quantity or are you referring to money? Haha.

 

[MikeHalloran]

Either way.


Mike Halloran
Pembroke Pines, FL, USA

 

[khonfahm]

I don't know why I didn't think of this earlier. The solution was staring me in the face....


I can get a high speed electric cylinder/cylcindrical linear motor... (

http://www.tacrockford.com/product/...otors/oswald-lins-cylindrical-linear-motors/)

couple this with a effective lever type mechanism, and get the required opening speeds!

So if this thing can open at a maximum of 1 m/s... then I couple it with a lever that moves 10 times the distance for every 1m this mechanism moves(

http://images.tutorvista.com/cms/images/38/lever.PNG)...

and then Boom! I have a device opening at 10m/s!
Only issue is the applied force, but my force requirements are quite low in comparison to what can be outputted in these devices!