Super High Speed Actuators (30-35 ft/sec)?!

[sgs550]

I'm looking for an actuator that can achieve speeds in the range of 30-35 ft/sec (360-420 in/sec) and handle a force in the 100-200 lb range.
I was considering a high pressure air cylinder type actuator but most of the companies I talk to say their cylinders can't go anywhere near this fast.
I found a picture of a setup similar to what I'm trying to build, and it looks like they're using some kind of pneumatic setup. Open to other options too though.
Does anyone have experience with super high speed actuators, or know of a product that might work for these high speeds?

 

[3DDave]

Doubling up pulleys can convert low speeds to higher ones at the trade off of using larger actuators to get the desired force.

 

[IRstuff]

How much stroke?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[hydtools]

You could do it with hydraulic power.

Ted

 

[CWB1]

I dont have much experience with electric actuators, but that speed and force is entirely doable with either pneumatic or hydraulic cylinders or a simple servo and ball screw.

 

[Jboggs]

You didn't mention several factors, including orientation, precision and repeatability, motion control requirements, price range. Just based on what you said I would probably first look at a belt style actuator in which the load is attached to a continuous belt which in turn is driven by a rotary motor. Several companies make them such as Bishop-Wisecarver, Parker, Tolomatic, and more. A water jet company I used to work for had a line of machines that used very high speed linear motors. So depending on your budget, that could be an option.

 

[hydtools]

Time to accelerate? Or distance to accelerate?
Time at speed?
Horizontal? Vertical?
Stopping?
One shot? Occillating? Frequency? Repetition rate?

Ted

 

[IRstuff]

Pneumatic cylinders can achieve extremely fast strokes; the design of the tank, piping, and valving is critical
> Aircraft carrier launch systems
> pneumatic cannons
>

https://en.wikipedia.org/wiki/Captive_bolt_pistol

> nail guns

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[EdStainless]

Air could do this (large lines, low restriction valves, receiver tanks) but it will not be super precise, after all the air is compressible. But will provide a little cushion against shock.
It would take careful balance on cylinder size. A larger cylinder is lower pressure and better flow, but more mass.
One problem with high speed air systems is that air cools when it expands so the pressure drops.
I recall seeing pneumatic cylinders undergoing life cycle testing at speed of 25ft/sec. The were SS body with crimped ends (SMC, Pneumatics, Bimba and such) the testing was being done using bulk nitrogen and not air. the ultra low dew point assured that there would not be condensation.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[LittleInch]

Speed and force mean nothing without that other data:

To paraphrase others here the critical items include:

Stroke length
Time or distance to reach speed, distance or time to decelerate
Positional accuracy
Is force constant or varying
What is the mass of the thing your force is connected to?

Give us a DIAGRAM


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[sgs550]

Thanks for the replies.
Budget is not a major concern; I'm looking for the best option in terms of performance (reliability and repeatability, in terms of achieving the same speed each time).
The stroke length would only need to be on the order of 6-12 inches. I added some labels to the picture, so hopefully that helps with some clarification of this application.
It's a one time deal. The cylinder simply needs to retract once at 30-35 ft./sec.
I don't think positional accuracy is an issue. I just need to measure the force required to uncouple a connector, one side of which would be attached to the piston.
The connector would be relatively low mass, I imagine less than 10 lbs. But the force to uncouple the connector could spike momentarily in the 100 lb range.
Once it's uncoupled (which would happen very quickly at 30 ft/sec since it's only an inch or so to uncouple the connector) the force on the cylinder would just be the mass of the connector.

As far as time/distance/acceleration, I'd like it to achieve the speed very quickly, within 1/2" of travel if possible. In the picture below, it looks like they allow 3" or so to decelerate by placing a piece of foam beneath the part being accelerated, so I imagine I'd do something similar.

So far I've been considering a "double purchase rodless air cylinder" that Tolomatic makes. It has an integrated pulley system to double the output speed.

https://www.tolomatic.com/products/product-details/double-purchase-pulley-actuators

They also suggested a belt drive actuator that's rated for 200"/sec. Could possibly mount a gearbox to the motor to double the speed to the range I need, but I'd be concerned about how it would hold up to being overdriven.

Thanks for the tip on using Nitrogen. We do have dry nitrogen available, just need to check what pressure it's supplied at. I guess alternatively we could use a tank of nitrogen, which we might need to achieve a high enough pressure anyways.

 

[EdStainless]

If you want it faster use He. The piping is a pain because of leaks but it will expand faster. Look up the speed of sound in different gases. (H is even better except for the flammability issue)

And electromagnetic actuator would be fastest, but they tend to be very short stroke.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[handleman]

Little sanity check here....
You are requesting final velocity of 30 ft/sec within 0.5" travel.

a=(Vf^2)/2d

where Vf = final velocity, and d = travel distance.

Being careful with units, this is roughly 170g accelearation. Nobody gonna quote you that.

 

[MintJulep]

Springs and a remote-release latch.

Compress and latch with something mundane. Hydraulic, acme screw, etc.

Release the latch and Newton is your friend.

 

[IRstuff]

hmm, I got 335.7 g

(360 in/s)^2/1 in ==> 129600 in/s^2
one g is 386 in/s^2, so in the 300 g range

Can the OP describe the actual test? That might easier than randomly looking for actuators

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[sgs550]

The test involves measuring the force required to disconnect/unmate a plug from a receptacle. The main requirement of the test is that the separation needs to occur at 30'/sec.
The attached picture shows a lab's setup for performing this test. I'm trying to get as much info as I can out of this photo and replicate a similar setup. They state on their website that they achieve speeds of up to 382 in./sec. within 6 in, so maybe the 1/2" is actually 6"!

https://www.dtb.com/testing-ordnance-lanyard.php

 

[sgs550]

The test has been performed in the past by dropping weights from the appropriate height to disconnect the plug at the required speed, but I'd like to build a setup like the one in the photo.

 

[hydtools]

I think you need to start with the Mil standard and specs to define and understand the test requirements before developing a methodology.

Ted

 

[sgs550]

Yes, I've become quite familiar with the Mil standard. That's where the 30'/sec requirement comes from. The 30'/sec disconnect speed is the main requirement of the test.

 

[IRstuff]

6 in would reduce the acceleration to 28 g's, a much more practical and plausible acceleration. With something that has 1.5in. diameter, that's like 137 psi.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm