Making a slow linear movement without motor? 2

[niklasw]

We are in need of a simple mechanical linear movement.

I wonder if anyone has tried making a good solution to a linear one-way motion without the use of electics? Perhaps you can give us some advice on where to start.

As it is now, we are using a DC-motor that transports a piece of 70x20mm plastic piece a distance of approximately 70mm. Since we want to minimize the use of electics in this unit, we are thinking of ways to do this without the motor.

Our idea is that the piece of plastic is pushed into a slot the full length (70mm) by hand and after release being pushed out automatically in a slow linear movement with hopefully constant speed. The full output motion should take 2-3 seconds.

I was thinking about some kind of viscous damped movement, but have never tried this. A spring could be compressed on insertion of the plastic piece.


Has anybody done this?

Or perhaps somebody has some good tips of simulation software to use

...Or links to more information...anything would be helpful.


Best regards,

Niklas

 

[alpharam]

Though I still hardly imagine the whole process after your description, I think that this link

http://www.bansbach.de/englisch/startseite_englisch_3.htm

may be useful. Gas spring would provide you with slow motion without any use of electricity, just as you have described.

 

[RichLeimbach]

I agree with alpharam that a gas spring would be a good choice. If you cannot find one that fits your cost / size constraints you can talk with Nye Lubricants. They've got a good range of damping greases that you can use with a custom design like you were talking about. You might be especially interested in getting one of their "sampler kits" thta has different greases that you can play around with to get the desired level of damping. Check out

http://www.nyelubricants.com/pdf/Nyogels.pdf

for more info.
Can't think of any good way to simulate it that would be any faster or more accurate than just making and testing different combinations. Hope this helps.

- Rich

 

[MintJulep]

Gravity.

alpha and Rich, How do you return the gas spring to the origianl position?

 

[TheTick]

If its a linear movement, you might consider something with a rotational damper and a rack. This is similar to what is used on some automotive consoles and CD drive doors.

Look up Delpro on the web.

"Great ideas need landing gear as well as wings."--C. D. Jackson

http://www.EsoxRepublic.com

 

[pennpoint]

How about a hydraulic cylinder connected to a air over oil reservoir. The hydaulic line to the cylinder would have a flow control valve (see below). As "niklasw" pushes his/her work piece forward 70mm (2.755&quot, retracting the piston and rod building pressure against the air. Upon release, the piston and rod move forward. Hydraulic fluid is metered back through the valve at the flow rate required to meet the lenear travel rate.

The flow control valve is bidirectional, however in one direction the flow is metered. In the other direction the flow is unrestricted. These type valves are available from many manufactures.

A bladder arrangement might also be used.

Regards
pennpoint

 

[alpharam]

Gravity is a good point.
As for gas springs that is why I said that I hardly imagine the whole process. As nicklasw wrote in his post, retracting movement shall follow inserting the plastic immediately, otherwise some control device must be attached to the system (mechanical, not hydraulic, most probably)

 

[m777182]

You could think about turning your manual mechanical work to some equivalent m*g*h work like -in the most simple case-using a pulley. It will counteract immediatelly and will eject your piece of plastic.A more luxurious model includes a spring. If you really need a constant speed movement at least at some distance,then you need a damper.Do not forget,that you will loose some millimeters to accelerate and a few more to stop the plastic piece. Secondly, if you are trying to measure some properties electronically during the linear move it will be more or less a heart attack EKG-like graph because of the friction effects.
m777182

 

[niklasw]

Thank you, everybody for repies to my post.

As I understood that I didn't fully describe the process, I will try to clarify a few details.

I wrote: "Our idea is that the piece of plastic is pushed into a slot the full length (70mm) by hand and after release being pushed out automatically in a slow linear movement with hopefully constant speed. The full output motion should take 2-3 seconds.
"

-The DC-motor was connected to a threaded rod that transported the tray, this is perhaps not the best choice since threaded rods could be quite expensive, but they give slow controlled motion.

-The plastic piece is put into a tray which is somehow connected to the linear drive that I need to design.

-This tray (includingg plastic piece) has a possible weight of 100g, probably less.

-To get an automated movement we somehow need to provide the system with the energy to move the tray. That is wy I considered that the tray i slid in by hand thus "loading a spring" or somehow building up enough energy in the system to push the tray out at command.

-The tray is possibly (probably) latched in the inner position and released on command (by releasing the latch).

-We want to perform different measurements at the piece of plastic, and that is why the movement should be slow enough to give enough "samples". It is acceptable to have controlled motion over a shorter length than the full distance (this could perhaps be fixed by extending the total distance travelled)

-The best would of course be a constant velocity movement, but I think we however need some kind of tracking of the position of the tray.

-The construction need to be cheap. Hopefully at least cheaper than the DC-motor with feedback solution.

-The tray need to be parallel to the ground


TheTick, This is one idea we have thought of, we didn't know how to do it or where to start, thanks for the link to DelPro. It would be nice to see how "automotive consoles and CD drive doors" work.

m777182, can dampers give a constant velocity movement, or will it just slow down the movement?

alpharam and RichLeimbach, Gas springs could be ok, simple solution and all. I have one on my table. This one is a bit to strong though, since it takes a lot of power to compress it. However, will they give the same velocity everytime released, I have a feeling they dont. Could perhaps be used together with position tracking.

pennpoint, your ideas sound interesting, but I do not understand it , would be happy for more details. Also a desciption of "a bladder arrangement"?


Gravity is a good idea, using the simple principles of potential energy, but it could be a problem since the tray is not allowed to be tilted.

 

[imagitec]

"Gravity is a good idea, using the simple principles of potential energy, but it could be a problem since the tray is not allowed to be tilted."

m777182's suggestion of a pulley is a way to use gravity without tilting the tray. How constant does your velocity need to be (this question is an oxymoron, but not even a servo system can produce constant velocity)? You could also use a constant force spring. It would probably be easier, however, to "tune" the system with a pulley and weights. Ideally, you'd want the applied load to be just shy of the static friction force. Then an impaarted force would set it in motion. In fact, if you're restraining until you want it to move, it could even exceed the stiction.

Depending on what kind of measurements you plan to take and how you plan to take them, you could go really retro and have mechanical feedback, e.g. a gear, trigger the measurement.

Rob Campbell

 

[alpharam]

Gas spring performs as normal spring providing smooth movement due to damping effect. It neither has a constant force during the stroke, nor constant velocity (moreover, these parameters can change with a change of ambient temperature).
On the other hand, it can give you time that you need, it can be contracted with low forces (as low as 10N) and it can be adjusted by supplier to suit your need more or less and, of course, it is cheaper than dc motor/threaded rod (AFAIK). It cannot substitute it though, by no means.
If you can arrange the system so that the tracking does the entire job try it, otherwise look at something else.
I would research clock mechanisms at your place, some of them have very nice solutions for damping movement (just an idea).

 

[griffengm]

niklasw,
It sounds as though you are trying to regulate the insertion speed to allow measurments to be taken while the part is traveling. Correct? Can half of the measurements be made during insertion and half during retraction?
Of what does the main mechanism consist? Is it like a press? Are you wanting to insert the part with only the force available from a hand? Is this an intermittent operation in that it occurs only occasionaly through the day or is it necessary on every part in a 2-3 sec time cycle? MUST the part come to a stop at a point in the cycle?
Could the part be loaded on a rotating table and passed under the scanner using a clockwork?

Griffy

 

[pennpoint]

Sorry! "niklasw" I didn’t know how descriptive I should be in my first post to you.
And I am not sure I’m going to do that much better this time, but I’m trying.
If I only had a drawing board! A picture is worth a 1000 words.

Using a hydraulic cylinder connected to an air over oil reservoir would entail the following. A double acting hydraulic cylinder connected only at the piston end to an air over oil reservoir or bladder. Air pressure is used as the pressurizing agent against the hydraulic fluid to “push” on the fluid to extend the piston of the hydraulic cylinder. The pressure required is based upon the load + fluid friction + resistance and all that stuff. The nice thing about it is you can adjust the pressure as required.
In the single line connected between the cylinder and reservoir place a flow control valve (Actually a speed control valve). These type valves are bi-directional; in one direction the flow is unrestricted. In the other direction the flow can be micro-metered.
Now connecting whatever you have to the piston rod end, you (manually) push against the cylinder compressing the hydraulic fluid against the air in the reservoir or bladder.
When you release the “thing” you want to move back in the opposite direction. The fluid passes through the metering valve at the metered flow rate you choose. Then the cycle starts all over again. Push to compress stroke, release and allow the hydraulic/air to advance the cylinder back forward. Cycle after cycle..
No electrical required.
The reason for the fluid to provide viscous flow characteristics. Using just air would not be adequate. Too hard to control at low speeds.
A bladder arrangement is nothing more than a thick balloon that isolates the fluid side from the airside. The fluid expands inside the bladder “balloon” and compresses the air on the outside of the bladder. When the motion needs to reverse the air pressure squeezes on the bladder to expel the fluid. Cycle after cycle.
Maybe someone who understands the concept of what I am describing can chime in and add some information to help enlighten. Maybe they can also tell me I am full of Camel poop.

With Regards
pennpoint

 

[notnats]

Old fashioned dial telephones used a return spring with a governor mechanism to allow the dial to return at a constant 10 pulses / sec. The spring was not too heavy and the speed control was good. Either a rack and pinion or drum and cable attached to a similar mechanism would probably do the job. However a lot of development went into them and a DC motor and screw mechanism would probably be simpler and cheaper unless you anticipate high production volumes to pay for the development.

 

[alpharam]

Sorry, pennpoint,
Isn't the scheme you describe just a big gas spring? And doesn't it bear the same disadvantages on it? You describe the valve that controls the flow, but will the viscosity and the flow be independent of ambient temperature changes during the day?
Just curious.

 

[pennpoint]

alpharam
Well! it surly will be influenced just the same as the device your suggesting, however my suggestion can be adjusted to compensate for some of it by increasing/decreasing pressure or fluid flow.

Regards
pennpoint

 

[tlee123]

If the friction in the movement is low, a pull damping air dashpot will work fine. You'll need a low spring rate spring to get (near) constant velocity. Gravity can also work well to provide the actuation force. Go to:

http://www.airpot.com

A rotary dashpot with a rack to drive it should also work well. Investigate how cycle rate affects the viscosity (and therefore increases the speed). Go to:

http://www.acecontrols.com

I also liked the idea with the Nye grease. Sounds good if you need to keep costs at a minimum and that you have the time to tinker.

Gas springs are not right for theis application. Forces change during the stroke and from cycle to cycle. Friction is high which is bad.

Standard disclaimer: I want to be upfront and tell you that I work as an engineer for Airpot an am not trying to sell here, I'm just telling something about a product that I know about.

 

[smckennz]

An air spring like is used on hatchbacks will do, but could be large and expensive one off.
A soft coil spring, oil filled cylinder (with compensating reservoir if required) and orifice/check valve will be the cheapest to custom make, but speed will change slightly with extension.
If constant speed is essential, do not forget the humble clockwork escapement mechanism, or the friction flyball governor (used on wind-up gramaphones). Both can be spring or counterweight powered.

Cheers

Steve McKenzie

 

[niklasw]

Thanks again for all good suggestions!

We are willing to evaluate different solutions. It will be interesting and will, independent of the result, be a good experience to try to make things work without electricity and motors.

We have already done some simple tests with a spring loaded gear connected to a home-made rotary damper filled with silicon oil. This is a simple solution and we will try to get in contact with some companies to try some "real" springs and rotary dampers.

We also have a 20N gas spring which is a bit to forceful for this application. However, It seems it will be difficult to have a smooth and repeatable velosity curve. And also the cost of these are relatively high even if you buy a number of them.

smckennz,

You wrote: "If constant speed is essential, do not forget the humble clockwork escapement mechanism..".
This and the other mechanism you mentioned would be really interesting to test out but we do not know how these are built or how these should be designed for our needs. Do you know where we could find an example setup for this or perhaps a description of how it should be done?


Thanks to all of you!

/Niklas