brainstorming

[boscar]

Hello, everyone!
It is just that I joined this forum and would like to submit the following problem to the various users.

for a school project I have to find the way to place cylinder1 and cylinder3 (see attached file) at the distance d=0.2mm +- 0.005 from cylinder 2.
cylinder 2 is fixed, so you can't move.
it must be possible to make small corrections of the distance by moving the cl1 and 3.it must be possible (when the cylinders are not moving) to move the cl 1 and 3 of 1 mm to the outside, this is cl1 to the left and cl 3 to the right.
cil 1 and 3 must exercise a force in the direction of cil 2 between 2kn-20kn.

together with my team we found different solutions, now I am pleased to see if someone has other ideas.....after all experience is the best teacher! !

thanks to anyone who dedicates time to this question! ! ! !

 

[sampom]

Hello, everyone!
It is just that I joined this forum and would like to submit the following problem to the various users.

for a school project I have to find the way to place cylinder1 and cylinder3 (see attached file) at the distance d=0.2mm +- 0.005 from cylinder 2.
cylinder 2 is fixed, so you can't move.
it must be possible to make small corrections of the distance by moving the cl1 and 3.it must be possible (when the cylinders are not moving) to move the cl 1 and 3 of 1 mm to the outside, this is cl1 to the left and cl 3 to the right.
cil 1 and 3 must exercise a force in the direction of cil 2 between 2kn-20kn.

together with my team we found different solutions, now I am pleased to see if someone has other ideas.....after all experience is the best teacher! !

thanks to anyone who dedicates time to this question! ! ! !

it is difficult to give solutions without knowing purpose and (especially) design limits; systems would be tens and all effective.
but what is (..uhmm, Switzerland.. a refiner?)? What's the point? (I see a file with three circles, a little bit..).
so to the eye you could say that it doesn't take who knows what I deploy technologies to do what you ask.
which meccanisni/automatisms can you/you have to use? dimensions and masses in play? What kind of controls?

your description is a little scary; It's not that we can guess solutions until you find the one that attacks you. The study should be much more targeted.
Maybe post what you have already thought, some more defined scheme.. and then it can be discarded.

I mean, let us understand.

greetings
Marco:smile:

 

[cacciatorino]

I would put the two reels 1 and 3 on two slides operated by pneumatic cylinders 1 or 2 mm. for the "all open" position you just need to call back the cylinder. for the "all closed" position you have to close the cylinders and send them to line on adjustable feedback, perhaps with threaded elements.

Bye.

 

[sampom]

I would put the two reels 1 and 3 on two slides operated by pneumatic cylinders 1 or 2 mm. for the "all open" position you just need to call back the cylinder. for the "all closed" position you have to close the cylinders and send them to line on adjustable feedback, perhaps with threaded elements.

Bye.

in fact it is a possibility (dimensions apart). only that for the forces that indicate (and for a "rigid" positioning) we would be better than the hydraulic cylinders (or a mechanical system).
Moreover if I understood what you are looking for would like to have a "finely" adjustable and quantifiable positioning probably automatic (now I do not know if of the same value symmetrically for the two cylinders), and perhaps with the guides and the laterally implementation on the shaft ends (then 2 groups, dx and sx). .
Then I repeat, it should explain better.

greetings
Marco:smile:

 

[cacciatorino]

Moreover if I understood what you are looking for would like to have a "finely" adjustable and quantifiable positioning probably automatic (now I do not know if of the same value symmetrically for the two cylinders), and perhaps with the guides and the laterally implementation on the shaft ends (then 2 groups, dx and sx). .
Then I repeat, it should explain better.

greetings
Marco:smile:

Clearly, you're right, it was just a conceptual idea that then our students can develop.
also because claiming to adjust to +/- 0.005 two reels under 20 kn spins (with all the problems of inflexions and elastic deformations that this entails) is a difficult question to solve.

 

[boscar]

hi marco, you're actually right!
here are more info:

purpose: press and/or mix oil with pigment (v attached)
roller size: 200mm diameter / working length 200 mm
Roller position: Horizontal
d distance: variable (manually or electronically) between 0 and 0.2 mm with positioning tolerance of +- 0.005 mm
the distance d must be separated for the two external cylinders.
transmission between rollers: via gears
to get the force you need to use pneumatic cylinders
pneumatic pressure provided by the network: max 5 bar
engine to rotate cylinders: electric, max 7.5 kw
cylinder engine transmission1: Strap on

these are the data provided, everything else is free!!
(the product is created only in 3d and we also have no cost limits--> clearly with head)

in the attached file you will also find some ideas we had.
I hope the data provided is enough! !

 

[boscar]

Hello, thank you 1000 for your help!

in the file that I attached in the previous reply you find some of our ideas.
(and more info rsul project)
We also thought about a solution like yours.
(currently I think it's one of the best)
However, we asked some questions about tyre pistons:

1) you can really get the required forces
2) Is positioning really possible with the required precision?
3) how to measure the distance to check the positioning--> do there exist such precise sensors? where to take the measure?

Now I don't demand answers, but just a few straight!
every advice is useful!

 

[Focus]

Hello, thank you 1000 for your help!

in the file that I attached in the previous reply you find some of our ideas.
(and more info rsul project)
We also thought about a solution like yours.
(currently I think it's one of the best)
However, we asked some questions about tyre pistons:

1) you can really get the required forces
2) Is positioning really possible with the required precision?
3) how to measure the distance to check the positioning--> do there exist such precise sensors? where to take the measure?

Now I don't demand answers, but just a few straight!
every advice is useful!

1) It is possible, perhaps with the use of air springs.
2) Yes, maybe using wedge systems.
3) look here the resolutions of the sensors http://www.mtssensor.de/industrial-sensors-br-r-g-e-l.218.0.html?&l=0If anything by looking at your pdf I don't understand the d solution, what does the cylinder need if the roller is bound in the approach movement by an endless screw?

Hi.

 

[cacciatorino]

1) you can really get the required forces

then, considering 5 bars and 20 kn it would take a cylinder large enough, but using one of these:http://www.tox-it.com/index.php?id=30&l=3You can get there comfortably. are pneumo-hydraulic systems: they are powered by compressed air but then inside them there is an oleodynamic system that multiplies the thrust. from the point of view of the control logic are such and as to a pneumatic cylinder. as cost we are on 1500-2000 euro.
for force change, you can use a pressure reducer party party, look for the series ms6, or a servovalvola to command via plc (series vx of the smc).

2) Is positioning really possible with the required precision?

depends on how you realize the mechanics. According to me, 0.005 mm require extreme construction accuracy, think about how much you can inflate the roller under a push of 20 kn, there's half a cent you'll lose it already on the start, then I think it will take a conditioned environment!

how to measure the distance to check positioning--> do there exist such precise sensors? where to take the measure?

you can use linear translators, such as this type:
http://www.gefran.it/en/products/product_853.aspxthe ideal point where to take the measure would be on the roller, but you take a measure with the exhaust system, then apply the 20 kn: the bearings drop, the roller bends, all the various games of the kinetic chain are reset, what end does your position? You have to try to put the mechanics of your sled as close as possible to the roller!

However you have to try to build a precise hyper mechanic, otherwise with all the games of the kinematic chain you immediately lose the required precision. think for example what it means to maintain a parallel accuracy of 0.005 on two 200mm long generators, of which one of the two mounted on a cart.
Are you sure you have to comply with these specifications?

Take a look at me. This is a job your teacher is doing for some client, and then he does it to his students? specs seem too precise for a didactic case....

 

[boscar]

Hello, thank you for your intervention!

in the d) solution it was thought to use the screw to place the cart with the roller.
once the cart is in place "fly the piston" so as to exercise the required force.
the piston should work in compression.

In any case, personally, I think this idea is to be rejected.
The team is gone and we will discuss it. :bekle:

 

[Focus]

Hello, thank you for your intervention!

in the d) solution it was thought to use the screw to place the cart with the roller.
once the cart is in place "fly the piston" so as to exercise the required force.
the piston should work in compression.

In any case, personally, I think this idea is to be rejected.
The team is gone and we will discuss it. :bekle:

doing so the force of the piston goes to download on the screw....

Hi.

 

[boscar]

Hi.
As for the tolerance to be respected on Monday (when the team finds itself) I want to talk about it with prof.
actually it is not obvious at all getting this value, perhaps the prof was wrong...... although I don't believe.
I am also very skeptical, the structure that supports the roller, although rigid that it is, will always have small deformations. especially if we apply the force of 20kn.

Finally I also wondered whether it is a job for an external customer or not. I've known the right prof for a few days and the only thing I know is that he's head of an ing studio strangely doing some internet search I didn't find anything.
I personally believe that this is a purely school work where you want to reach the maximum. I hope we are not used for its industrial works!! (I think he would warn us on departure)

 

[boscar]

You're right!
I'll talk to the creator, maybe he has something else in mind. even if I don't think so. I remain of opinion: idea to discard! !

 

[stefanobruno]

You're right!
I'll talk to the creator, maybe he has something else in mind. even if I don't think so. I remain of opinion: idea to discard! !

the push goes on the screw without end and, from there, through the rack passes to the support of the roller.

a similar application, with wider movements and tolerances not so pushed (I agree with the fact that 5 microns with 20 kn are an exaggeration ) I have done so:

sliding support pushed directly from the cylinder (in my hydraulic case) with contrast between the same support and a screw jack that, with its fine movement, imposes the stop to the desired quota.
control with absolute encoder on the movement of the screw jack.

Bye.

 

[zeigs]

I say mine, hoping not to change
for the question of rigidity you can think of supporting the two working rollers with 2 or 4 on the diameter decidedly higher that they will take care of the thrusts. It is a method normally used in rolling mills when the diameter of the rollers becomes too thin.
another method is to create the bombed work rollers so that under effort they present themselves, along the working generator, cylindrical. suffers from the problem of having a very constant and well defined force, for the rest it is a relatively academic calculation.
for positioning, considering the imponable stroke, I would opt for membrane cylinders: are practically two plates with an intercapedine that fills with pressure air. especially in the case of metal membrane allow to have very high forces with very short stroke (I don't know if they come to mm...). If they were hydraulic and there was no need for high dynamics once positioned, they could almost be considered fixed-point. I emphasize almost, given the required precision. is to predict something that allows reopening, unless you consider to throw away part of the production in repositioning (it is the material that enlarges the rollers passing inside, but they are out of range during). if the force is very variable you can with a compressor and put in exhaust with a regulator that keeps the pressure in the room, so to large lines.
for the sensor, I remember that for the thesis I used a magnetic position sensor with a range of 2 mm and a sensitivity around the micron. It's been a few years, I guess they've improved their performance in the meantime.
the dentate wheels for the drag are fine, for the race set the inranamento should not resent it

I don't understand a few things: in practice it is a crusher, even if at high precision. what are 3 reels for? Aren't two enough? If on one side the dust goes down, on the other it is pushed up and would not be crushed. then: admitting that you work in both ways, the fact of adjusting the two side rollers leads to think of a different piece, consequently different forces on the sides of the central wheel. only for this reason I think we go out tolerance for how good is the positioning of the side rollers

Hi.

 

[Focus]

a ø200 roller with a 200 board assuming an interasse of 350 bearings, loaded with 20kn, has a straw around 0.0001, I don't think there is need of support rollers or bombings for what it has to do.


Hi.

 

[boscar]

Bye!

the number of rollers was imposed by the prof and the processing process. it should be taken into account that the fluid (oil and pigment) slips through reels 2 and 3. then thanks to fluid dynamic effects it is dragged to reel 1 (it is still liquid). between roller 1 and roller 2 all solidify flat. at the end you get a colored "tip". (roll number according to file attached in previous posts)

to tell us that the rollers are slightly bummed, and not to think that under the influence of force they deform. In practice they have already been studied for application.

Could you tell me pf who is the manufacturer of the "diaphragm cylinders" you mentioned and the sensor brand? (always you remember )

 

[zeigs]

focus: as I said at the beginning, I spree not to change. (ii) I had only felt, I had not made any accounts above

boscar: some examples to understand, the only one I have ever used I had to buildhttp://www.directindustry.it/prod/bimba/cilindro-a-membrana-7046-352188.htmlhttp://www.meplas.com/piane.htmlI'm sorry for the sensor. I can't tell you. However google remembers better than me

 

[boscar]

Hi.
I looked at the cylinders.
In the annex you find a scheme, could you tell me if I have well interpreted your pf suggestion?
Do you need a valve connected to a plc or something?

 

[zeigs]

You can use that pattern. or put the regulator in the exhaust, you have a faster dynamic (the valve is never closed but it regulates continuously, while in the first case in addition to regular it must exchange in send and unload) but the calculation of the load losses in the ducts is complicated.
take into account that this type of actuators, not having creepy seals, implemented with air have the bad attitude to swing, having a very low damping.
Yes, we need a control system. but it would serve in any case you do not use a mechanical locking.
a hydraulic implementation in my opinion would obviate this: Once in place the fluid is incomprehensible, so once you define the walls of the room and the cinematic chain of a rigidity enough to guarantee tolerance, it no longer needs regular, just a control in position decidedly simpler to implement. and to command
or something like a bite: with pneumatic cylinders moved a sled on which the rollers are mounted, which is blocked by friction by another actuator, always pneumatic, with open/closed command of sufficient force. if you opt for this, remember to pressure the actuator relatively slow, or make sure that the movement is equipped with a viscous dampener, or give the sled a fairly high friction, to the limit with something studied purpose, because it risks that the sled moves to shots. then, let's mean, I've never worked on races so short, maybe I'm putting on problems that don't actually exist or are not implementable

Hi.