@maurodelazeri how are they doing this now with the computer controlled knitting machines that are on the market? I've been spending quite a bit of time trying to figure this out, looking at a lot of sites that sell them trying to see how it's done now. So far they all talk of a binary needle position with the V and flat bed knitters. And with the number of needles your trying to use it seems like this is an industrial project not one for home use.

Wiith the way it looks like in the places Ive looked so far there is no way that extending a needle to a differnt farthe position will do any thing different than a two position binary movement, it just would make the needle move farther befor it latches, which to me seems like the loop would end up looking like a snag in the finished fabric.

The closest I could find to doing what you seem to want are the ones that move the bed to make a bigger loop, not moving the needles. Is there a name for the type stitch your doing? I never before realized the number of different stitches in knitting.

 

The common problem of solenoids are that at wide gap position the force is very light, but at gap near zero it craps like a dying man under electrocute. So, one excellent solution is to produce the central magnet-core very conical, example 15 degrees slope. Then air gap will have much lesser impact on force.
Solution with stepper seems bit expensivish, if no steppers are under 10-15 USD then 15x100 pieces is very much money.

 

RE:""The problem with using three solenoids is space""
Any movable metall pair in Space conditions have surprizing problem - any even far different metalls are welding together if only take into contact, because of not existing monomolecular gas layer. Especially problematic it stays for bearings, shafts etc. Exists very expensive very special antiadhesive coatings against it

 

I don't know

@maurodelazeri how are they doing this now with the computer controlled knitting machines that are on the market? I've been spending quite a bit of time trying to figure this out, looking at a lot of sites that sell them trying to see how it's done now. So far they all talk of a binary needle position with the V and flat bed knitters. And with the number of needles your trying to use it seems like this is an industrial project not one for home use.

Wiith the way it looks like in the places Ive looked so far there is no way that extending a needle to a differnt farthe position will do any thing different than a two position binary movement, it just would make the needle move farther befor it latches, which to me seems like the loop would end up looking like a snag in the finished fabric.

The closest I could find to doing what you seem to want are the ones that move the bed to make a bigger loop, not moving the needles. Is there a name for the type stitch your doing? I never before realized the number of different stitches in knitting.

@shortbus this days this machines use something called carriage that moves back and fourth



That is basically a solenoid....

I'm not trying to do a specific stitch, there are many that can be done https://blog.nobleknits.com/blog/20...ting is based,variation of those two stitches.

 

Today I came across something that I think is worth sharing:

http://openactuator.org/

http://solenoid.or.kr/index_dosa_open_2d_eng.html

 

this days this machines use something called carriage that moves back and fourth

I understand that part, it is basically like in weaving where you send the weft through the warp. That solenoid you point to is what grabs the yarn and moves it across the needles to make the next row of loops, I get all of that. What I'm having a hard time understanding is the reason for a trinary(3 positions) of the needles instead of the normally used binary 2 positions.

It just seems like you want to make things more complicated for just the reason of doing it. All of the existing binary V-bed knitting machines can and do make all of the commonly used stitches, so you have invented a new stitch? One that needs a three position needle? I'm sorry if I come across as being argumentative, that is not my intent, I'm trying to understand why commonly used methods aren't good. And why the three positions have to be available on the fly instead of being set up like in weaving, something I do understand.

Having a three position needle set that could be changed out when a knitting set up is made could be a better/easier thing to do. Like when another color is added to a pattern.

 

I understand that part, it is basically like in weaving where you send the weft through the warp. That solenoid you point to is what grabs the yarn and moves it across the needles to make the next row of loops, I get all of that. What I'm having a hard time understanding is the reason for a trinary(3 positions) of the needles instead of the normally used binary 2 positions.

It just seems like you want to make things more complicated for just the reason of doing it. All of the existing binary V-bed knitting machines can and do make all of the commonly used stitches, so you have invented a new stitch? One that needs a three position needle? I'm sorry if I come across as being argumentative, that is not my intent, I'm trying to understand why commonly used methods aren't good. And why the three positions have to be available on the fly instead of being set up like in weaving, something I do understand.

Having a three position needle set that could be changed out when a knitting set up is made could be a better/easier thing to do. Like when another color is added to a pattern.

@shortbus there's a reason, I wouldn't do that just to make it more complicated

 

there's a reason

You could tell me but then you'd have to kill me things?

 

You could tell me but then you'd have to kill me things?

not necessarily hahaha, but I would have to explain to you all the design and other parts that also are affected by this decision and this is not the focus of this topic

 

Here's an idea that might be worth considering: A 3-cam system for each needle with the cams in parallel, but each cam with an internal clutch that could be activated either electrically (i.e. a solenoid) or pneumatically.

 

Here's an idea that might be worth considering: A 3-cam system for each needle with the cams in parallel, but each cam with an internal clutch that could be activated either electrically (i.e. a solenoid) or pneumatically.

pneumatically I wouldn't do it, with a solenoid is something to think about it but even, I would have to go beyond that what I would like in terms of space

I liked the idea that Danko brought with the paper below

Design of a Solenoid Actuator with a Magnetic Plunger for Miniaturized Segment Robots

 

pneumatically I wouldn't do it, with a solenoid is something to think about it but even, I would have to go beyond that what I would like in terms of space

I liked the idea that Danko brought with the paper below

Design of a Solenoid Actuator with a Magnetic Plunger for Miniaturized Segment Robots

I meant for the pneumatics to be used to actuate the clutch of one cam in a set of three ... that would qualify as micro-pneumatics.

 

One comment: perhaps I should've used the word "engage" instead of "actuate" in my previous post, to make my point clearer...

 

One comment: perhaps I should've used the word "engage" instead of "actuate" in my previous post, to make my point clearer...

@cmartinez I got you, thanks... what are your thoughts on a voice coil motor ? overkill ?

 

@cmartinez I got you, thanks... what are your thoughts on a voice coil motor ? overkill ?

My thoughts on that can be reduced in two words: Bulky and Pricey.

And again, controlling a solenoid's speed profile is no easy task (mainly because of inertia and current buildup in their coils), and the problem would be exponentially compounded if one tries to sync the actuation of a set of solenoids.

Here's another candidate idea: consider using a galvanometer for what you want. They can be extremely fast, although their motion is oscillating instead of linear, and it could prove difficult to move an object more massive than a needle and might be susceptible to positional errors due to friction.

 

@maurodelazeri, Had some time today so went to Youtube and looked at some videos of these type knitting machines, to get a better idea of the working of them.

So have a new questions;
1. Does this idea do away with the cam box on the carriage?
2. Or are you just wanting to have a way to drop out individual needles, for a pattern?
3. Are you going to make needles of your own design or use an existing type of needle?
4. If an existing needle what brand/type?

Really am trying to come up with a way to help you, I know more now about knitting than I ever did.

 

@maurodelazeri, Had some time today so went to Youtube and looked at some videos of these type knitting machines, to get a better idea of the working of them.

So have a new questions;
1. Does this idea do away with the cam box on the carriage?
2. Or are you just wanting to have a way to drop out individual needles, for a pattern?
3. Are you going to make needles of your own design or use an existing type of needle?
4. If an existing needle what brand/type?

Really am trying to come up with a way to help you, I know more now about knitting than I ever did.

I don't know how this questions are related to the topic

 

@maurodelazeri, I think shortbus is really trying to help you out. He has obviously spent some time doing research on your application. You have explained very little of your project requirements to us (probably because you want to keep them secret for potential future commercial application), but this makes it quite difficult for us to offer help.

If you are able and willing to describe the reason why you must control the needles in this way, and the method in which they must be controlled, then you will likely get more directed feedback on your questions. More people may join in the discussion and offer ideas that you haven't thought of yet. The more details you provide the more help we can give.

If you need to keep this secret, we get it, but don't expect to get much help on your project.

 

@maurodelazeri, I think shortbus is really trying to help you out. He has obviously spent some time doing research on your application. You have explained very little of your project requirements to us (probably because you want to keep them secret for potential future commercial application), but this makes it quite difficult for us to offer help.

If you are able and willing to describe the reason why you must control the needles in this way, and the method in which they must be controlled, then you will likely get more directed feedback on your questions. More people may join in the discussion and offer ideas that you haven't thought of yet. The more details you provide the more help we can give.

If you need to keep this secret, we get it, but don't expect to get much help on your project.

@mcardoso what you said is just not true, there are plenty information here, I explained with details, pictures, I even posted a 3D model of a needle moving... I really appreciate all the help im getting from this topic but the @shortbus questions have no relation with the problem that is Controlled linear motion with 3 position stages, 10% tolerance is acceptable, speed is important, ~20ms would be great, force of ~5N

 

@mcardoso what you said is just not true, there are plenty information here, I explained with details, pictures, I even posted a 3D model of a needle moving... I really appreciate all the help im getting from this topic but the @shortbus questions have no relation with the problem that is Controlled linear motion with 3 position stages, 10% tolerance is acceptable, speed is important, ~20ms would be great, force of ~5N

One of the most critical things is going to be the size constraints and whether the needles move in unison or not. This greatly rules out lots of different kinds of actuators. Does the needle need the full force over the entire travel length? Only at the end? Does the needle always fully retract? Do the needles move in unison? Can they be grouped? Does each one move completely independently and programmably? Are these configured at the start of the run of product or are they changed dynamically? How quickly must the mode of operation change? Is position feedback required? Do you need to control the velocity and acceleration profile? Does it need damping? Is the 20ms one way, or in and out?