Problem with a Loom!

[wizard]

Hi guys,

Today I were at a sack producing factory.
At the woven line there was a circular loom (like the typical below picture but it had 4 shuttles).
I am not sure how many of you have seen a such device, but it mixes
the warp and ther woof and produces an integrated sack.
The device stops working automatically when a woof (those innumerable strings which seem to be the radius of the circle in the picture!) is broken or
cut off. But unfortunately there is not a such option for warps which come out of the 4 SHUTTLES (I have shown a shuttle by a blue line), when the shuttle's string cuts off the device keeps working and produces an odd sack.
Unluckily the shuttle rotates both along its axis and along the circle (the raising sack) (I have shown it by those 2 read lines).
The electric technician of the factory asked me if I am able to solve the problem?
I used a mirror and an INFRA RED system to solve the problem. The mirror were mounted on the shuttle's string so that it were hidden till the string were cut off, after that the mirror came out and reflect the infra red light which were emitted of the infrared system and by this trick the loom were stoped.

Now they want me to think of another way to do a similar task but does not CONTACT to the shuttle's string.
As I mentioned before the shuttle has 2 kinds of rotation so I am not able to handle it by any external power. When shuttle's string is broken the shuttle stops rotating along its axis (not along the circle, as you know).
Does every body know any solution please?

Thanks a bunch

 

[Ubergeek63]

got a video of it? preferable slow motion. I have a few ideas but can not tell from your description if they will work.

The most difficult would be if all six motion in a continuous 360 around the bag.

Dan

 

[wizard]

They are 4 really.

Yea I have gotten a small video, but not sure if it is helpful or scarer

Yes those shuttles rotate around the sack by 360 degree continuously.

I thought of induction sensors but unfortunately it seems they do not work too.


The technician wants a strong trick for the shuttle.

 

[Ubergeek63]

Grrrrr....I thought I responded. Oh well.... it makes sense that the thing would be fast... how much space is there around the shuttles?

If you can paint a stripe on the shuttle and mount an opto in range of each one you can use a couple AAs to run a system that starts emitting magnetic pulses when one stops spinning.

The technique is used on fitness equipment for remote heart beat sensing. it is only good for a few feet, but that is ok since you only need to get out to a stationary receiver.

Dan

 

[eblc1388]

Another suggestion is to drill a hole in the shuttle and fix a tiny tiny Neodymium cylindrical magnet. Make sure the magnetic pole is pointing in a radial direction to the axis of the shuttle.

The small physical size of the magnet would not unbalance the shuttle in operation.

A biggish coil is placed over, under or near the shuttle traveling plain. Because of the strong magnetic field of these small devil magnets, the positioning of the coil can actually be quite far away.

If the shuttle is rotating while traveling in a circle, then the coil will pick up several magnetic pulses by the passage of each shuttle. If the shuttle is not rotating, only a single pulse will result.

With some clever IC logic or via a microcontroller, such condition can be detected and acted upon.

 

[Ubergeek63]

If the shuttle is rotating while traveling in a circle, then the coil will pick up several magnetic pulses by the passage of each shuttle. If the shuttle is not rotating, only a single pulse will result.

With some clever IC logic or via a microcontroller, such condition can be detected and acted upon.

He is looking for the cessation of cylindrical motion so he can stop the rotating.

He has two axis to deal with. The shuttle is spinning in the Y direction as the thread or what ever is pulled off and the six shuttles are racing around the piece being made. He want the machine to stop when any of the shuttles run out of thread.

I would love to hear your explanation for how to distinguish between the race and the spin. I know of only one way, but with out more information I can not tell if it will work never mind being reliable with out more information.

Dan

 

[bassnut]

mm, Ive had a bit to do with shuttle controls on looms. They have extremely fast and sophisticated control systems. Are you sure the manufacturer hasnt got a solution?.

 

[Ubergeek63]

mm, Ive had a bit to do with shuttle controls on looms. They have extremely fast and sophisticated control systems. Are you sure the manufacturer hasnt got a solution?.

his does not, did you see the video clip? It is a purely mechanical tube loom.

I believe you are remembering fancy automated pattern looms, all he has is a tube weaver. While mechanically intricate, electronically minimalistic.

Dan

 

[Pommie]

Dan,

You should reread eblc's post. He posted what I consider a valid solution to the problem but you appear to have misinterpreted it. To reiterate his post, if the shuttle is rotating then multiple pulses will be observed, if stationary then only a single pulse.

Mike.

 

[wizard]

Thanks for all responses,

As I mentioned before, my first trick was using a REFLECTOR (a small mirror) under the shuttle arm where the shuttle's string goes for the vertical sack, So when the shuttle string was broken or finished the reflector came out of the arm and in this way the fault was detected by an IR system mounted above at 10cm of the shuttle. The trick was very nice. So we did not have any electronic device on the shuttle, the fault was detected after just One 360 degree of rotation or maybe less, and the mirror was strong enough to do not be damaged when the shuttle arm would take for repair.

Yesterday when I have been told by phone that they need a CONTACTLESS trick, the first think I thought of was putting a Neodymium magnet on the shuttle and controls its behavior by inductive sensors or just a coil, so it seemed a simple task till I went and see the device gain. I did not care of its circular rotation and thought it just has a single axis rotate along its axis. using of permanent magnets on the shuttle is the easiest way to go but I think it is not so easy for a device which has a 2D rotations.

Ubergeek63,
The factory does not like to be involved with replacing or charging the batteries.

Bassnut,
The factory is a small one and produces not many sacks to buy just modern machineries. Tthe devices are old enough to not have a solution for such problems. They are made in China.

eblc,
Do you want me to use a microcontroller and configure it to WAIT for a period of time for numbers of pulses it detects? I am thinking of the possibility of your trick yet. But do you know we have MULTIPLE pulses without regarding to the broken shuttle? There are 4 magnets on 4 shuttles which rotate around the sack too, and would be detected by the sensor (i.e. coil) as multiple pulses.

 

[Pommie]

Thanks for all responses,
eblc,
Do you want me to use a microcontroller and configure it to WAIT for a period of time for numbers of pulses it detects? I am thinking of the possibility of your trick yet. But do you know we have MULTIPLE pulses without regarding to the broken shuttle? There are 4 magnets on 4 shuttles which rotate around the sack too, and would be detected by the sensor (i.e. coil) as multiple pulses.

A microcontroller would be required as this is a complex problem. The timing of the pulses will define if it is a stopped shuttle going past the detector.

What mechanism stops the machine? A PC? If so then the PC can count pulses and no uC is required.

Mike.

 

[xanadunow]

Hi wizard,

This is a production line I don't have much understanding of, so let me ask you a few questions.

a. the direction of production line.. is a string being woven onto the rotating shuttle or from the shuttle?

b. is there only one string on one shuttle or are there more of them (I understand that whether there is only one or there are many, any string will travel accross the shuttle within the shuttle's width)?

It makes me think that if detection of the presence (or not) of a singular string is factually to hard because of it's constant position change and relatively small diameter, maybe we should think of monitoring this in an alternative way, like:

c. detection of the string's tension
d. detection of the load change of the shuttle's drive with and/or without the the tension of the string
e. as the position of the string changes so much at the shuttle end would it not be easier to monitor it optically at the other end i.e. in the centre of the big weel?

What do you think?

Regards,
xanadunow

 

[Pommie]

Eblc's solution works, use it. Or explain why it doesn't work so I can rest easy.

Mike.

 

[Ubergeek63]

Dan,

You should reread eblc's post. He posted what I consider a valid solution to the problem but you appear to have misinterpreted it. To reiterate his post, if the shuttle is rotating then multiple pulses will be observed, if stationary then only a single pulse.

Mike.

But the structure he posted will always have multiple pulses on the fixed sensor since it is the axis motion he wants to sense and not the motion of the shuttles around the piece being made.

There is no good non contact way to get the power to the shuttle assembly and get the signal back saying that one of the shuttles stopped spinning

Dan

 

[Pommie]

If the shuttle is spinning then you will get multiple pulses per quarter rev. If stationary then you will get one.

Mike.

 

[Ubergeek63]

If the shuttle is spinning then you will get multiple pulses per quarter rev. If stationary then you will get one.

Mike.

Look at the video... string is coming off the shuttle causing the spinning that he wants to sense. in addition the shuttles are all spinning around the unit being manufactured, so even if ALL the shuttles stop spinning you will continue to get pulses from the motion of the structure.

You can not rely on multiple pulses as the shuttles pass the sensor with out more information, namely the ratio of the weave radius to the shuttle radius.

Dan

 

[Pommie]

The (my) assumption is that the shuttles spin multiple times per revolution of the main unit. If per revolution you normally get 50 pulses then all is well. If you get 40 then one of the bobbins has stopped. If you get a little more clever then you may see 12 pulses per quarter and when a bobbin stops you will see no pulses per quarter.

We have the ability to count pulses at megahertz, counting pulses in a quarter of a revolution of any mechanical device is trivial.

Mike.

 

[Rolf]

Hi guys,

[snip]
I used a mirror and an INFRA RED system to solve the problem. The mirror were mounted on the shuttle's string so that it were hidden till the string were cut off, after that the mirror came out and reflect the infra red light which were emitted of the infrared system and by this trick the loom were stoped. [snip


Thanks a bunch

Don't understand why the string has to touch the mirror, as you state in a later post, please explain.
Sounds like you have a working solution that could be modified slightly for satisfactory use.

Rolf

 

[wizard]

Pommie,
We must take care of the starting point when the loom starts to work, besides; the loom does not rotate with a constant speed at all times, I am not sure if the trick you and eblc mentioned is practical yet. Besides you should consider that every shuttle is exposed at the detector at a quite small period of time, so I am not sure if the counter is able to count the whole pulses!!?

There is no PC just a circuit is doing so.



Xanadunow,
Just one string is coming out of any shuttle, just like a spool.

When I was at the factory I was thinking of several solutions.
I had several suggestions in accordance with the tension which whole were refused. A Suggestion was putting a small solar cell connected to a TX 'IR' diode on any arm and then send an alternating signal into solar cell by the constant IR system, this trick needs a key to be connected on the shuttle string (and reduces the tension of the current reflector system) this suggestion was refused too.
So the single solution seems to be working on the shuttle rotation.

I am not sure what do you mean by controlling the position of the string and monitoring it? Please be noted that there are numerous strings there (if we suppose the 4 shuttle's strings as warps you must consider that there are a lot of woofs...)

Rolf,
The mirror is mounted on a surface with a HOLE across the surface, when any string of any shuttle is cut off or finished then the surface has no support (leakage of the taut string) and will be exposed due to centrifugal force. Make sense?

 

[nickelflippr]

Rolf,
The mirror is mounted on a surface with a HOLE across the surface, when any string of any shuttle is cut off or finished then the surface has no support (leakage of the taut string) and will be exposed due to centrifugal force. Make sense?

Do you have a picture of your shuttle string to IR/mirror setup?, also string variables in terms of space/distance to IR sensor. Not knowing the shuttle string environment its hard to say what alternatives are available. Photo interrupter, reflective object sensors, etc.?