I'm working on a little project that requires me to wind up a long strip of VERY thin and delicate material. Basically plastic (hence foiling my original plan of using a continuity tester-like circuit) thinner than aluminum foil. I want to come up with a digital or gentle mechanical solution to finding the edges of the foil as it winds up. The sensing needs to be done on the foil plane BEFORE its wound into the roll.
(so that I can ensure that the strip stays perfectly centered)
I need something cheap (maybe $100 max?) and quite small. Kind of like an edge finder in machining, but preferably some sort of sensor.
Some ideas I've brainstormed are:
Using a strain gauge/piezoelectric sensor on either edge of the foil to detect if one side is starting to receive more force than the other
Concern: Foil is so thin that I'm not sure the sensors will detect any meaningful output from the contact
Using a infrared diode + receiver on either edge of foil. Measure if foil is starting to cover more light on one receiver
Concern: Not certain how well this can be calibrated/ how easy this will be...
Pray to the gods.
Concern: I'm not religious...
Any guidance on this project would be GREATLY appreciated!
I'd use a laser beam with a LDR to detect if the foil is moving. If you install such a solution on both sides you get an immediate digital signal if the centering is not within the limits.
Using a strain gauge/piezoelectric sensor on either edge of the foil to detect if one side is starting to receive more force than the other
Concern: Foil is so thin that I'm not sure the sensors will detect any meaningful output from the contact
Using a infrared diode + receiver on either edge of foil. Measure if foil is starting to cover more light on one receiver
Concern: Not certain how well this can be calibrated/ how easy this will be...
Pray to the gods.
Concern: I'm not religious...
Of the three ideas you present, IMHO the second is the best to pursue. You are correct that for such a thin material any weight based system (at least any that is under multiple thousands of dollars) is going to be at best unreliable. Although pylon's suggestion is even better, especially if the extra information is useful for you.
With either a laser or IR based system the detection of the edge will depend a lot on how much light is transmitted through material at the given wavelength being sensed. If it's completely opaque to the give wavelength, it's as easy as distiguishing between HIGH and LOW on a digital I/O pin. If it's translucent, well then you might have to use an analog input and experimentally determine the proper threshold. Also on the subject of wavelengths, it could be possible to do this with visible light LEDs as well. You should also know that most LEDs are capable of acting as passable detectors for their output wavelengths when reversed biased.
Hmm, I have a slight problem in that the material (it is opaque) is resting on a tray. I don't think I have any space to install a LDR underneath the foil (assuming laser is above the foil). Are there ways to have both laser+receiver on the same side?
pegasixi:
I want to come up with a digital or gentle mechanical solution to finding the edges of the foil as it winds up. The sensing needs to be done on the foil plane BEFORE its wound into the roll.
(so that I can ensure that the strip stays perfectly centered)
I'm guessing that the underlying requirement is to end up with the material centered on the roll after it is wound up, and I wonder whether it may be possible to come up with a scheme to make the foil self-center rather than measure whether it's off-center and then try to do something (what?) about it.
Unfortunately, without having any idea of where the foil is coming from or to or what the physical properties are, I can't begin to guess how you'd achieve that.
Hmmm, the above two posts bring up an interesting point. Ideally, I'd devise a system like the following:
However, my problem is slightly complicated by the fact that I'm sandwiching this very delicate material between two wider sheets. I'm trying to center this foil in relation to the wider sheets which are all being wound together into a roll.
To describe the setup: My foil is laid upon a tray and (currently) hand-guided into a mandrel that is winding up the two wider sheets by means of a powerful servo. The foil is approximately 1m long.
My current plan of attack is to just try and get my foil PERFECTLY lined up initially and not do any corrections as it winds up. I figure that 1m doesn't really give me much leeway for corrections anyways. However, any and all suggestions are welcome! ^^
To clarify, that video shows an industrial system that automatically aligns itself while winding. I'm working on a homebrew solution for cheap.
My foil is approximately 54mm wide, and the sandwich sheets are 56mm wide. I need to keep the deflection across the 1m wind less than 1mm either way (The 1m length of foil cannot "drift" more than 1mm either direction during the wind... Meaning I have to align it pretty well from the beginning)
Is it feasible to suspend all three parts vertically and let gravity align them all? I guess they'll all be super light and I'd imagine hanging something on the bottom end to hold them all straight down, although what form that would take would depend how delicate the material was.
The current setup does something quite similar: It uses a tensioner to pull all three sheets (kinda like what the weight would be doing in your scenario). Unfortunately, using gravity is not an option due to the machine's stupid layout. So... instead of letting gravity do the work of aligning it, I have to find a way of manually lining everything up... Ugh, I really hate this machine.
Out of curiosity, can you/anyone recommend a good analog LDR with good resolution in a small package? I'm along the lines of using a cheapo laser, pass it through a filter, pass that through a pinhole (to decrease the size of the dot and increase accuracy), and have the LDR detect the tiny remaining LED dot. Am I on the right track here?
pegasixi:
Out of curiosity, can you/anyone recommend a good analog LDR with good resolution in a small package? I'm along the lines of using a cheapo laser, pass it through a filter, pass that through a pinhole (to decrease the size of the dot and increase accuracy), and have the LDR detect the tiny remaining LED dot.
If you are using a bright light source of with a known output primarily at or around a specific wavelength (like a laser or LED), you could use discreet photodiodes or phototransistors. They can come in very small packages and can be biased to specific threshold levels. Here's an example of a photodiode is made to detect red and near IR and a small and thin package. To give you an idea of cost, Digikey's price is $3.64 for volumes under 25.
pegasixi:
The current setup does something quite similar: It uses a tensioner to pull all three sheets (kinda like what the weight would be doing in your scenario). Unfortunately, using gravity is not an option due to the machine's stupid layout. So... instead of letting gravity do the work of aligning it, I have to find a way of manually lining everything up
Is the tensioner based approach not working, then? I'd have thought that was your best bet. Trying to detect tiny alignment errors and then compensate for them (how?) feels like a much harder problem to solve.
Are there ways to have both laser+receiver on the same side?
Put a mirrored surface on the tray and bounce the light of the mirror?
A daft thought;
1m is not that long. Could the foil not be tensioned, placed on the lower sheet (allowing alignment to be adjusted), the upper sheet applied, and then the sandwich rolled? Maybe an electrostatic charge could hold the foil to the lower sheet once it is positioned.
I suppose a lot depends on the physical properties of the foil, does it tend to stick to things and deform easily.
pegasixi:
Out of curiosity, can you/anyone recommend a good analog LDR with good resolution in a small package? I'm along the lines of using a cheapo laser, pass it through a filter, pass that through a pinhole (to decrease the size of the dot and increase accuracy), and have the LDR detect the tiny remaining LED dot.
If you are using a bright light source of with a known output primarily at or around a specific wavelength (like a laser or LED), you could use discreet photodiodes or phototransistors. They can come in very small packages and can be biased to specific threshold levels. Here's an example of a photodiode is made to detect red and near IR and a small and thin package. To give you an idea of cost, Digikey's price is $3.64 for volumes under 25.
Thanks! That looks very very promising! All the ones I'd seen so far came in huge packaging.
pegasixi:
The current setup does something quite similar: It uses a tensioner to pull all three sheets (kinda like what the weight would be doing in your scenario). Unfortunately, using gravity is not an option due to the machine's stupid layout. So... instead of letting gravity do the work of aligning it, I have to find a way of manually lining everything up
Is the tensioner based approach not working, then? I'd have thought that was your best bet. Trying to detect tiny alignment errors and then compensate for them (how?) feels like a much harder problem to solve.
Well, I think it WOULD work if I could get it aligned perfectly from the get-go. That's what I'm trying to detect the alignment; I want to use them to align the foil perfectly before I start the winding process. I don't think I can develop a compensating system.
Are there ways to have both laser+receiver on the same side?
Put a mirrored surface on the tray and bounce the light of the mirror?
A daft thought;
1m is not that long. Could the foil not be tensioned, placed on the lower sheet (allowing alignment to be adjusted), the upper sheet applied, and then the sandwich rolled? Maybe an electrostatic charge could hold the foil to the lower sheet once it is positioned.
I suppose a lot depends on the physical properties of the foil, does it tend to stick to things and deform easily.
Hmm... A reflective surface... I guess I could try and calibrate my readings based on the reflections. That may be a good option.
Unfortunately, the machine's layout prohibits lining up the three sheets from the get-go (they're spaced apart for some weird reason).
I guess I could try and calibrate my readings based on the reflections
If you were using a laser there would not be much need for calibration, if the foil obscured the mirror the beam would be cut off.
Alternatively the mirror could be foil width and on the perfect track. If the beam gets 'made' the mirror has been exposed (I think I prefer it being cut).
The video showed an essentially continuous process. What I was trying to suggest is that perhaps a batch approach might be better if you are only rolling 1m lengths.
It sounds as if you have an existing machine though. What is the actual application, and what is the current machine design if that is not asking too much?
