Want to measure rolled goods in a factory
avsa242
Posts: 452
Greetings,
I'm a department head for a defense contractor and most of the products in my department use various rolled goods that are sewn together. As the lead, I am also responsible for inventory of the floor stock in my department. While some items, such as various screws, nuts & bolts are relatively easy to keep a handle on, I've found rolled goods a daunting challenge to maintain what is in my floor stock physically versus what our MRP system says is in my location. With cycle counts becoming a regular part of my responsibilities, it is even more important that my numbers line up.
One of the ways I would like to establish control, or at least to verify usage against the BOM, is to implement a simple counter that I could in some way physically attach to the spools of material. As an example, one of the materials is a nylon binding tape that is fed into the sewing machine simultaneously with the thread. The rolls vary in width from .75" to 1.5" depending on the product, and about 8-10" in diameter when new. Initially I was thinking of just finding a bicycle odometer and retrofitting it somehow, but I really need to be able to measure down to the inch and couldn't find anything appropriate. Now I'm interested in experimenting with an encoder and using one of the Propeller boards I have to make up a simple counter, but I wanted to ask if anyone else has ever had to track usage of materials like this in a manufacturing setting, or if they've worked with or seen devices like what I'm thinking of putting together? What type of encoder would you recommend for something like this?
Any ideas would be greatly appreciated!
Cheers,
Jesse
I'm a department head for a defense contractor and most of the products in my department use various rolled goods that are sewn together. As the lead, I am also responsible for inventory of the floor stock in my department. While some items, such as various screws, nuts & bolts are relatively easy to keep a handle on, I've found rolled goods a daunting challenge to maintain what is in my floor stock physically versus what our MRP system says is in my location. With cycle counts becoming a regular part of my responsibilities, it is even more important that my numbers line up.
One of the ways I would like to establish control, or at least to verify usage against the BOM, is to implement a simple counter that I could in some way physically attach to the spools of material. As an example, one of the materials is a nylon binding tape that is fed into the sewing machine simultaneously with the thread. The rolls vary in width from .75" to 1.5" depending on the product, and about 8-10" in diameter when new. Initially I was thinking of just finding a bicycle odometer and retrofitting it somehow, but I really need to be able to measure down to the inch and couldn't find anything appropriate. Now I'm interested in experimenting with an encoder and using one of the Propeller boards I have to make up a simple counter, but I wanted to ask if anyone else has ever had to track usage of materials like this in a manufacturing setting, or if they've worked with or seen devices like what I'm thinking of putting together? What type of encoder would you recommend for something like this?
Any ideas would be greatly appreciated!
Cheers,
Jesse
Comments
Just to help clarify in case the original poster is not familiar with sanding drums, etc.... the sanding drum is the rubber cylinder that the sanding sleeve slides over. The sanding sleeve is the gritty thing that does the sanding. The sanding drum is smooth but has a bit of a rubbery "grip" and probably won't hurt whatever you're measuring.
Welcome, Jesse! HTH, especially since you're interested in using a Propeller board. Parallax's President Ken is always eager to assist with special pricing and I'm sure he would love the opportunity to reciprocate for the many bargains the USA has received from defense contractors over the years. For instance, from http://www.globalspex.com/bartlow/afaquisition.htm :
... the Defense Department did buy a diode for $110...
we bought a claw hammer for $435 ...
we bought the $9,600 allen wrench...
Just kidding, honest. I'm a former Pratt & Whitney GPD man myself. Now your encoder idea has merit, and you have received some good ideas already. I've bought fabric for various projects, and I know most cloth shops have specialized mechanical "odometers" to measure cloth distance. Could something like that work for your application? Google "fabric measuring machine".
http://www.polymaster.co.uk
http://measuregraph.com/Product_Description_HTML_pages/Measuring%20Machines.htm
Here's a neat 1922 vintage unit on Ebay now: http://www.ebay.com/itm/Antique-sewing-Machine-fabric-measure-Store-Simplex-computing-1922-/370806948723?pt=LH_DefaultDomain_0&hash=item5655d10373
And if you need a Propeller board, I have a nice pair I can let go for just $11,500 each.
S.R. Hadden: "First rule in government spending: why buy one when you can have two at twice the price?"
I'm sure that Rich (W9GFO) will chime in here.
[video=youtube_share;_xsFci_7TTA]
-Phil
If you know the distance between WIF Pulses, the length of material (Red) removed from the roll may be determined.
If you count how many WIF Pulses occur for every Roll Pulse, the diameter of the Material can be determined.
After a number of Roll Pulses, the thickness of the material can be calculated from changes in circumference.
Since you now know the diameter of the Material Roll and the material thickness, calculating the area of the face of the material roll minus the area of the core, dividing the area by the material thickness will give you linear remaining on the roll.
And lastly if you know remaining material, rate of consumption, you can calculate time remaining before consumption of roll.
Encoder Products makes the TR1 that I use quite often. Wheel, encoder, and torsion arm in one single assembly.
Something like sheet steel rolled on a coil has nearly no stretch at all when measured.
But some fabrics, such a mosquito netting might actually stretch quite a bit while being pulled through a measuring device. It would require a more sophisticated drive system to be assured that you actually are not distorting your measurements as you go.
I strongly suspect that the mosquito netting would be better off weighed for inventory control purposes. You might just drive yourself to distraction trying to get the right system established.
erco, My favorite quote from that movie
The two old mechanical meters could possibly work for our .75-1.5" wide materials, but for the big rolls (50-60" wide, several hundred pounds each) I'm not sure how I'd go about attaching them. That Polymaster site caught my eye, though, especially the first picture on the left with the measuring device.
kwinn and ElectricAye,
Yes, something about that size is what I was thinking for the binding tapes, which are fairly small rolls (anywhere from .75" wide to 1.5" wide depending on the part, and about 8-10" in diameter when new). I was thinking the easiest physical connection would be at the axle of the roll, although I think this would require some additional calculation; I think part of what pogertt is talking about addresses this.
JordanCClark,
I think I would be able to do this with the binding tapes, as they're thin, narrow material. I'd just have to be sure they straighten back out when feeding into the sewing machine. Slippage was one of my concerns when I was thinking of the small rubberized wheel for the encoder.
The laser-based encoder looks awesome but I doubt my bosses would go for something like that, especially for all four of our cutters.
Loopy Byteloose,
The small rolls are nylon binding tape (sewing two or more materials together, and you are left with a nasty, "unfinished" look along the edge where they're sewn together, so this binding tape is sewn over it). We also have 50-60" wide, several hundred pound rolls of material that are cut on a mechanical or laser cutter. With the exception of one, none of them are particularly stretchy. I hadn't thought of it before, but that one material may be more difficult for just the reason you mentioned. Thanks for bringing it up.
Thanks all again for taking the time to reply!
I will post updates, especially if I end up going the Propeller-based route (I'm really itching to
Cheers,
Jesse
That is one thing I learned about computer systems - sometimes a good old fashioned paper index card system is the best solution for a "management information system".
With that said, if you are a large contractor, perhaps you could get your rolled goods suppliers to mark the rolled goods with measurements of some type? Like along the edge or back side, maybe mark feet left in the roll or whatever. Maybe this could be marked with something which is easily removed with one washing?
And if this is a problem with an entire industry, you can get together with other manufacturers and come up with some sort of standard marking system.
I was picturing the roller/encoder either riding on the top of the material roll, between the encoder roller and a pinch roller, or as pogertt's picture in post #7. I thought pogertt's method of calculating the amount remaining was quite ingenious even though it does introduce an added source of measurement error.
For the wider fabrics perhaps a long roller drum floating in a U shaped loop of fabric might work. In the case of stretchy material you might have to do something similar to what was done on magnetic tape drives to feed the material without stretching it.
Thank you for your compliment kwinn.
Depending on the distance between each WIF Pulse the smaller the distance, the more accurate the calculated values.
For example: WIF Roll diameter of 1.5" = Circumference of 4.7123 inches.
With a 1024 PPR encoder and edge detection resulting in 4096 edges per revolution, each of the 4096 pulses will measure 0.00115 inches of material being measured.
An 8 inch diameter roll should then produce 21845.3333 pulses.
If the material removed is .045 thick, diameter is reduced is reduced to 7.91 after 1 turn, providing 21599.5733 pulses.
Depending on the precision of the floating point math results, will affect accuracy of prediction of material remaining.
In the industrial application of this calculate thickness and area scheme a roll of paper approximately 60,000 feet long was calculated to a length within 15 feet of actual length after measuring 10 turns of the roll being unwound for an initial error of .025%.
Pogertt