I have been tossing around a new project idea - a newspaper box audit system.
It could be valuable to know how many newspapers are taken from a box at what times and possibly when the box is sold out. This could help distribution, ad sales, and performance analytics for both the business staff and news/design staff.
The catch is, I am thinking primarily of my school’s paper which is free, and not necessarily a regional paper (that may change if there is a demand, but thats business mumbo jumbo talk). Our school paper is free and a lot of the boxes do not have a door, they are just covered boxes you can reach in an open side to get your paper.
My idea is to have a pressure sensor (or maybe four - one on each corner of a platform) that would weigh the stack of papers and a microcontroller that can log when a paper is taken off the stack.
I was looking at some force sensors and my question is, will the tolerance of these sensors be within an acceptable range? It would be great to be able to timestamp when an individual paper is taken, but I could back the spec down to 5 or 10 papers. I haven’t weighed the stacks or the individual papers, but they are pretty light. Also, the weight may change slightly with humidity.
Does anyone have any suggestions on the best option for the sensor/s and the arrangement? Is this idea realistic or is there a better implantation?
I can flesh out a few more of my ideas if anyone is interested in the project.
Best thing to do would be to get some force sensors and do some experiments with them I think.
You may not be able to detect exactly how many papers are there, but you may be able to detect changes in the force sensor to indicate a paper being removed.
Thats the plan. I just wanted to put some feeler out to see if anyone had any suggestions and ideas. I’ll post some updates as I go along. I think I’m going to order the sensors and probably an arduino after my last exam tomorrow.
I think measuring weight has real problems. Newspapers vary in the number of pages. The weight will change based on differences in the type of paper used. Also, humidity will be a factor in the weight.
Why not use a distance sensor? Place it in the upper part of the case pointing downward? You know what the distance is when it’s empty. no need to factor in anything else.
a distance sensor, I didn’t think about that one. I wonder what the resolution would be for that. I imagine you could estimate on the order of 5-10 papers?
RE: weight sensing - don’t look at absolute weight; look at changes in weight.
The weight should change in obvious steps as people take papers. Any slow drift would be sensor drift or humidity absorption. Also, you could look at the size of the step to determine how many papers were taken - the step size would always be a rough multiple of 1 paper.
I am definitely thinking about playing around with different ways of calculating it, and I think some variation of measuring the steps will probably be the way to go.
I talked to one of my professors about it and he said an alternative to force or load sensors would be to measure the pressure of a balloon or bladder under the platform for the papers. That might be something to look into.
Lastly, I just realized there is a third type of box in addition to the hinged door and the open steel box. The cheapest box is a rack that has shelves of wire (I think three) that hold the papers and a plastic base. Think of what you typically see free pamphlets and papers on. I don’t think the weight or the distance sensors would work in this case, at least not with some serious rethinking. Any ideas for that type of box? I’ll have to go look at one this week to brainstorm, I’ll try to get a photo.
busonerd:
RE: weight sensing - don’t look at absolute weight; look at changes in weight.
The weight should change in obvious steps as people take papers. Any slow drift would be sensor drift or humidity absorption. Also, you could look at the size of the step to determine how many papers were taken - the step size would always be a rough multiple of 1 paper.
I’m not trying flog the distance sensor idea but it seems like anything that requires a modification to the box isn’t a good idea. Have you thought through installation?
Why do all that processing and filtering when distance is a very simple solution? Occams Razor, baby!
It could be built as a complete unit. Installation is simple - just install the unit at the inside top of the box facing downward. Load cells, bladders or similar would take a bunch of installation work.
Crazy! I just bought an arduino to work on this exact project. I’m pysched there is someone else interested in this project!
Background:
With a local vegetarian advocacy group, I manage a handful of free literature stands around my city. We use this model of stand: http://www.goplastics.com/ss-5c.shtml In them we offer free vegetarian starter kit magazines, with recipes, health info, reasons why people become veg (actually, like this one: http://www.goveg.com/ORDER.asp).
To track distribution numbers I bought a cheap digital scale on ebay. I measure the weight of the remaining magazines before I refill the stand, then weight the stack again afterwards. Take the difference to know how many I added and divide that by the weight of one magazine to get the number of mags.
About the actual circuit:
I think weight sensors are the best option. For my application, just measuring once a week when I show up to fill the stand would be sufficient–no need to have a tweeting stand or take measurements daily to get a more precise rate. But knowing the daily rate would be kinda neat too… Anyway, I’ve been looking at the flex sensors: http://www.sparkfun.com/commerce/categories.php?c=143 Not sure what other kinds of weight sensors exist.
I do not think a height sensor would work well because the magazines often do not stay in a perfect stack. Being outside, they often get bumped and moved. Sometimes they get crumbled or folded. The stacks don’t stay vertical very well. I guess I could mod the stand with wood to make the magazines stay more orderly. Seems like using weight is less obtrusive.
Another thing I foresee being an issue is taring. If the weight sensor is placed permanently under the newspapers/magazines the circuit may not have a zero starting point. Not sure how to address that.
