Wireless scales

I am new to arduino in every way. I will soon be attempting to produce a set of 4 wireless scales, and would like any guidance available. Each scale will consist of a 15" x 15" x 1" aluminum pad, hollowed out, with a strain gage glued to the inside. This will be my load cell. I have ordered the strain gauges, voltage amplifier boards, several NRF24L01+ transceivers, and an Atmega 328 arduino with a 2.4" TFT touch screen display. I would like for all 4 scale pads to transfer data to a wireless handheld display unit that will simultaneously display the weight on each scale, as well as display "left side percentage" and "rear percentage" I will also need a tare function on the touch screen display. These scales will be used for my Dirt Late Model racecar. there are some companies that sell a very similar product that starts around $1800. I figure if I can make this work I will have only about $300 involved.

Will this work? What other hardware do I need? and is there any videos out there of anyone doing this exact same thing?

Have a look at this Simple nRF24L01+ Tutorial. The second example (using the ackPayload feature) can be extended to work with several slaves. I would arrange for the handheld unit to act as master and call each of the others in turn to collect their data.

...R

Load cells usually use four strain gauges in a Wheatstone bridge arrangement.

How do you propose to use just one strain gauge?

Jremington: I purchased full bridge strain gauges. So each one is a full Wheatstone bridge in itself. I hope this will work?

Also wondering, will I need an arduino board in each scale? Or can I simply hook up the strain gauge to the voltage amplifier to the NRF24L01 board and transmit that signal without an arduino board in each scale pad?

The standard NRF24L01 breakouts do not include an MCU, are not programmable and do not have an ADC. They must be driven by an MCU of some sort. Arduino is fine. Are you thinking of some other type of module?

Jremington: I'm so new to Arduino I haven't even received my arduino yet. I ordered it on eBay. I understand very little electronic lingo, but will have an Arduino project next semester in one of my engineering classes. I didn't know if I needed to be able to program the transceiver, but suspected that may be the case. So to clarify the components that I will need: (Each scale pad)= 1 full bridge strain gauge, voltage amplifier,arduino 328, transceiver, and a bunch of wires and solder.

I may not build a display unit as I have downloaded "Iraceweigh" from Intercomp. That is a free app courtesy of Intercomp that works with their wireless scales. The frequency it operates on, per their sales rep is "ISM 2.4GHz, 802.15.4"

Can I communicate with that via the scale pad configuration I described above?

Thanks in advance...

This app wants a channel, PAN ID, and encryption key for each scale. Will I be able to program these things on the arduino?

Can I communicate with that via the scale pad configuration I described above?

This app wants a channel, PAN ID, and encryption key for each scale. Will I be able to program these things on the arduino?

Probably not. The app and the wireless scales likely use a proprietary communications protocol.

I would recommend following one or more of the on line tutorials and learn how strain gauges, strain gauge amplifiers and Arduinos work before worrying about the display technology. Here is one example: https://learn.sparkfun.com/tutorials/load-cell-amplifier-hx711-breakout-hookup-guide

Thanks for the shared link. I enjoyed that. I do have a good understanding of the math of how the strain gauge system works, and have made it through Physics 1 and 2. My biggest downfall is that I have no experience with assembling electronics. I can however read a schematic. The strain gauges I have purchased are "1000 ohm full-bridge EB foil for analog sensor" on eBay. These have 4 wires. I am hoping to be able to use only 1 in the center of my scale pad, but am not against using 4 if need be.

For a little background on myself, I am a 35 year old, Project Engineer at a mid size, aerospace machining and engineering company. Before taking this job 2 years ago, I was a CNC machinist and programmer for 15 years. I am currently pursuing a mechanical engineering degree, of which I am about half way through.

Attached are pictures of the strain gauge, and a screen shot of the scale pad as designed in solid works. This screen shot shows the displacement of the aluminum pad, from the bottom side, with 750 lbs applied to a 36 square inch area centered on top of the scale (our racing tires are 12" wide and I estimate a contact patch of 3") According to solid works, the pad will be displaced .005" I would think this should be enough to get a good reading with the strain gauge.

well the attachments didn't work.

Chrisrw28: According to solid works, the pad will be displaced .005" I would think this should be enough to get a good reading with the strain gauge.

Don't guess - calculate it. As a mechanical engineering student, you should know that strain gauges don't measure "displacement." They measure...strain (change in length divided by original length). Does SolidWorks tell you what the strain is on the bottom of the pad at the center under your design load? Will it (the strain at your design load) be "enough" (or too much?) for the rest of your system to use?

I will see if Solid Works will give me the strain value. As far as the rest of the system, it says I have a safety factor of over 17.

I know that strain in change in length over initial length. When I used the term displacement, I skipped past explaining that displacing a flat plate in the center would create an arc of longer length than the original flat plate. BUT, I have not yet calculated that in terms of length. But I will.

Can anyone tell me how to determine the range of a given strain gauge? I ordered 1,000 full-bridge strain gauges. What is the correlation factor between ohms and amount of strain that can be measured?

Chrisrw28: I know that strain in change in length over initial length. When I used the term displacement, I skipped past explaining that displacing a flat plate in the center would create an arc of longer length than the original flat plate. BUT, I have not yet calculated that in terms of length. But I will.

Your reply would have been more reassuring if you'd used words like "modulus of elasticity," "stress," "moment," "linear range"......you know, "mechanics of materials 101" terms.

Chrisrw28: Can anyone tell me how to determine the range of a given strain gauge? I ordered 1,000 full-bridge strain gauges. What is the correlation factor between ohms and amount of strain that can be measured?

The datasheet for the gauge in question is always a good place to start. And a quick Google search on "strain gauge basics" will lead you to some very helpful material for self-education...including the answer to the "correlation factor" question.

PS: if you are studying to be a "mechanical" mechanical engineer (as opposed to an HVAC/R type), pick up a copy of Roark's "Formulas for Stress and Strain." I know these days it is easy to build computer models of "things," but it is always good to check the output "by hand," and Roark's book has a wealth of information that will allow you to do that.

In fact, it has data for rectangular plates centrally loaded with a rectangular patch... ;)

Normally fullscale on a strain gauge is about +/- 3% in length and therefore resistance. But they have a large temperature coefficient, and are generally used in a bridge arrangement of 4 to cancel this out.

Normally to measure a force whose exact position is unknown a load cell is used - google it

regards

Allan

" Does SolidWorks tell you what the strain is on the bottom of the pad at the center under your design load? Will it (the strain at your design load) be "enough" (or too much?) for the rest of your system to use? "

[/quote]

Solid Works analyzes the strain in the area where the strain gauge will be as 2.066 x 10^-5 (micro meters/meter) Will this be enough for the strain gauge to measure? I can't seem to find an answer to this question. I did find a maximum strain on the MFGR website, that value is 20,000 micro meters per meter

DaveEvans: Your reply would have been more reassuring if you'd used words like "modulus of elasticity," "stress," "moment," "linear range"......you know, "mechanics of materials 101" terms.

The datasheet for the gauge in question is always a good place to start. And a quick Google search on "strain gauge basics" will lead you to some very helpful material for self-education...including the answer to the "correlation factor" question.

PS: if you are studying to be a "mechanical" mechanical engineer (as opposed to an HVAC/R type), pick up a copy of Roark's "Formulas for Stress and Strain." I know these days it is easy to build computer models of "things," but it is always good to check the output "by hand," and Roark's book has a wealth of information that will allow you to do that.

In fact, it has data for rectangular plates centrally loaded with a rectangular patch... ;)

Thanks for the reference to Roark's book, I will check that out. I haven't made it to mechanics of materials yet, just finishing up D.E. And Chemistry 111. Wrapping up the associates degree at the community college before transferring to UAH in the spring.

2.066 x 10^-5 (micro meters/meter)

Please double check that figure. That would be less than an atomic diameter per meter.

Even if it is really 2.1 x 10^-5 meter/meter (or inch/inch, or furlongs/furlong), or 21 um/m, then that corresponds to a stress of only about 200 psi in aluminum (stress = modulus of elasticity x strain...for linear elastic materials...which aluminum can be considered to be, for practical engineering purposes).

200 psi is tiny. For comparison purposes, the tensile "yield" stress of unwelded 6061-T6 is 35 ksi (35000 psi). The strain associated with 200 psi (that is, 21 um/m) is unlikely to be enough to be reliably measured by you. Consider that your strain gauge apparently can take up to 20000 um/m. If the strain reported by SolidWorks is in the ballpark (check it with Roark, or even just do a simple beam calculation), then you need to make your plate thinner to increase the strain.

How much thinner? Well that depends on the rest of your system. If you would take the time to do some self-education (as previously suggested, google strain gauge basics), you would be able to figure that out on your own. It's not hard....

And regarding "2.066"...let's not kid ourselves about knowing this strain to that many significant digits.