Suggestions for a very cheap sensor to measure elongation of many springs

[gonsp]

Hi everyone!
I'm choosing a design for a matrix of sensors and I think I need some help!

TLDR: Check the last 3 bullet points

For a personal project, I need to build a multiplexed matrix of 560 sensors (14 x 40), each one measuring how elongated a spring is. That is, I have 560 springs, placed vertically with the bottom side fixed and the top side supporting a variable force. Hence, the top side of each spring will be farther or closer to the bottom side depending on the experienced force. The maximum distance (length of the spring without force) is 7cm, and the minimum distance is 2.5cm. That is the range that I want to detect. I can place a sensor on the bottom of the spring, on top, or on both.

I originally considered to use a variable resistor as a sensor, like a sliding potentiometer, and to multiplex the sensors as follows:

In that diagram, the arduino (mega) would iteratively sample the resistors by pushing 5V to only one of the rows while reading drop voltages with the analog inputs.

An important constraint is my budget: since I need 560 sensors, each one needs to be really cheap so the total cost is not too large. I want each of the 560 sensors to cost less than 0.7€ each. I've found several cheap sliding potentiometers like this one, but they don't fit in the space that I have... (ask me for details if you need more info about the space constraints).

So, I'm in the search of alternatives:

• Sliding potentiometers don't fit although some are cheap
• With a force-sensitive resistor on the bottom of the spring I could measure the force, and by knowing the Hooke's constant of the springs, I could obtain the current length of the spring. However, they are too expensive (3.7€ each).
• An ultrasonic distance sensor also doesn't fit and is expensive. However, I have considered to just put the transmitter speaker on the bottom of the spring and the receiver speaker on the top of the spring. This way, I would decrease the size and the overall cost, since each spring only has two speakers and not the entire sensor (then I would reuse a small number of "sensor boards" outside the sensor matrix to control the speakers). However, not all the speakers fit and they are also too expensive (3.37€ each pair of speakers)...
• A VL53L0X Time-Of-Flight laser sensor would be ideal, but they are too expensive as well...

Since I'm desperate in my search, I'm also considering the following alternatives:

• Build my own sliding potentiometer. I would love to receive some feedback and indications about how to mount 560 of such potentiometers myself, and what to use as a resistor. I'm not very inclined about this approach since the accuracy of DIY potentiometer might not be great, and it can take me a lot of time to build 560 of these...
• To use a led and a photoresistor to measure the distance between them. This would use the fact that: There would always be constant darkness, the maximum distance is "small" (7cm), and that the led and the photoresistor would be perfectly aligned all the time. Do you think this is a feasible solution?
• The same as the previous option but using a laser instead of a led. Would that work? Would the laser intensity degrade enough in 7cm travel to be able to consistently detect it with the photoresistor?

I would love to hear your feedback and suggestions that I haven't contemplated.
Thanks!

[gilshultz]

I would suggest you go online to Aliexpress and check, they are in your approximate range. Find two and then contact the supplier with the quantity you want and expect some good pricing. You could also try the parent division Alibaba.

[DaveEvans]

Maybe you could do something with pieces of this stuff under each spring: Pressure-Sensitive Conductive Sheet (Velostat/Linqstat) : ID 1361 : $4.95 : Adafruit Industries, Unique & fun DIY electronics and kits

Or this: Conductive Rubber Cord Stretch Sensor + extras! : ID 519 : $9.95 : Adafruit Industries, Unique & fun DIY electronics and kits

[wvmarle]

You can not multiplex pots like that, they will all influence one another. Each needs to be read independently.

What are your actual requirements in terms of force and/or distance measurement (actual forces, actual distances, required resolution & accuracy)?

At your budget of 0.7 Euro each you may not be able to get FSRs, but close - the cheapest I find with a quick search is about 1.3 Euro a piece; at your quantity it'll be less.

The cost of the sensors will pale in light of the rest of your development cost. Your hours, for starters. You will easily spend a couple hundred hours total to complete this project, from sourcing the components to going through several iterations of testing the sensors (do they even sense what they should sense, how to read them accurately) to slowly building up to the total 560 sensors, all the way solving problems ranging from where to route all those wires to getting the data fast enough and in a useful format. Then you have to buy all the ADCs you need to read all those sensors (one per sensor). The equipment to read those 560 ADCs. The wiring and other rigging.

Seriously, I hope you have at least a 5-digit budget for this. A good 4-digit one if your hours are not counted against it. Otherwise, have a word with your employer on what is realistic in terms of time and money investment.

[hammy]

Why do you want to do this ? There may be a different way