How to detect usage of height adjustable table?

Question

Hi at element14

We have been trying to solve this matter for quite some time now, and we simply haven't been able to come up with a satisfying solution. We are working on a little "bot" placed on top of a height adjustable desk. The bot has to measure whether the desk is in a sitting or standing position. So we do either need to measure the movement and direction of movement or the height of the table. We have had some luck using proximity sensors (IR and US), but they are very dependent on the environment and placement, as e.g. a desk lamp can make the readings worthless.

We are now working with an accelerometer and/or a barometric sensor, but the acceleration of a table is very small and the height difference of the table is only 20-40 cm's. So the sensors have to be very precise and sensitive.

Soo, do you maybe have other ideas to what sensors we could use? Or maybe some insights into how we can get the accelerometer and barometric sensors to work properly? - would be a of great help

Best regards Mads

mcb1 · Answer

So we are trying to register everytime the users changes working position from sitting to standing and from standing to sitting You could use an ultrasonic sensor and detect the height to the ceiling. That would give you a relatively clear path and is uniform. Some code to establish when it's stationery and then record it, or transmit it to a collection device. Cheers Mark

DAB · Answer

Hi Mads, Have you considered using a capacitive sensor mounted to the desk? The TI sensor system can accurately detect objects up to about 50 cms. Another way would be to use a light and a camera to measure spot size, but now you are adding some complexity. Maybe an ultrasonic sensor? DAB

vandia · Answer

On the note of what you're trying to do, you've giving detail of your method, but it still leaves your end goal rather vague.... especially with "acceleration of a table". It leaves the questions: "Is this to be done on multiple/different floorings under the table", "is the table moving", and "how frequently does the table change heights". If you want to stick with the accelerators: I haven't used them before to know their accuracy, but I remember an Arduino SHARP library that averaged 5 IR scans into one value.... you could use 2 accelerators, one on each end of the table, and run each end through that same thing to alleviate some of the fluctuation. IR has the highest fluctuation of any simple sensor I've used. It varies based on ambient light, and even EM waves. Sonar varies by texture of what it's to hit, which is why a question is the flooring. I also have an idea for determining if you're on carpet or hard flooring if you wanted it.

beacon_dave · Answer

Are there any restrictions on fitting sensors to the table ? Are you trying to have a device that just sits on the table and logs the information, or can the sensors be physically attached to the table ? Are the tables of the same design ? If you can attach the sensors to the table then: - mechanical limit switches would be a simple choice - a linear encoder could give you more detail about the heights the table was adjusted to - a rotary encoder with rack and pinion arrangement could also be used If the floor covering is giving problems with the likes of ultrasonic sensors then perhaps locate the sensor on the floor so that the reflection is coming off the underside of the solid table rather than off a carpet.

koudelad · Answer

Hello, is the table height adjusted by some electronic control or a manual handle? I use two types of these tables (both electronically controlled) and the manufacturer offers them with two types of controllers. The first one has only up and down keys, the second one has up and down keys, 7 segment display to display the height in centimetres and four keys to store / restore value from a memory. These controllers are interchangeable which means there is probably a rotary encoder on the motor shaft and only one of these controllers can make use of the output The controller of course has then no idea about the ground clearance etc., it just knows the relative position to the lowest position. (Note: I haven't personally dismantled the mechanism.) I would try to do my own version of the controller in this case. If you wanted to attach the measurement sensor to all kinds of mechanical tables (parallel with the floor), an ultrasonic sensor is one of the options. You don't even need to measure the absolute ground clearance, just the difference between the positions.

balearicdynamics · Answer

Mark, I think this solution is totally reliable with a very low impact. It has been successfully used by the first drones to detect their height and avoid terrain obstacles. There instead a sensor needed together. Suppose you stay sit on the the desk and it is not rare that your feet impact the sensor. Ok, it can be placed in a very good position but the risk always exists; When I sit to a desk I can have to put downside my backpack, my skateboard, my shoes to wear a more comfortable pair while I am working and so on (these are all real situations I saw while working in the Netherlands past year). So to avoid the ultrasonic (the best solution anyway) is wrong a second sensor should be added to confirm that it is the table moving and not the ground or some part of it. Also for this I see more than one easy solution: If electronic movement it is sufficient to detect when the motor is on (no matter where it moves, the sensor detect the distance from the gorund (or the relative distance if something is on the ground in that moment). If manual movement a cap sense placed in proximity of the desk unlock lever is another possible solution for bare mechanical movement An inertial accelerometer is sufficient put beside the table to inow that it is the table moving instead the rest. Not only but the ultrasonic sensor is able with fewl ines of code to self-calibrate to a relative height when something is put under the table. Enrico

svendsen2032 · Answer

I'm not sure I do fully understand. Do you suggest to use a capacitive sensor to measure the electrical field of the motors? Or, how do you figure it's possible to detect movement with a capicitive sensor? And a camera might be a bit too complex and we do experience users linking cameras to surveillance, which ain't good. We have been using ultrasonic sensors and they do work, however they are very dependent on the environment.

How to detect usage of height adjustable table?

Top Replies