I want to build a telescopic device that measures closure between a roof and the floor. The device must be able to measure with a resolution of 0.5mm or smaller but no bigger and it must be very accurate. Any idea's of how I can build such a device?
Leadscrew
calvingloster:
that measures closure between a roof and the floor.
Do you mean distance, or are you talking about an opening roof?
When you mention a resolution of 0.5mm what percentage of the total distance is that?
...R
Robin2:
calvingloster:
that measures closure between a roof and the floor.Do you mean distance, or are you talking about an opening roof?
When you mention a resolution of 0.5mm what percentage of the total distance is that?
...R
Well the distance between the roof is about 1.5m. And the roof moves down slowly over about 6 months. I need to monitor how much the roof comes down relative to the floor
Sorry I meant to say the distance between the roof and floor is about 1.5m. Also the maximum closure anticipated is about 20cm. So the device does not have to measure more than that
I must be misunderstanding something because I can't imagine how it takes 6 months to close a roof. How long does it take to open?
What sort of physical measuring system can be considered?
Can it be in place all the time?
Could it be a pillar reaching from the floor and nearly touching the roof?
Or do you want something portable that is only in place for a few minutes to take a measurement?
A rough drawing of the system would be useful.
...R
Robin2:
I must be misunderstanding something because I can't imagine how it takes 6 months to close a roof. How long does it take to open?What sort of physical measuring system can be considered?
Can it be in place all the time?
Could it be a pillar reaching from the floor and nearly touching the roof?Or do you want something portable that is only in place for a few minutes to take a measurement?
A rough drawing of the system would be useful.
...R
It's for use underground. To measure closure of tunnels and excavations. Closure happens over periods of 3 to 6 months depending how fast the excavations are created or advanced. We need to monitor the closure to ensure the roof is coming down. If it does not come down as the tunnels develop then we anticipate a seismic event. I will draw a picture and post it now
Here is exactly what I want to build. We are currently using these but they are stupidly expensive. So I want to build my own
OK, things are moving forward - but very slowly. I feel like I have to fish for every single piece of information. How about writing a 10 or 20 paragraph essay that describes the entire system that you need.
If you drawing is a guide it looks like you envisage devices that stay in place for 6 months or longer.
Do they have some way of continually reporting data or does someone enter the tunnel periodically to read the measurements?
This has implications for whether you want an unattended system that is very reliable in the face of water, dirt, insects etc ?
How serious would it be if one of the devices failed during the 6 month period?
A possible way to measure would be to have a device (perhaps a screw) that could be extended until it detects contact with the roof (or contact with the floor if a roof mounted device would provide better protection from water etc). A stepper motor could be used to extend the screw and the number of steps until contact should give an accurate masurement. Also there is nothing in principle against having a long screw so that accuracy and distance could be combined. It might also be desirable to retract it when not needed so as to protect it.
It should be possible to program an Arduino so that it would be apparent if it had failed.
A standard Arduino Uno is unlikely to be suitable as the plug-in connections would not be reliable. But there are several alternatives with screw or solder pad connections.
...R
Sorry that u keep asking for info. This device gets put underground and it logs closure on a time scale so we can see exactly when it moved and how much it moved. The data transferring and all that is not a problem. The only problem I am having is making an electrical device that measures the movement. I was thinking maybe a resistive piece of wire with a wiper on it that changes resistance as the device gets pushed closed. Almost like a potentiometer.
It must be very simple and must use very little power. The device would ideal sleep and take a reading every 5min or so then log it according to the time. I cannot use motors or anything that is making things way to complicated
I also had in mind maybe creating some gear system that turns a potentiometer slowly as the roof moves down. Ideally I would use a pot that rotates as many revolutions as possible. I'm not saying this will work but this is what I'm going for. Simple as passive if I could say that? The arduino should apply a voltage to the circuitry and read a voltage of a voltage dividing circuit or something(i don't know what, that's why I'm asking) to determine movement
sound like you have some opportunities.
the first that comes to mind it to put a ping sensors in a tube. the inside of the tube coated to prevent echos.
they do make single horn ping sensors. the easiest and cheapest place to get them seems to buy a wireless tape measure and hack it.
another possibility is to use a laser, but as I understand they are not able to get the resolution you are looking for.
using a mechanical device, you could use a ball screw. some ball screws also have a very fast spiral.
http://www.kammerer-gewinde.com/products/ball-screw-threads/
the ball-nut will free fall along the length. using an encoder that would offer output based on rotation, as the shaft rotates, the encoder will output pulses (or serial) that will report the rotation.
using a screw that is spring loaded to hold it up to the ceiling, the encoder would then report rotation. a screw that spins one rotation and one rotation is 1 inch, that has an output of 1,000 pulses per inch would offer resolution be about 0.0254mm. you could get higher pulse counts, but as you increase resolution, the costs increase. the serial ones start with low resolution of 11 bits, you can get 22 bits, but costs increase.
http://www.dynapar.com/uploadedFiles/_Site_Root/Products_and_Solutions/AC110.pdf
using one of these, you would put in your instrument, set it, zero it, and then visit it when you wanted.
if you look inside of a old floppy drive, they have such a fast spiral screw. very small device.
I started with ball screw. the reason is that even a non-fast taper will allow a nut to 'fall' by gravity. A common screw will lock the shaft under weight. they do make multi-start ACME screws so you could get similar results with a less expensive device.
+++++++++++++++++++++++++++++++++++++++
actual measuirng devcies. look a linear glass scales like they use on lathes and milling machines.
easly read to 0.001" and not expensive. much less than an encoder before a screw !
if that is broken, search; iGaging 24
iGaging is the brand 24 is the longest they have listed.
costs about $50.
you could buy a caliper and cut the jaws off and then hack into the serial data. there is a lot of information about how to do that on the net.
calvingloster:
I also had in mind maybe creating some gear system that turns a potentiometer slowly as the roof moves down. Ideally I would use a pot that rotates as many revolutions as possible. I'm not saying this will work but this is what I'm going for. Simple as passive if I could say that? The arduino should apply a voltage to the circuitry and read a voltage of a voltage dividing circuit or something(i don't know what, that's why I'm asking) to determine movement
this came to mind as well. the screw would need to have fast spirals so it slipped when pushed. and instead of a pot, an encoder. encoders allow continuous rotation and are widely available.
calvingloster:
Sorry that u keep asking for info. This device gets put underground and it logs closure on a time scale so we can see exactly when it moved and how much it moved. The data transferring and all that is not a problem. The only problem I am having is making an electrical device that measures the movement. I was thinking maybe a resistive piece of wire with a wiper on it that changes resistance as the device gets pushed closed. Almost like a potentiometer.It must be very simple and must use very little power. The device would ideal sleep and take a reading every 5min or so then log it according to the time. I cannot use motors or anything that is making things way to complicated
a linear encoder will offer absolute location information. location is based on the glass scale so will not be lost if power is lost.
low power, runs on batteries.
Widely used in home machine shops. lots of guys who have hacked these and some that have used an arduino
You could possibly use a multi-turn potentiometer - they are a lot better than single turn versions.
But I think an encoder would be the best option if you don't want to use motors.
Another option might be to have a visual sliding scale and use something like a RaspberryPI to take pictures of the scale. It has the advantage of being self calibrating.
I suspect that a reading a few times a day would be sufficient.
...R
PS I guess "closure" is common jargon in your business - but not in the Arduino world. Be careful that you understand our jargon.
Thanx guys, great idea's. I think I'm going to try get a linear optical encoder or something. I just have no idea where I can get them in south africa. Can anyone maybe explain what a linear glass encoder is? I've gathered that it's some type of linear optical encoder but as to why it's called a glass encoder I have no idea? And why they so expensive?
calvingloster:
Here is exactly what I want to build. We are currently using these but they are stupidly expensive. So I want to build my own
That thing is so expensive because there is death liability attached.
It has to be right.
if you understand how rotary encoders work, the glass scale are the same, with the white and black sections laid out in a line.
http://www.parkermotion.com/dmxreadyv2/blogmanager/blogmanager.asp?post=motor-feedback-options
and
http://www.zsinstruments.com/scales_about.aspx
they cost a lot because there is no high volume production and the application is for high end business. the customer will pay a lot so they charge a lot.
I think that the magnetic ones use technologgy like they use on hard disks. this offers some features like working if they are dirty, that the glass ones do not.
if you have access to rotary encoders, the screw method might be easier to make and cost less.
For the life of me I can not find the example I am looking for but there are inexpensive linear scales http://www.ebay.com/itm/12-DIGITAL-DRO-TABLE-READOUT-SCALE-for-BRIDGEPORT-/310224430119?pt=BI_Heavy_Equipment_Parts&hash=item483ad12427 that people have written libraries to read them. The scales provide pretty decent 0.001" accuracy and hold the readings for long periods of time under power. Sorry I think in inches but 0.001" is about 0.025 millimeters so I believe your requirement is met and I am pretty sure they come in at least 25 cm lengths.
Someone else can probably chime in and direct you to the the code examples to read these scales.
Hope this helps,
Hey man, where in SA are you?