h a l f b a k e r yI didn't say you were on to something, I said you were on something.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
This looks like a standard laser pointer, perhaps a bit wider to accomodate a sensor, and a length dial.
The dial is used to adjust the length of a line that the beam projects on a surface. Rotating the dial makes the line longer or shorter.
The sensor detects the distance from the pointer to
each end of the line, then uses simple trigonometry to determine the length of the line projected on the surface.
The pointer will then constantly adjust the laser and lenses so that the line projected will stay the same length even as the distance or angle to the surface changes.
In use, the user will adjust the projected line to a known length object, say a one foot ruler, then be able to project an exact one foot long line on remote surfaces.
Specifically I am thinking of this as a way to check spacing of studs and fasteners for code compliance. An inspector could set the pointer to project a 12 inch line, then check the spacing of screws on a cathedral ceiling without using a ladder. Then adjust the beam to sixteen inches and check the spacing of screws on the walls.
Like one of these?
Dambuster_27s_20laser_20bargepole
[AbsintheWithoutLeave, Jun 26 2006]
At the end it gets a tiny bit relevant.
Resume_20Rifle
My idea to keep the orientation of a projected image. [GutPunchLullabies, Jun 26 2006]
Like one of these?
http://i4.tinypic.com/15rfxc2.jpg
[nihilo, Jun 26 2006]
Dual light source locating and tracking system
http://www.freepate...ne.com/4734572.html
One of many "patented" variations on a theme, used in camera rangefinders and particulate detection systems. [reensure, Jun 30 2006]
Disto laser measuring tool.
http://www.leica-ge...cpd/en/lgs_3849.htm
Really one of the best things for measuring buildings. [zigness, Jun 30 2006]
Various laser stuff
http://www.lascolaser.com/index.php
A lot of interesting things, but none quite like this idea. [zigness, Jun 30 2006]
[link]
|
| |
How about a pole with a ruler stuck to the end of it? |
|
| |
/How about a pole with a ruler stuck to the end of it?/ |
|
| |
Tried it. He kept wanting beer and sausage breaks. |
|
| |
[Absinthe] No, not at all. My idea creates a line of a specified length at any distance from the operator. Set it to a one meter width and it will create a one meter line anywhere in its range. |
|
| |
So when you get in a barroom discussion, and tell your buddy that his arse is as wide as that pool table over there (or that biker) you can check without leaving your table. |
|
| |
But how does the pointer know the distance and angle? (Looking for something more than distance and angle sensors, BTW.) |
|
| |
For distance, the same way an autofocus camera does would be one way. Or pulse the laser and time the reflections. |
|
| |
Angle is calculated, The distance measurement at each end of the scan gives the length of two sides of the triangle. The desired length is the third side. Simple trigonometry lets you calculate the appropriate scan angle. |
|
| |
I'm not sure that will give you the angle since it will be moving all over the place, but if it did, you could show a reticule on a camera display more easily, simultaneously getting photographic evidence of compliance or non-compliance. And no need for the laser. |
|
| |
Sorry, [ld] I wasn't done with the anno. I intended to answer both questions. See above. |
|
| |
Here is what I don't get - I understand that a laser pointing down each of the sides of this imaginary triangle could measure distance and calculate the length of the "base" (assuming there is a straight line between the two points). But how does the device know what the initial angle is between the two lines? I can understand a device where you set an angle, then it can give you a readout of the space inbetween, based on the readings from the two sides, but it would take lots of computation and correction for it to be able to adjust the angles of the lasers in real-time to be able to keep a given width. I have to give a fish, because I think you could get a reading of a width from the 2 lasers, but it would be much more difficult for a little pointer to adjust the width under described conditions. |
|
| |
Maybe a bit more technical noodling would let this attract more buns. |
|
| |
Imagine a laser. Shoots a nice straight beam of light. |
|
| |
Add a tiny vibrating piezo electic crystal, or perhaps a tiny rotating prism to make the beam scan a section of a plane. |
|
| |
Now use a lens to restrict the width of the scanned planar section. This lens is on a linkage allowing it to be moved relative to the rest of the assembly. The farther it gets, the narrower the angle. Call this angle Theta. |
|
| |
At each end of the scan, a sensor measures the signal return time for the laser and calculates the distance to that point. |
|
| |
Call this distance A. When this point is captured, the old value of A is shifted to be called B (this old value is distance measured at the opposite end of the scan. |
|
| |
We already know the desired length of the third leg of the triangle, that was defined by the user, call this C. |
|
| |
For each Length A, B, and C there is a unique Theta. (Is this the "Law of Sines"?) |
|
| |
So the system can adjust the prism/lense distance to create that value Theta. |
|
| |
This proces might take a few tenths of seconds, but that should be barely perceptable. |
|
| |
Most small lasers that draw a line do so by spreading a single laser point with a lens of some sort (don't know what they are called), so you are adding a mechanical device to spread the beam up to say 90 degrees. |
|
| |
So your device is then supposed to measure the distance to all the points (sampling) on that line, as the laser is moving (by whatever method), and compensating for the lens, which I guess it has to sense the position of, and then restrict the movement only to the section of the desired width. |
|
| |
It still sounds complicated - although a bit better. You still have the problem of figuring the initial Theta - although the laser could start wide and shrink until it hit upon the right size. Just sounds tough to me for a pen-sized device. And your estimate of tenths of a second sounds OK, but I imagine it would be based on the speed the lenses and the spreading mechanism. Sounds almost like the distance meter on a camera, but a bit more complicated. |
|
| |
I still think for a simpler device (and less costly), you could opt for being just a distance reader, with two distinct lasers you could angle apart. Thanks for explaining though. |
|
| |
I have not torn apart a laser pointer, but I know that some laser devices use prisms to form the line. Other mechaisms are possible. |
|
| |
Oh, and using two lasers (or splitting the beam) would work too. |
|
| |
The initial theta should be known from the design of the lens system. |
|
| |
If you use two lasers and restrict
yourself to measuring distances parallel
to the baseline of the lasers this would
be easy. You could mount two lasers
parallel to each other on the edge of a
ruler. Sort of like an extended pair of
vernier callipers. |
|
| |
If you want this thing to deal with
angled surfaces you're in for a difficult
time. Say the suface is 10m away and
slopes at 45°. You want the points
14cm apart. You are going to have to
accurately rangefind two distances
10cm apart at a distance of 10m. A 1%
error in rangefinding would be a 14cm
error in the length... and that's the
distance you're trying to measure. |
|
| |
Sounds great - but not sure it would work.... |
|
| |
Doesn't it assume that the surface you are projecting onto is 'flat' - doesn't slope away or towards you in anyway ? |
|
| |
[mono] No such assumption is made. the internals should be able to calculate for any flat surface. Running around a corner, or curve would be much more complicated, but a straight line distance not so much. |
|
| |
I can stop a laser at a specified distance by pointing it at a wall. |
|
| |
//no such assumption is made. the internals should be able to calculate for any flat surface// |
|
| |
Any *flat* surface granted - I was wondering when target surface is curved,like the side of a car) - or a bumpy surface like brickwork. |
|
| |
It might not matter that much, but the straight line measurement couldn't take into account the bumps - or could it? |
|
| |
I think based on the intended use, that a fairly simple sonic range finder combined with a spread laser beam would do quite well. This would not be intended for exacting measurements in unholy conditions but as a spot checking device in the hands of a professional trained user. The error caused by angle differences would be realtivly small if used properly(ie stand on opposite wall from cieling being measured and move side to side while checking so you are roughly directly in line with the joist). |
|
| |
Also use of a calibration standard could increase accuracy(ie program unit with standard dimensions of common building materials such as drywall or lumber, electrical boxes, PVC Pipe, Bricks etc., then just tell the unit to cal on something(this is the long edge of a 4X8 drywall sheet from my position) the unit then knows where to set the line based on the distance to the target and the "known" distance at the same angle. it can then adjust from there. It could also be equiped with a tilt/orientiation sensor that it could use to gauge changes in the angle of the surface being measured(if the rate of change of distance from the surface changes by more than a set amount the unit can warn the user that the measurements could be drifting.) |
|
| |
Remember the goal is not analytical precision but to quick check approximate spacing, if the need for ball busting arrises then you get out the ladder and the calipers and go to work. |
|
| |
[monojohnny] Yup, flat surfaces only. Basically you are showing the distance between two points, not measuring along a surface. |
|
| |
[phlish] Nope, heavy sabres only. You light sabre wusses need a workout anyway. |
|
| |
[jhomrighaus] You have got it. And yes, calibration would be something you would expect to do. |
|
| |
Basically this is a device to see if something is "close enough" to not require putting up a ladder and checking each little thing. |
|
| |
I think Galbinus_Caeli means width or height rather than length, which would make much more sense. |
|
| |
Shocked and appalled! Shocked *and* appalled at the thought of building compliance inspectors not doing their very valuable job in an exacting way. This 'quick reckoning' of yours is nothing more than half-assiness.
I say make them get their ladders out, and none of those new aluminium ladders either, it’s cast iron for you matey. Make them get their measuring tapes (those with the sharp edges and the always-too-strong spring recoil) to protect us from the big bad developers that are in it to cause suffering to the public.
We need more input from Council; where’s [crash]? <men in white coats come to persuade him that it’s nice inside, that getting close to a computer is not such a great idea, that there is work to be done and the halfbakery is not helping much> |
|
| |
Why draw a line? Why not just two dots the specified distance apart? Just have two range finding lasers, with a variable (precisely measured) angle between them. One to be the "start" point, and the other, the "end" point. First laser measures distance 'a' to the first laser point from the unit. Second laser measures distance 'b' similarly, and calculates the third side of the triangle 'c' using the exact angle alpha.. Simple control logic will allow the system to vary the angle until 'c' matches what you want. With the simple calculations involved (with a pretty tight PID rationalising the slightly varying inputs), you could probably measure 10-100 times per second, and therefore however shaky your hand is, it should keep a pretty good track of the distance. You don't even have to be perpencidular to the portion of wall being measured, which is the real beauty of the system. As long as the wall/suface is flat. |
|
| |
This is worthy of a patent, and is definitely not magic. In fact it'd just be a compilation of existing technologies. I'd bun it twice if I could. |
|
| |
Thanks [Custardguts], Two dots would work too, but I think that ergonomically a line would be more clear. |
|
| |
I don't have the time or energy to have two patents cooking at once, so if someone else runs with this, toss me a bone after your first million. |
|
| |
It would be easy enough to add a laser line inbetween the dots of the rangefinders. Just simple transfer of the needed angle to the beam spreader or prism angle. |
|
| |
Brief comment on stated use: |
|
| |
Inspectors just count how many nails per panel. For 12in spacing, if there are 3 nails in the field (the central part of a piece of drywall or plywood), that's roughly 12 inches. Not hard to do, and it doesn't need to be any more accurate than that. |
|
| |
For stud spacing, you can spot a wide bay in a group pretty accurately by eye, once you get enough experience to be an inspector in the first place. |
|
| |
Could this be done with a laser diode, an angled spinning mirror/prism, a single range finder, and a plate with a hole in it? |
|
| |
To elaborate... laser points at mirror/prism, creating an 180-degree-or-less sweep of light. The device has a rangefinder (laser, radar, sonar, whatever) pointing along the centre the segment. The laser 'sweep' is directed through a hole of known diameter, and the distance of the hole from the laser adjusted to allow a line of desired length at the distance measured. |
|
| |
Essentially: you know the width of the hole, the distance from the hole to the wall, and the desired length of line, so interpolate to adjust the distance of the hole from the diode. |
|
| |
Oh, and + for the concept. I want one. |
|
| |
You have the idea correct, but you need to be able to measure the range to each end of the line. |
|
| |
Imagine you are standing directly in front of the wall shining a regular flashlight at it. The lighted area is a circle, right? Now move so that you are pointing at the wall at a 45 degree angle. The lighted area is an elongated ellipse. |
|
| |
The line (or pair of points) projected by the constant length laser pointer is always the set length apart, no matter what you project to the wall. |
|
| |
You know, this idea is actually really brilliant. I think it's been a little lost on some of the other bakers... |
|
| |
Even so, these are my thoughts. |
|
| |
I've used the Leica Disto laser measuring device quite a bit. (I have the bluetooth model -- see link) It actually has a feature that lets you measure a wall by zapping the corners. It's out and back accuracy is to about a millimeter. The length accuracy is not so good since it depends on the user hitting the corners as well as the perpendicular point accurately. |
|
| |
I've been building a hacked up experiment using this thing as a quick and dirty digitizer to get large object models into the computer (i.e. buildings and rocks). |
|
| |
Since this device already knows how far away it is, it could report that to another device in realtime. (And it does, quite well with Bluetooth. I use an HP Pocket PC to capture the data) |
|
| |
The second device could respond by projecting the laser line using a variety of well established mechanisms (spinning mirrors etc). |
|
| |
I like this one. I wish the 'bakery had a lot more of this kind of thing some days. |
|
| |
[phlish], I'm disappointed in you. There's no magic in this at all. |
|
| |
[zigness] Thanks! That made my day. |
|
| |
If all of you people are here... there must *not* be a Trekkie convention in town...wow |
|
| |
[zigness]//get large object models into the computer// |
|
| |
Wow - you got any results of this on the net ? Sounds great. |
|
| |
With enough of these things working in parallel, you more than just measure stuff, you could 'sample' stuff in 3d. |
|
| |
Which would be great for getting computers to 'see' in 3D - something they struggle with I believe at the moment (resolving 'flat' images into the actual 3D profile). |
|
| |
I'd bun this again if they let me. |
|
| |
Nothing up on the net to share yet with the homebrew digitizer... those 'bakers that are curious about it can email me. |
|
| |
Okay, with you on that, [Galbinus]: could you solve that problem by having two parallel laser range finders to calculate your angle to the wall? And adjust the spread accordingly? As with [trekbody], I'm trying to think of a way to avoid having to sample a moving laser... |
|
| |
[m_Al_com] Simple matter of timing. Just do a range check at each end of the scan and use the previous two measurements to adjust line length. |
|
| |
In terms of the lens thing... I think this would work a lot better using a spinning mirror technique. The mirror spins, the laser is pulsed appropriately based on the position of the mirror in realtime. The unit could extract range from any point -- pretty much on demand. The rest is a software problem -- easily solved. |
|
| |
Your links are interesting, [zigness], thanks. But as you noted none of them are really very similar to this concept. Basically all those are measuring a distance where you can place the tool on at least one point of the measured item. This is not true with my idea. My idea lets the user measure or check the distance between two remote points. |
|
| |
[GC],I think you might have missed my intent on my last anno... check back later... I have a much more detailed thought on a way to build this... just no time to type right now. |
|
| |
Oh, my response was to the links, not the anno. |
|
| |
A sextant of sorts would do a great job at this too. You just need to carry a calculator or a look-up table for the trig. |
|
| |
Somewhere, up there, was a comment about using a prism to make a line from a laser. As far as I know, a diffraction grating is used for that purpose. Sorry, doesn't help with the idea, which is already excellent. |
|
| |
A shape that's easy to "eyeball" (such as a perfect circle) could be formed in real-time as the device is moved along the wall, so you can see that it's found the angle correctly, as a quick way to double-check it, so you know that the "range finder" system has accurately identified the angle. The laser device should be easy to mount to a fixed object, and it should have special signals so you know when accuracy is reduced. |
|
| |
If you've got the range finders, don't limit the system to one flat wall. You could measure across corners, too. |
|
| |
//using a prism to make a line from a laser. As far as I know, a diffraction grating is used for that purpose//
I thought you just used a simple cylindrical lens. |
|
| |
Actually, I'm willing to be corrected as I'm quite interested in using an application that requires a green laser to be 'split' into a line. I can't seem to find a solid reference. If I used a cylindrical lens, would it need to have a very short focal length? I reason that since the laser beam is very narrow, the cylindrical lens would look almost flat over the width of it. |
|
| |
[Ling] It is a very long time since I looked, but back in the 80's I looked at the alignment laser (He-Ne back in those days) on a CAT machine I was working on, and all it appeared to have was a single rod.
Like you, I'm happy to be corrected. |
|
| |