Laser cross-hair pointer

[pete]

Some new Delta and Sears (Craftsman) drill presses have a new method of using laser cross-hair pointers to pinpoint the location where the drill bit will strike the material. The new laser pointers are mounted on the drill press column which is touted to be an improvement over the previous method of pointers mounted on the spindle. That method was ok if you had a small drill bit - but a larger drill bit or forstner bit would obscure the laser crosshair point.

I think the new laser pointer could be adapted to the ShopBot – mounted to the Z axis frame for pinpointing the X & Y home positions. Here is a link to a Woodcraft web item where the independently adjustable laser pointer is for sale – $39.95. http://www.woodcraft.com/family.aspx?familyid=5475
Is it worth a try do you think?

[richards]

My trig skills are lacking, but it seems that you'd have to adjust something every time you changed cutters (distance from router/spindle to actual end-point of the cutter) in order for the lasers to work properly. Other than that (and the fact that it is powered by a battery), it looks good to me.

[steve4460]

Hi Peter

Check out this link about the laser pointer.
http://www.talkshopbot.com/forum/messages/29/10296.html?1134216508

SV.

[dvanr]

Try pointing the laser straight down. A good location would be on the y-car towards the 0,0 position of the table.

Measure the distance from the middle of the bit to the middle of the laser spot. Using the x and y offset measured, you now know how far to move the bit to be where the spot was.

What would be even better would be short routine to do all of this for you.

I haven't written a routine yet as I am still fiddling around with a b&w camera mount to accomplish the same thing as the laser.

[gerald_d]

Dick, that is the most sensible way of using a laser that I've ever heard! (That is if you have the routine that Bill Young is going to write for us: zlaser.sbp with keyboard shortcut ZL
)

What is it with the clever Dick's all being on this side of the Atlantic?


The camera version, so that you can zero the job while you have your back to it, is going to blow their minds. You have to add pattern recognition so that it will find a pencilled cross automatically!

[stevem]

Dick, that’s ingenious. It’ll even work when the area around the bit is obscured by a dust shoe or pressure foot.

Gerald, you should be able to make this work easily with Mach.

[gerald_d]

Simplicity is the ultimate sophistication.

[bill.young]

Hey Dick,

What kind of camera are you working with? I use the Logitech 4000 Pro Eyeball cameras pointing down at the table and the Record function in the Keypad to manually digitize large paper plans...it's great for recording keypoints. My son's also using them for prototyping some laser scanning applications...it'll pick out a red laser dot really easily.

I like the Logitech because it's cheap, has decent resolution (1.3 megapixels), and doesn't have any real moving parts that dust will mess up. The downside is that it has manual focus and is USB so you have to boost the signal if you have to go very far.

Bill

[mossie_jim]

"What is it with the clever Dick's all being on this side of the Atlantic?"

Easy, the day starts earlier over there. We sleep in and "borrow" your ideas. . . and sage admonitions.

Cheers,
Jim

[gerald_d]

Bill, Dick is probably fast asleep by now (the effects of time zones, beers and barbies). He is not going to look for the laser spot with the camera - he is just going to look down straight at the job for a reference; maybe a corner or a mark.

We can live with that, Jim.

[bill.young]

Hey Gerald,

That may be his application now, but there are a lot of neat things he can do with a camera on a ShopBot once he starts playing with it. For instance if he shines the laser straight down on something but looks at the dot with the camera mounted at an angle, he can do some calculations and find the height of the dot...could make a slick Z-zeroing routine. The same if he projects a laser line onto a 3d object...he can generate a cross-section.

Bill

[gerald_d]

Bill, what sort of accuracy do you expect in that type of z-zero?

But I think we are drifting the thread and detracting from Dick's totally simple way of using a laser to reference x,y position.

[daveiannone]

Hi guys,
I use 2 Zircon Laser balls...1 mounted to the y car and one to the x car.....the alignment of the 2 doesn't change with bit depth or material thickness because they always point at exact alignment of the router center....It really works great and I wired them to a switch so they come on and off together with one flip of switch.
They were 35 bucks at COSTCO but am sure they are widely available.

Dave I.

5402
5403

[mikejohn]

This (http://www.talkshopbot.com/forum/show.cgi?tpc=7&post=19473#POST19473) thread shows the relationship with the Z axis and the table, and the relationship between the Z axis and the x and y cars.
This suggests that in some instances the true z position may not always be the crossed laser position,(where the laser is attached to the x and y cars) but would depend on material thickness, and bit length.

............Mike

[gerald_d]

Mike, you have lost me. I don't see the direct connection between that thread and this one.

The basic principle is that folk want to see the point on the job/wood/table where the center of the bit is going to plunge. All the laser systems (till yesterday) relied on two lasers each projecting a vertical plane of light. The vertical plane of light would make a line where it intersected a horizontal surface (the surface of the job/wood/table). The two lasers would be mounted so that the vertical planes of light intersected each other more or less at 90 degrees (not critical). The critical thing being that planes of light are as vertical as the router/spindle axis and the z-slide (here Mike's link has relevance).

Peter Meacham opened this thread asking if a "twin laser" would work on a SB. Of course it would. And, if properly mounted, it will be insensitive to Z-axis height, job/table/wood height, cutter length whatever. But the key is that it should be properly mounted - the laser planes must be vertical to the table.

Dick comes along and points out that we don't need a crosshair line projected under the cutter, like the drill-press folk need. We have a calibrated tool capable of precise x,y movement. We can mount a single dot laser, pointing straight down, offset (x,y) from the router/spindle axis. We can zero the spot over our target and then tell the bot "now zero the spindle/router over the spot". Very similar to using a bot with twin routers on twin Z-slides.

Stephan Voelkel has this (http://www.mechmate.com/Forum/messages/47/307.html#POST77) arrangement. One of his light planes is vertical, the other is slanted - and therefore the system is height sensitive, as he notes. He would be better off pointing straight down and then using a "ZL" command to make his bot move directly over the spot where the laser was pointing.

As Steve notes, this will avoid all light obstructions caused by dust shoes, etc.

[mikejohn]

Gerald
You are correct! (as usual!)
I had wrongly asumed (I'm getting good at making mistakes) that the lasers were set square with the x and y car.
I now see this is not true.
However, I won't apologise for my post, it's possible others might make the same wrong asumption! They can now understand fully without the embaraasment of looking a complete twit


..........Mike

[daveiannone]

I guess any of these ideas will work well...I just like the simplicity of where the crosshair is , that's where the bit center is..Also, unless I'm mistaken about Dick's idea (a good one by the way) at some point because of the x,y offset you may not be able to reach with the laser dot everywhere on the table that you can reach with the router.
Dave I.

[gerald_d]

How to "trammel" a down-pointing laser....

1. put a slightly opaque window horizontally about an inch below the laser.
2. lay a mirror flat on the table
3. the opaque window will show the laser pattern on its way down, as well as its reflection on the way back.
4. Adjust the mounting angles of the laser until only one pattern can be seen.

[dvanr]

Gerald,

Been on Barbie and uncorking duty yesterday, another hard day in paradise.

Bill,

The camera is a very small security camera from Swann , its called the Secura view and comes as a kit. Camera, small monitor, and 18 meters of wire connecting the two. I was thinking along the same lines wth the camera or laser at 45deg. I will email you the link for the DIY laser range finder. It is a uni students website, he even has VB code as an example.

I'm still not sure how practical the camera is at this point. The need for it came out of a job that I was unable to do, tight tolerances, having to cut in specific areas and with no square edges as a reference.

The Image sharpness is very good , and depth of focus is a about 2 inchs (50mm).Focus is adjustable.

With the camera sitting about 4-5 inchs off the table (vacuum dust shoe in contact with table) the magnification is about 2x , with 300 lines of resolution you can adjust very close to a pencil mark.

A felt marker was used to mark the monitor screen with a crude x-y axis and then subdivided into smaller marks.

The lens and image plane are not ideal . The lens has a slight fish eye to it , the effect is that the spoilboard looks like it is curved slightly. It feels like your googling the table with your own satellite.

In addition the image plane or CCd is not aligned properly in the camera either (yes , it is made in China)I made an adjustable platform to plumb the optical center of the monitor to the table and that worked, but I was amazed at how far to one side the camera had to be tilted.

The method to align the camera to the z-axis is .

Mark the table ,

1)Move camera to it's lowest point , position camera using x-y movement so that the table mark centres on the crosshairs of the monitor.
2)Move camera up 4-5 inchs. Adjust tilt of camera to move crosshair back over spot.
3)Repeat sfrom step 1) till the spot stays aligned on the cross hair at the top and bottom position.

The original camera platform works but needs a redesign to prevent it from turning. When I have the new one made I'll post some pics.

[hfjim]

Bill,

Can you provide some more details about how you digitize the large paper plans? I do some work with a stained glass company and they always hand me a big paper pattern that I'd like to digitize. I thought about the camera idea but wasn't sure how to implement it? Are there crosshairs in the camera, or what do you use to align things?

Thanks

[dvanr]

A camera thread has been started here (http://www.talkshopbot.com/forum/messages/27/12790.html?1145329976#POST34953)

[larry_r]

After I installed my air deflector for my PC router, I found it difficult to visually locate an exact spot on the table where I wanted to Zero the x and y axis. So a search located this thread on the forum where Dick’s above solution seemed the easiest. I bought a laser pen at Office Depot, drilled out a piece of pvc pipe to fit the pen. Glued the pipe to a thin piece of acrylic. I was fortunate to have some rare earth magnets lying around. I used a total of 4 magnets. I placed two magnets on the backside of the Z frame and placed the acrylic piece with the attached pipe and laser over the magnets and then the other two magnets over the plastic in line with the previous two magnets.

I then installed a V bit and cut a fine groove along the X-axis. I was able to move the magnets/laser combo (with some effort) to get the laser point to align with the groove below. That now gave me my Y-axis location. I next marked the location of the laser pointer with the groove below and I was then able to measure the distance from where groove started and the marked location. This measurement of 5.342 inches told me I would have to moved the X-axis by that distance to locate my starting point.

Here is my procedure:

1). I align the laser with the starting point on my table.
2). Zero X and Y-axis.
3). Jog X 5.342 inches.
4). Zero X
5). Run C2 command to zero top of work.

It has worked great, though the use of a wireless keyboard helps a lot.

Here’s where I would like some help if possible. I would like to automate steps 2 through 5 with the use of a program.

Thanks Larry

5404

[beacon14]

Save the following as Custom5.sbc in the C:\SbParts\Custom directory (you can use any number 1-9 besides 5, find one that is not already being used)

Then when you have the laser over the start point type: C5



'Move X axis 5.342" and zero X and Y

SR
JX,5.342
SA
Z2
C2

[larry_r]

Thanks Dave. I did not know it was that easy.

[odulfst]

david,
i belive you need to zero before and after the move command. I'm not very experienced, but i think JX, *** moves to that position. it doesn't just move that amount. so if you were at an x cood of 10, and said MX, 2, you would get a lateral move (to the left) of 8 inches, not 2 to the right.


maybe it already does that, seeing as i don't know what the SR command does right now.

tony.

[beacon14]

[S]et [R]elative sets the SB to Relative Mode, so any move is calculated from the previous position.

In hindsight, it would have been slightly shorter and maybe a little more obvious to write it as such:

'Move X axis 5.342" and zero X and Y

Z2
JX,5.342
ZX
C2

The net result is the same

[odulfst]

at least i knew my foot was gunna go in my mouth ;-)