Page 1 of 1

A bass-ackwards approach with a forward result

Posted: Sat Dec 24, 2016 9:22 pm
by fred ungewitter
IMG_3241.JPG (1017.73 KiB) Viewed 407 times
Will you look at that burn test? It's hard to believe that it came from my machine, but it's true, and on Christmas Eve, no less!

The upper right is of only minor consequence, as I believe it's related to the warped machine and if it isn't, I'll work around it, mostly.

Here's the news flash. Despite Mr. Russ' wonderful laser pointer brackets/clamps for mounting a laser pointer on the laser tube end, I could not avail myself of his great research, design and work. The end of my tube is not a cleanly machined surface, but a bubbly curved piece of melted glass.

Deep inside my alleged mind, a new concept appeared, what happens if I shoot the laser pointer in the reverse direction?

It was a darn sight easier to work backwards. The inside diameter of my nozzle tube is about 15 mm, give or take. The outside diameter of my cylindrical presentation laser pointer is 11 mm, more or less. Fire up the old three-d printer, known in the county makerspace as the threedy printer, and I have a cylinder nearly the right diameter to hold the laser pointer snugly inside the nozzle tube, also snugly. So darn snugly that when removed and replaced multiple times, the dot (more of a slash) appears in the center of the target each time.

The threedy printed part took three or four iterations, as my printer is not precisely 1.000 to 1.000 and I'm too lazy to calibrate it. Fifty millimeters length of the tube provided appropriate insertion at the nozzle tube. The laser pointer actuation button engages the inside of the tube and makes the pointer run throughout the test period. Fresh batteries also helped after the light died mid-testing.

With the dot in the target above the nozzle centered on the crosshairs, it was replaced with the mirror (#3) and the target was placed into mirror #2 socket. Following Russ' guidelines, the objective is to get the dot at full extension to match the location of the dot at minimal extension. With such ease of targeting provided by a constant laser dot, I added the objective of getting the dot in the center of the crosshairs as well.

In the case of my machine, and I would hope this happens to no one else, I discovered that the assembler or alignment person must have used clamping pliers (vise grips) to put things into initial aligment, as the slots holding the mirror bracket were bent out of shape and partly chewed up inside. For my machine, it was also not even close to being perpendicular as in Russ' recent video.

Lucky for me, I was able to carefully slide the mirror bracket on the 45 degree slots to the correct location, after what felt like ten tries and two breaks during the day, then slide the y-slots the appropriate amount. This resulted in the laser pointer dot nearly spot-on, for both max and min extension locations.

Now to mirror number one. Paper target inside the bracket showed me that I wasn't even close and had to return to the 45 degree and y-slots, but finally got it in the ballpark. Even though the laser dot was in about the 2 o'clock position and 4-5 mm off center, it didn't stray when min and max extensions were tested.

I have a threedy printed sleeve to keep the number one mirror in place vertically when the lock screws are released in the mount on the machine deck. Think candle holder where the candle wants to drop deeper and the candle has a t-shaped top. The sleeve made it easy to release the screws, crank up the sleeve threads and get the dot into the 3 o'clock position. The base screws also have slots cut, enabling me to get the dot into the center of the crosshairs.

Somehow I got distracted and moved one step too far. That step is to remove the paper target and place tape or similar target material over the end of the laser tube. That would have enabled me to adjust the screws on mirror number one to put the dot in the center of the laser tube.

Just think! If you have a laser beam running from the center of the nozzle to the center of the laser tube, when you reverse it, it's going to be fairly close to on the mark.

Because I forgot the above step, I moved onto the target strip shown in the photo above.

I fired a quick pulse and there was nothing. Another quicky and still nothing. Okay, press it a little longer. Yikes. Sharp burning on my upper arm, just above the elbow. Good thing that was a short pulse. I was able to bend low and look into mirror number one from the front of the machine and see the reflection of the bore of the laser tube. Sheesh. I am ALWAYS wearing my laser safety goggles when the laser is hot, but the cover is not always closed.

In short order, the candlestick mount was twisted and tuned into position for a clean burn on a paper target in mirror number two. I knew that mirror two and three were properly aligned and I would not get another shot in the arm. The target strip was put into place and all the shots were a smidgen too far to the left. A couple notches on the left/right screw on mirror number three and the results are in the photo above.

I suppose I should have taken photos during the process, but I made so many mistakes and had so many false starts, reverses and repeats that it wasn't foremost in my alleged mind. At one point, I nearly re-manufactured the mount for mirror number one before I realized I could adjust mirror number two to get things under control.

Because of the construction of this machine, a redsail clone, it's possible others' machines would not have the same ease of insertion of a pointer, but it is likely to be easier to create an appropriate sleeve for just about any machine.

Re: A bass-ackwards approach with a forward result

Posted: Sun Dec 25, 2016 12:49 pm
by Michael Dunning

I haven't been too happy with my recent (re)alignment, since having to move our laser to our new studio. Since we have identical machines, I'd like to volunteer as a guinea pig to replicate your method. Can you post a picture of your "threedy" lens holder? I think I follow the rest of what you described. :D

Re: A bass-ackwards approach with a forward result

Posted: Sun Dec 25, 2016 2:30 pm
by fred ungewitter
IMG_3243.JPG (901 KiB) Viewed 280 times
I hope you don't find the sophistication of the design to be overwhelming. Russ' designs are marvels of engineering and I'm amazed that his brain cells can dance so well. Mine aren't so talented, so we get cylinders as a result.

The short cylinder was the last test to determine that the print fit the laser pointer snugly and fit the inside of the nozzle tube snugly. Snugly enough that I wondered if the cylinder would snap at the layer lines. No snap and a good solid fit.

At first I wondered if I was on the wrong track by figuring the laser alignment for the pointer was centered and axial to the center of the outer housing. It doesn't matter, as long as it doesn't shift. It doesn't matter, because the initial alignment is based on the paper target, shown for scale (20mm) and placed on the cross hairs.

The red button on the pointer is depressed, but now that the holiday is over, it is cheering up a bit. Oops, I mean it's depressed by the cylinder, keeping the laser illuminated throughout the process. Getting the fit snug means also that the carriage and gantry movement does not shift the dot. I used also the securing screw that is part of the nozzle tube to further snug the cylinder in place.

It's amazing to compare the amount of labor involved in firing the laser in the forward direction and checking and changing the paper targets to the amount of labor not needed when you can twist an adjuster and watch the dot travel properly.

Because mirror two mount was damaged during assembly, the work load was higher than if it had been properly installed and aligned. Even an incorrect alignment without damage would have reduced my workload. I realize now that had I been unsuccessful in getting that corner aligned, I could have machined a replacement from aluminum stock and assured a clean surface. The existing bracket was chewed up inside one of the slots as well as chewed up on the top and/or bottom of the slot area. It's a bear to carefully slide something when the surface is akin to cobblestone paving stones with sharp edges.

I hope you'll learn in advance from my screw-up with mirror one. I was so thrilled to have the dot perfectly on that target that I moved on to burning my arm. I suppose if it had hit clothing, there could have been smoke, fire and even more excitement.

A strange aspect of the tuning is that the dot from the pointer bounced from mirror three to mirror two, starting in the center and ending in the center, but sort of orbited the primary point as the carriage traveled from far right to far left. I think the orbit diameter was less than a millimeter and a half and not anything to worry about, but peculiar nonetheless.

Even though I am somewhat verbose, I summarized the sequence quite a bit. If you have a segment in question, ask away and I'll expound until the cows come home.

Re: A bass-ackwards approach with a forward result

Posted: Sun Dec 25, 2016 4:21 pm
by Michael Dunning
I'll give this a go when I get back home Tuesday. I put some folded cardboard cut to the proper length inside my mirror number one holder to hold height, so I like your collet idea to hold that mirror up during pointing. Definitely something I want to redesign at some point. (Preferably with adjustment screws. Thumb wheels?)

I also got "bit" by my machine for the first time last Sunday during a low power engraving, so I feel your pain. Literally!

Re: A bass-ackwards approach with a forward result

Posted: Sun Dec 25, 2016 6:32 pm
by fred ungewitter
IMG_3244.JPG (910.46 KiB) Viewed 273 times
The collet I created is a simple cylinder again, with internal threads on the primary and external on the secondary. The inner diameter of the primary matches the outer diameter of the mirror mount stem with sufficient spacing for the threaded portion, while the inner diameter of the secondary is the same as the stem for the entire length.

You may also note the quick release mirror plugs. Having had so much trouble with alignment, I wanted a faster way to move between mirror and paper target. By having the paper target in place of the mirror, the front surface of the target is in the exact plane as the mirror surface. Using any other location causes parallax error if the mirror mount is at an angle, adding just a tiny bit of uncertainty or misalignment to the entire project.

This collet came in handy with the most recent alignment, as I had to raise the mount about two or three millimeters. The laser tube also had to be kicked up three to four complete turns on the locking nuts of the mounts. Now that I've typed this, I have reason to believe that I did not lock the bolts holding those mounts!

Re: A bass-ackwards approach with a forward result

Posted: Tue Jan 10, 2017 6:54 pm
by Michael Dunning

Have you watched the Episode 90 "Christmas Alignment Special" yet? I wonder if the offset towards the right side of the machine can be accounted for using your method.

Anyway, when you give it a watch, let me know how you think the two methods compare. I can see advantages to both.

Re: A bass-ackwards approach with a forward result

Posted: Tue Jan 10, 2017 7:23 pm
by fred ungewitter
I did see the video, although it's hard to take Russ seriously when he puts on a non-traditional disguise! (grin)

As I watched it, I was thinking at the time that the reverse method I used required none of the special jigs and allowed me to adjust for proper alignment in real time visually. Even after seeing it a second time, I can't say that the version 90 video would have helped me to locate the damaged mirror mount.

One can use the laser-pointer-in-the-nozzle method to ensure a square reference (I believe) from mirror three to mirror two to mirror one, all with a pretty red dot. As an example, if the dot moves on a paper target in place of mirror two, when moving the nozzle from one side to the other, and you can't adjust the nozzle spacing from the gantry (as I cannot), the dot on the target allows one to adjust the moveable base of mirror two. Of course, if the base of mirror two isn't adjustable, all bets are off.

Obviously there are compromises based on the laser machine design and implementation of mirror mounts, brackets, etc. My method came from the frustration of not being able to put a red dot laser on the end of the laser tube, compounded by the screwed up mount for mirror two. I might not have been as bent out of shape as the mount for mirror two if the mount for mirror two was not bent out of shape, but at least those two things are back in shape and no longer out of whack. I can't say everything is in whack yet, though, as the laser tube is dead and I'm dragging my feet on getting a replacement.

Re: A bass-ackwards approach with a forward result

Posted: Wed Jan 11, 2017 12:29 pm
by Gene Uselman
Hi Fred- I have been thinking about your method and decided to contact Russ and see what he thought- he answered and kindly allowed me to post his email to me:

Hi Gene
I have investigated the merits of the backward method very early in my beam setting frustrations (just like Fred) As he says, there is a good location point at the bottom of the nozzle/extension tube. The probel I found was that although it was great for going back through mirrors 3,2 and 1, there was no real way to pick up the axis of the laser tube itself. Ideally you would be trying align your tube to reflect your red dot off the fully reflecting mirror at the anode end of the tube and that way you would be assured of end to end perfect alignment.I gave up and tried several other methods over time.. I fully understand the principle of beam alignment but the Chinese have made it so difficult by using a kit of standard parts the are fully flexible and designed to suit any machine. This became so apparent when I messed with the settings on my Lightblade machine. That is a red sail clone type machine that uses this "standard mirror kit" see ... SwfcVUJBlw

Even understanding the principles of what I was trying to achieve it was an absolute nightmare trying to nail something down as a reference to work from. My Xmas beam setting special was an attempt to use setting jigs to remove many of the variables and it worked reasonably well. Fred pointed to the fact that he could not use the laser pointer because of an unsuitable reference face on his tube. I agree that is a problem for some folks BUT for a "from scratch" setup there is nothing (as yet) to replace the absolute accuracy of the scorch mark method. working from mirror 1 to 3. That's one of the reasons I designed the little target holders that sit on the mirrors, it does allow quick and pain-free scorch tests. As I found out with my Lightblade set up there are adjustments all over the place except the places that really matter.
That Xmas special was very much an interim "fix" it does not solve the fundamental problems of beam setting. I have been chasing partial solutions for almost 2 years and have finally decided to stop digging as I'm never going to climb out of this hole with someone else's half cocked solution. If I was designing this machine I would want beam setting to be a simple logical set of steps that, even at the factory setup stage, was a simple non-fiddly process. Even as I am typing this response I am 80% of the way through designing something that people can make for themselves that will remove all these unnecessary variables that the "standard kit" forces upon us. I also plan to solve Fred's reference beam fixing problem at the same time. I am currently waiting on new mirror mounts from China so it may be a few weeks before I upset Thinklaser by (in their opinion) unnecessarily re-engineering the machine. In some ways they are right, once the machine is set there is no need to keep fiddling with it and for their customers, if things are not right they call in a service tech to fiddle with it and set it back to correct. It's a RedSail clone and there are lots of non-Thinklaser users that would like an alternative solution if it's out there somewhere.
Watch this space

Best regards

Re: A bass-ackwards approach with a forward result

Posted: Wed Jan 11, 2017 4:44 pm
by fred ungewitter
Thanks for pushing that over to Russ. I certainly recognize his points, as well as that he recognizes mine. Perhaps I simply got lucky, but alternatively, I had already attempted the scorched target method which likely set the laser tube in a good position to the first mirror. Of course, I also leveled the laser tube to match the level rails during that process. I can't say that I recall Russ every referencing that the tube needs to be level and raised or lowered appropriately to match the target on mirror one. By reversing the laser pointer to a 3-2-1 firing direction, keeping the dot centered in all targets, I had a suitable aiming point for my leveled tube had it not been pointed appropriately.

If ever I purchase a replacement tube and receive it, I'll know pretty quickly how well this works out, as I should have to only level the tube and lift it uniformly to the center of the target mirror vertically. If it's off by left-to-right, it's back to wrestling with the mirrors a bit, as my tube brackets have no left-right adjustment (relative to the mirror), but I suspect I won't have that sort of trouble.

I like how Russ has put out a bit of a teaser. He's darn ingenious when it comes to engineering. His stuff ranges from simple shapes to near-rocket-science modeling and almost always is aimed to be cut by his laser.

What could it be that would "upset Thinklaser?" That should be exciting to see too!