Mirror 2 to Mirror 3 isn’t consistent along the entire Y axis. When Y is at 0 and at the maximum the lasers aren’t directly aligned.
How do I fix this?
From the position of the laser head in the photo, this looks like a classic Fourth Corner problem.
If it’s already aligned so that:
- The beam from the tube is hitting Mirror 1 dead center and
- Mirror 2 has a single spot at either end of the Y axis travel and
- The spot is dead center on Mirror 3 with the laser head in the front left corner
Then put the laser head in the front right corner and adjust Mirror 1 very very very very slightly to move the spot to the center of Mirror 3.
If it’s not already perfectly aligned, then make it so before trying the Fourth Corner tweak.
An earlier discussion covers the process:
That isn’t at all necessary.
You need to hit the 1st mirror at the height where the beam should enter the laser head. That’s not necessarily the center.
This is very important.
This is not true either. It needs to be hit at the point where it would then reflects at a 90 degree downwards, centered on the lens. This isn’t per definition the center of the opening.
Trying to center it, rather than trying to get overlapping dots, won’t get you a good alignment.
Your “classic fourth corner problem” also isn’t solved by first getting a perfect alignment from 1 to 2, and then somehow messing with 1 afterwards again.
It’s much more likely that the rails aren’t perfectly coplanar and the head is thus dropping out of alignment. No amount of 1st mirror fiddling will fix it, other than average out the error over the length.
This technique of putting tape on, burn and check is unnecessary and highly confusing. I agree that going back and changing mirror 1 just adds to the confusion.
I switched over to the reverse alignment process and would never go back. I can align my machine in less then 5 min including finding wrenches and adjusting screws.
Also what’s the objective of a CO2 alignment? For the beam to pass thru the center of the nozzle over the entire bed. That’s all that matters.
Please confirm, are you showing us spots at M3 taken at Y-max,X-max and then Y-min, X-max? Is the upper spot from Y-max, X-max?
Could be an uneven support frame. Try shimming/raising just the front right (Y-max, X-max) corner up by 2-3mm and then start the mirror alignment again.
Yes, (X-max|Y-min) and (X-max|Y-Max)
I tried shimming the right Y axis and it seemed to have no effect on the dot at all
can you give more information on the reverse alignment process/tools? I’ve been using the tape burning method and it’s very tedious
I’m just trying to understand here. All the 3 below are good
The beam from the tube is hitting Mirror 1 dead center and
Mirror 2 has a single spot at either end of the Y axis travel and
The spot is dead center on Mirror 3 with the laser head in the front left corner
With this being said, the 4th corner burn is down and to the right. Which lead screw should I adjust on mirror 1 to try and edge it into the middle of Mirror 3 at the 4th corner?
It’s usually intuitive for me but adjusting mirror 1 for mirror 3 adds a little bit of confusion into the mix and which way the beam will travel after hitting mirror 2.
Standing behind the machine and looking forward, adjust Mirror 1 to move its top back (toward you) and its right side forward (away from you). Do that in separate steps, checking after each adjustment, because I have been known to be wrong in public. 
The total adjustment will be on the order of 1/16 turn of the screws, so this is not the place for dramatic motions. Tighten the locknut after each tiny adjustment, because taking the slack out of the screw threads will change the results.
When you think you’re done, verify the alignment all over again and show us pictures of the scorches at each mirror so we can check your work.
What is the overall working size of the machine?
The support frame under your machine looks unusual, was it modified? how is it connected to the machine? Try shimming the front right corner up even more, by the way I meant at the base of the machine not directly under the Y-rail. Does the back right of the machine lift? Keep testing while gradually shimming up the front right corner, but stop if the front left or back right starts to lift.
Or, if you have a laser level, like a self levelling oscillating one a builder would use to throw a 360deg line out on a wall, mount it in the middle of the machine’s working area, and then level the machine so that the laser marks the front and back of each Y-rail in the same place, it should also mark the nozzle at the same place in all corners. The purpose of this is not to make the machine level (though a useful bonus) but (given the issue you are facing) it’s just a means to ensure each end of each Y-rail are in the same plane.
To align mirrors in reverse order the only tool you need is a laser pointer that replaces your lens tube. I found a shot gun shell (mine is 20mm) that fits perfectly but anything will work that positions the beam perpendicular to the bed and in the center of the lens tube and straight up to mirror 3. The pointer must generate a very small beam. I block mine with a penny that has a very small hole. You could also print something if you have access to a printer. I think a company makes a unit just for this task.
Let’s pause here and see if you can find something first. The process is simple but you need this laser pointer that basically simulates your actual beam going towards the work material
By the way you won’t need to power your laser up until after this alignment process is over.
The reverse tool does absolutely nothing that you couldn’t do without it.
And just about the only useful thing it does is show you where you need to hit the head.
You still need to do the whole back & forth with the axis, marking on tape, and adjusting accordingly. No different from the traditional process. It’s no magic bullet.
Besides, that’s not going to do much if your frame is out of whack to begin with.
I’ve never used one but they certainly do look useful - especially with regards to M3 as discussed in some very interesting videos online by Steve Walters at American Photonics, it’s cool that you made your own?, can you show a pic?
So in my case the lens tube is 20mm. If you pull up 12 gauge laser boresighter in Amazon the diameter is slightly over 20mm. But they are made from brass and with patience you can sand/file them to fit. Again that’s how I did it but there are smaller boresighters and other laser pointers. My method would be to 3d print an adapter for other diameter laser tubes.
You could even lay a laser pointer on the bed straight up into the lens assembly with no lens, but you wouldn’t get it centered and can’t easily move it around.
These boresighters, and most laser pointers are strong so you can see them at a distance. You want weak and focused here. That’s why I blocked the output of my laser with a disk & small hole.
Finally there are videos on this but just like this thread there is an issue with adjustment of mirror 1. Let’s see here if one of you prepares the tool.
So moors7 it’s interesting how different the English language is from the US to the UK. If one of us talked about “hitting the head” we would be saying to go to the bathroom. So it doesn’t mean the same thing obviously.
I think this method is quite an advantage over the burn and tape method. If you own a CO2 laser set it up and prove me wrong.
Yes it won’t fix a bent frame but I think it would give you a clearer idea of what’s going on.
Thankfully I’m neither from the UK nor the US then. Because where I’m from, when we’re talking about lasers and aligning them, referring to “hitting the head” means “hitting the head”. As in, the laser head, one of the primary components you need to care about when aligning. If that instead makes you think about taking the piss, then so be it…
If you use the tool, you still have to go back and forth with your axis, marking them at either end. Whether you use a sharpie to mark that, or the way more accurate laser beam itself, you still have to go back and forth.
I’ve designed and built my own 24x48” 80W CO2, thank you very much. That means you have to start alignment from scratch, no guarantees that your frame is co-planar, nor that the mirrors, head, or tube are in the right place… Fun though, gives you some more experience on the matter as opposed to sticking something up a ready made machine 
I’m not saying the tool is useless. I am saying you can definitely do without, and the only thing it is good for is showing where to hit the head (the thing on your laser, that hold the lens and last mirror).
The problem with these is you have to assume that where you put the emitter is 3 dimensionally square with the actual laser axes of the tube. This is not possible in a real world environment.
It’s much like adjusting sights of a rifle from the target end, it will get you in the ballpark but you likely won’t have as good of an alignment if you start from the tube end, which is where it all begins.
It might be nice for a setup with all the mirrors out of whack, but it still depends on the head and tube having the proper orientation.
I struggled with mine for a while, getting pissed off and grumpy, but once I figured out how it works in the three dimensions, I can check alignment of my tube in a couple of minutes. Even with a complete change of mirrors, it’s less than 15 minutes.
Once aligned, it shouldn’t need any help, assuming the machine optics are tight. If it does, then you have something mechanically loose.
You can read (watch) many videos on how this is much more complicated than sticking an led up the hole and doing an alignment.
Russ Sadler has a number of videos, but this one is a through the lens pointer and how it isn’t correct either.
If you go to his site, there are lots of videos about alignment, lenses and other interesting items.
I don’t use one and mine works fine. You just need to understand how the beam works in 3 dimensions, which is not so simple.
@moors7 don’t know where you live, but I was hooked on sailing and as an amateur operator I’ve spoken to people all over the globe. Any of them that have boats know what/where the head is. Do you have boats where you live?
The slang word use of the word “head” came from the facilities on board a ship. But all slang terms come and go so I have know idea if that’s known by younger generations. Like “I’m down for that” is not something I can’t relate to.
A 48x96” machine is impressive. I can’t imagine getting the frame in perfect alignment, or exhaust smoke properly and not have the beam grow by the time it hits the lens. Very challenging It probably requires a lot of adjustment points and very straight components. I’m talking CO2 here.
So you have a method that works, great. You’ve been able to work the angles out, great. Others find it frustrating so that’s where this method comes in.
Alright, fine. PLEASE do explain how your method works, and is better? You say you don’t like the tape method, then please do explain how you not have to do that when sticking a laser up the head? What does the laser hack do that the tape method can’t?
Jack I’m always using my phone when on these forums and recently any YouTube videos have been blocked from opening. Some BS about bots. So it’s a pain to open YouTube videos these days.
Anyway I encourage anyone who has a process to stick with that. Since you’re advising folks all the time on this site I would say give it a try and find out for yourself.