Although I am still able to do what I want to do using my K40, the M2 Nano board, K40 Whisperer and Inkscape I am coming to the time when I would want to move to Lightburn. As a result, I will need to change the motherboard to one that supports Lightburn. I am presuming it does not matter which new motherboard is chosen as long as it is reliable as all the functionality I require will come from Lightburn – right? So, I am looking for a high reliable board at the lowest cost. The M3 Nano would be the best buy as from what I have read, in principle, it has PWM capability but at the moment you do not know how to invoke and control the PWM functionality. It seems surprising why the hardware developer has added this functionality but not bundled the software functionality in with the hardware. If you were to learn how to invoke the PMW functionality, I note that @Scorch would probably provide the functionality into K40 Whisperer but there is no timeline in being able to achieve this – right? So, if I want to move more quickly my only option would be to choose another Lightburn compatible board. In this regard, there may be some who can provide their own experience of the boards they are using with Lightburn so that I might make an informed choice. Ideally I am looking for a simple plug in replacement without any hassles and one that can be shipped to the UK.
Have you tried Meerk40t? It won’t give you control of the M3 Nano’s hardware PWM (@Tatarize has written here about attempts to work out the PWM protocol without luck so far) but it does have a software implementation that might give you what you want: Tech: Raster Pulse Modulation PPI · meerk40t/meerk40t Wiki · GitHub
It also has the ability to let you use Lightburn with your M2 Nano board by acting as if it is a Ruida DSP controller on the network. 
Perhaps try one or the other or both of these to see whether they meet your needs?
The best controller is with a dsp type controller… They are expensive compared to most of these other controllers. Maybe more than you wish to spend… the 6442g is about $400… it works best with Lightburn, compared to other dsp controllers.
To get the best out of a laser, I think you need the best control over how it works with existing hardware… that would be with something like a Ruida.
The dsp uses a pair of signals to control the laser, whereas the others use a single pwm control line that doesn’t control the lasers current. It is manually controlled, so you can’t change the current or laser output power, via software.
Good luck
Will do… thanks…
I looked at your blog for a procedure to set the pot on a k40… didn’t seem to find it… Found plenty of ways to change or fix the existing one…
Set the pot in a position and then do a test fire while watching the mili-amp meter. This is the maximum current setting.
I suggest putting a voltmeter on the pot. It gives you a numerical value for the pot position.
Also if you characterize your tube [potVolts vs. current] you will have a map of its current at a particular pot [IN] voltage. This characteristic curve also provides a foundation for determining tube deterioration.
Yes this was one of the first things you said when I was setting up my K40 so I fitted a voltmeter and determined the graph when I fitted the new tube. I found that it is 3.8v for 15mA and that, as the tube deteriorates over time, I will need to increase the voltage to continue to achieve 15mA - right? After test cutting a lot of materials at varying thickness over the last few weeks I am happy to consider 15mA to be the maximum power in order to prolong tube life.
I didn’t want to prompt in my post but I have had my eye on Mini Gerbil v3 which is half the price of the Cohesion 3D. It seems to get a good write up so if it provides the same same Lightburn functionality maybe this is the one.
I am not familiar with these controllers so I will take a look. This is a long term commitment so even though these boards are expensive if they do the best job and provide a high degree of control particularly fine engraving then I would not want to dismiss. If this were the case would you consider other changes such as steppers, mirrors and lenses as I current have stock items?
Yes I have been trying to track the technical detail of the recent M3 Nano posts hence trying to determine the current understanding so what you suggest looks like it might be an interim step to using Lightburn on the M2 Nano which I had not understood so thanks for that.
Of course, you can run power levels below 15ma programmatically from the software depending on the job characteristics.
Yes, you will need to adjust the max [pot] as the tube wears.
I have one in the wings to use with my Laser diode machine if I ever finish the build.
You have posted a lot about raster engraving, so there’s a detail you might care about.
The Cohesion 3D and Gerbil are both gcode controllers. Both are, to the best of my knowledge, fine products. However, be aware that gcode for engraving is verbose and so takes additional processing power to parse. The Ruida controllers, as well as the M2 and M3 Nano boards, use custom protocols that use less processing power for engraving.
Moving to a gcode-based controller might actually make things worse for you, at least on the low CPU performance end.
So it’s definitely worth trying out the less expensive option first.
And then @jkwilborn has a lot of experience that I would not lightly overlook. If you want high engraving performance, it might be a case of pay once, cry once?
For cutting and line engraving this just doesn’t matter, as far as I can tell.
There are lots of decisions you can make…
The first is probably to do the math and see what is physically capable before you invest in something more precise than can actually be achieved.
As you’ve mentioned, if the best dot size a laser can make is 0.10mm, then you are looking at a lower interval and restriction of you can accomplish. Shorter lens, smaller dot, shorter depth of field. Can’t expect 5,000 dpi or even a 1,000 dpi, with a beam that’s 0.10mm… only 254 dpi, best case.
The other is that most dc excited machines have a limited response time and the fact that these won’t lase down to 1%. It seems the larger a tube gets, the higher the minimum power is. I looked at a 60W RECI and the specification advised a power range of 12 to 60W… That relates to 20% for low power. Don’t expect the low end <20% to be a readily available… Although mine will lase down to about 9.5% and it’s a 40W tube.
If a project needed only 5 watts to do the job, then 12 watts would be over twice as much power… You can increase speed, but that has physical limits also…
Big isn’t always better, especially with these lasers. High power has pretty much one use, cut through thicker material… which is just a real deep engraving 
I use a compound lens to get a dot down to about 0.05mm, close to and exceeding a lot of dpssl lasers.
Might want to see what you can do to get a better dot for best engraving.
Here’s a couple of video by a wizard in lasers and a major help to many of us, Russ Sadler… Might be a good show to watch.
First is a preview of what he’s up to…
We are discussing how the interface on the lps works with a grbl board and a Ruida or dsp type… Please join in if you haven’t…
Good luck
You’re absolutely right in that I moved to a K40 for its cutting capability and I can do everything I need to create designs with the existing setup but perhaps made the mistake of thinking engraving would also be enhanced and as I have certainly see some great results produced by the K40 I know I will get there in the end so it is indeed worth trying the less expensive option first. Many thanks for your advice.
I just want to be clear that I’m not experienced, I am just summarizing what I’ve learned from reading lots and lots of posts while administering this forum. Definitely trust folks like @jkwilborn, @donkjr, and others with lots and lots of practical experience if they tell you anything that conflicts with what I say.
Yes I have looked at quite a few videos on the subject and some of those in the excellent series from Russ so thanks for the recommendations. It was from one of his that I discovered the active ingredient for tile engraving (TiO2). Not sure I am up to the intellectual challenge of the ins and outs of Ruida vs grbl but I will keep and eye on how the debate unfolds.
These in’s and outs that you mention is very close to how to chose a good controller.
This isn’t on the differences between controller, in all actuality.
It’s about how the two control signals work on the lps. The one that controls current is removed and done manually for the K40/grbl variants. Without this being under computer control, you have lost automated current control. All you have left is period.
If you want a good controller, I think you need to know how well it controls the tube operation…
IMHO, the simple difference in the two controllers when burning a line at 50% power is do I want to run 100% power for 50% of the time or 50% power for 100% of the time it’s drawing the line…
I’m suspicious that this is the reason K40s seem to go through tubes more quickly …
Trying to get Don involved… I highly respect his knowledge and abilities… not to mention he’s been pretty intimate with the lps hardware.
But, but Oz and Ray worked together and now there’s a version of Smothieware firmware on the Cohesion3D site and a switch in Lightburn which speeds up engraving. It’s a clustering option which combines lots of the same moves into a single command.
One thing to consider is that the K40 machines are FAR FAR more often shipped with under rated tubes so while you can possibly eek out 40W it takes greatly over driving it and that decreases lifespan of the tube. So for the first tube, it’s best to pick about 80% of full power as the 100% position and the tube lasts a good 5 years or so depending on hours used. Get a new grade A tube and it’s a whole new ball game and K40 owners are often surprised at what a real 100%=40W machine is like.