This is my first post here, I’ve had a quick search but apologies if i’ve missed a specific thread that already exists for this!
I’m looking at buying a K40 Laser from ebay (Probably an OMtech Clone) and after having done some research it appears that Lightburn seems the best option for software with these laser cutters. I do know however that the majority of these come with an M2 board which is not compatible with lightburn.
I’ve seen people talking about the GRBL and smoothieware controllers which seem suitable but i’m wondering if there are other options and/or if anyone has had success with a clone or an “aliexpress special” control board?
I’m looking to order everything at once so that I have the Machine, new control board and air nozzle mods/water chillers etc all set up for first use.
I believe that @Almurray42 was specifically looking for options other than Smoothware or GRBL based boards. That would eliminate Cohesion 3D.
I guess I have the fundamental question of why you’d want to do this as these are some of the best supported options and easiest to integrate. You’ll have more footwork to do. Which is fine if you’re up for it but whether or not it’s worth it will depend on your motivation for doing so.
Really this is going to leave you with a few options:
DSP solution like Ruida/Trocen/Top Wisdom
Marlin based solution with some of the multi-purpose boards targeted at 3D printing.
Suppose you could do a PC based solution with LinuxCNC.
Anyhow, if you can elaborate on your design goals that would help narrow the discussion.
Thank you both. I’m not particularly looking to avoid these boards, but as cost is a factor, I wondered if anyone has had success with cheaper control boards on the market, or if it is worth the investment in getting a genuine Gerbil Board.
Although I’d prefer to save some cost here, I’d much rather have a stable reliable board that has a good set of features.
There are number of boards that take GRBL firmware or variants that can be had for relatively cheap. There may be some work in getting the right components to integrate with a K40 but totally doable.
Makerbase MKS DLC32 is one mentioned here fairly frequently.
There are a bunch of ESP32 GRBL shield options. You could even do an old arduino 8-bit option but honestly at this stage there’s really no reason not to go 32-bit for performance reasons.
As far as I can tell the only advantage a 32 bit machine usually use a 16 bit register for the pwm generation. Most of the Arduino use an 8 bit register.
I know the 328p (Arduino) has at least one 16 bit pwm register. It must be used elsewhere or it probably would have been implemented with the Arduino based machine.
The primary advantage is not about size. It’s about performance. The Atmel 8-bit chips used on the Arduinos are way slower than alternative 32-bit offerings. This limits controller calculation performance and thus how fast engraving can happen.
I think this is the primary reason why most of the laser manufacturers have moved to 32-bit solutions.
Considering the controller is ‘waiting’ or busy elsewhere for the machine to complete it’s current operation I can’t think you could see any possible speed gain.
Comparing, even these tiny computers to the mechanical operation of the laser is not really possible. Any mechanical movement or operation is going to take much more time than even the smallest micro controller will take to control the requested operation.
Since the micro controls the stepper motors, I can see that it runs both Y and X axes when it homes, after raising the Z axis. It controls the stepper for both axes and you can’t visually see it.
Is the machine going to run faster if it’s connected to a super computer…? I highly doubt it.
The only other observation is that some will take a 4 Gb SD card. This advantage, according to them, is that it allows them to ‘receive and store’ more of the gcode. The machine speed isn’t going to change.
You may be speaking theoretically but the limits of the 8-bit arduino platform has run its course in the 3D printing world and in the laser world as well largely because of performance considerations. This is especially true for more computationally intensive operations like arc moves. This probably won’t be a limiting factor for most diode cutting operations. But for things like engraving and traversal moves this could very much be a limiting factor.
I’ve seen a number of discussions on speed considerations for this.
Take a look at this excerpt from the NEJE wiki. They acknowledge the speed advantages of the 32-bit platforms:
About STM32 and atmel chips？(2020.10.31)
Regarding the 32-bit stm32 and atmel chips, atmel is the original grbl dedicated chip. The stm32 chip can increase the main frequency and can indeed increase the movement speed of the machine, but there is no increase in cutting speed and engraving speed! why?
Compared with atmel, the actual increase in cutting speed and engraving speed is not obvious. , Because the cutting speed is determined by the laser power, the engraving speed is determined by the weight of the module and the sliding module. The performance of atmel is completely sufficient!
So, don’t worry about your atmel performance is not strong enough.
Later we will upgrade this chip on all master machines. The chip performance is 50% stronger than stm32. The improvement that can be brought to the current machine is the increase in motion speed. Motherboard interface is compatible with existing machines.
I am in the work of upgrading my controller ( among all other upgrades)
I am using the Mks DLC32
ESP32 MCU
Grbl FW
Successfully tested out
wIFI connection with LB
All the stepper motor homing and dry movements
Will need to re-align my gantry
And connect the main board to the K49 Laser’s PWS
And test fire when I have time.
