Preparing to chuck the moshi board. Anything I haven't considered

[Andrew Fisher]

HI all.

Thanks for all the buildlogs, and info here I decided to take the plunge, grab one of these guys off ebay and have got things going. To date have added a lid interlock, 3d printed an air assist nozzle and got that all sorted with an aquarium pump. Can now cut 4mm acrylic slowly but perfectly fine. Had a bit of drama with an X-axis beam not cut straight so all my rectangles became parallelograms so got that all sorted. 

So now I can cut reliably and some reasonable materials, it's time to chuck that moshi board and go more open source.

For reference I want to control the board using Printrun (Pronterface) to interpret GCode (so it's the same as my 3d printer) and I'd like to go from Inkscape for design to probably CamBam to do the tool set up (set power and speed for different parts of the job etc) which will then produce the GCode as output. 

My next step though is to get a new board in place. 

After a lot of reading and assessing my budget I've opted for a RAMPS 1.4 board and Arduino Mega. I have some standard motor drivers (A4988) that I can use for the motors. I also have some new optical limit switches (I have some mechanical ones too) which I can mount.

To make sure I'm not missing anything can I confirm the following:

• RAMPS board needs the poly fuses removed and shorted
• RAMPS board needs D1 diode removed so that it doesn't try to drop 24V to 5V (does this need shorting??)
• RAMPS board can take 24V but the Arduino can only take up to 20V right so I need to get something in range 7-20V to supply it yes?
• Remove the current potentiometer that controls power as that will be controlled by the RAMPS pwm going forward
• Add the new X and Y axis end stops and bring them back to the RAMPs board
• Remove the flat ribbon cable and rewire the X-axis motor back to the stepper driver on the ramps board
• Laser PWM and Laser fire get brought to RAMPS

From a software side, do I then just run Marlin on the arduino? (Bearing in mind I'm not going direct from inkscape so don't need that part sorted).

Anything I'm missing in this next step?

Cheers

Andrew

[ajf]

Hi Andrew,

I've answered some of your questions inline below.  Dan's buildlog has a nice wiring diagram for hooking up RAMPS to the PSU, but note that there are several different variations of PSUs in the K40s.

On Friday, June 26, 2015 at 3:55:47 AM UTC-7, Andrew Fisher wrote:

HI all.

Thanks for all the buildlogs, and info here I decided to take the plunge, grab one of these guys off ebay and have got things going. To date have added a lid interlock, 3d printed an air assist nozzle and got that all sorted with an aquarium pump. Can now cut 4mm acrylic slowly but perfectly fine. Had a bit of drama with an X-axis beam not cut straight so all my rectangles became parallelograms so got that all sorted. 

So now I can cut reliably and some reasonable materials, it's time to chuck that moshi board and go more open source.

For reference I want to control the board using Printrun (Pronterface) to interpret GCode (so it's the same as my 3d printer) and I'd like to go from Inkscape for design to probably CamBam to do the tool set up (set power and speed for different parts of the job etc) which will then produce the GCode as output. 

My next step though is to get a new board in place. 

After a lot of reading and assessing my budget I've opted for a RAMPS 1.4 board and Arduino Mega. I have some standard motor drivers (A4988) that I can use for the motors. I also have some new optical limit switches (I have some mechanical ones too) which I can mount.

To make sure I'm not missing anything can I confirm the following:

• RAMPS board needs the poly fuses removed and shorted

Yes. If you're going to power the RAMPS with 24v you will need to remove the poly fuses and either short the connections or install real fuses (you can get mini-blade fuse holders/fuses.) 

• RAMPS board needs D1 diode removed so that it doesn't try to drop 24V to 5V (does this need shorting??)

• RAMPS board can take 24V but the Arduino can only take up to 20V right so I need to get something in range 7-20V to supply it yes?

 Removing the D1 diode prevents the RAMPS from powering the Arduino. DO NOT short, just remove the diode.  

The laser PSU should also have 5v which you can use to power the Arduino (see Dan's buildlog noted above.)

There are also a couple of other options:

If you don't already have an Arduino, you could get the Taurino from RepRapDiscount which which can tolerate upto 35v (I'm not quite sure what the difference is between the Power and Classic versions, other than $10, you might want to research that....)  This would mean that you could leave D1 on the RAMPS in place and not worry about the separate 5v supply.

You could also use a standalone 12v PSU or a 24v-to-12v DC-DC convertor to power the RAMPS/Arduino. Note: This can be problematic if you are using active LOW to fire the laser because, if Marlin isn't booted up before the laser PSU is online, the laser may fire.

Or you could forget about Marlin altogether and go with Smoothieware.  The boards (official, Smoothieboard and Azteeg X5 mini) are more expensive, but there are some compatible boards out there that are closer to the RAMPS/Arduino cost. (I'm not linking the cheaper boards since the vendors are not really giving back to the opensource community as far as I can tell. But, I'm sure you can find them with a little searching.) 

• Remove the current potentiometer that controls power as that will be controlled by the RAMPS pwm going forward

Yes. 

• Add the new X and Y axis end stops and bring them back to the RAMPs board

You should be able to use the existing endstops. 

• Remove the flat ribbon cable and rewire the X-axis motor back to the stepper driver on the ramps board

Again, check Dan's buildlog.  I think you should be able to breakout the motors and endstops from the ribbon, but there is some variation on how these machines are wired. Mine didn't have a ribbon, just plain old wires. 

• Laser PWM and Laser fire get brought to RAMPS

Yes. 

From a software side, do I then just run Marlin on the arduino? (Bearing in mind I'm not going direct from inkscape so don't need that part sorted).

You will need a laser compatible Marlin. Again, Dan covers that...

[Andrew Fisher]

Hi AJF - thanks for that set of answers. 

I was toying up with the smoothie board but at this point couldn't justify the price. I also had a RAMPS board spare because I got 2 for a 3d printer (always like to have a spare floating about) and also had things like motor drivers for it etc. Figured I would at least start there from a broad familiarity perspective and then upgrade over a bit of time. The smoothie boards do look very good  though (though I was a bit perplexed by reports of not being able to do 5V PWM)...

Thanks for the info re power - I was considering this morning getting a separate 12V supply for the arduino so knowing that I have to ensure the arduino is booted first does cause some concerns. Having said that, how is that being prevented in the case of it all being on one power supply? Basically just that the arduino turns on at the same time, Marlin takes a second or so load so the window is very small?

Will post back with progress or questions. I've been documenting as I've been going so should have some good notes to again expand the pool of knowledge around this for those that follow behind.

Cheers

[ajf]

Andrew,

I've become a big fan of Smoothie over the past year.  All my printers are deltas, so the extra horsepower makes quite a difference and not having to deal with the Arduino IDE is a plus too.  The price difference is significant though and, for a laser cutter, performance isn't an issue.  Yes, Smoothie-compatible boards use 3.3v logic so you need to level shift to interface with most of the laser PSUs (I think some can handle either 5v or 3.3v, but I'm not sure and mine could not.)

Regarding power: When I was running RAMPS, I had my 12v PSU connected to mains power independent of the laser, so the issue was really that if I forgot to turn on the 12v PSU before the laser, it would fire. Now I'm powering the Azteeg board from the laser PSU and that hasn't been an issue.  Suspect that it won't be an issue with RAMPS if it's getting powered from the laser PSU or, if using a separate 12v PSU which is tied to the laser's power switch. However, I would suggest keeping an eye on that during initial testing just to be safe. I've kept the laser on/off switch in place so I can manually shut off power to the laser tube.

regards,

aj

[Andrew Fisher]

That's some good feedback re smoothie board - reckon it stays on my upgrade list.

If I have that a keylock on laser tube on interlock loop that would stop it from firing yes if I run two separate power supplies? Slightly academic as I've literally just removed the polyfuses and D1.

On the arduino power side of things. I'm not planning on running it unless it's connected to a computer so presumably arduino power can actually come over USB? Now I've removed D1 that is so circuits are now separated...

[Andrew Fisher]

Another couple of questions I've just come up against:

Has anyone got a flow sensor working for the cooling and then had it link back up to the controller board so it can show current flow rate etc? Ideally with cutoff if it drops too low.

Related to that - what is generally done with temperature sensors? Thermistor and just attached to water in and out on the laser tube?

[Andrew Fisher]

Update from the field.

Did all of the motor controller work over the weekend. Ripped out the end stops that shipped with my cutter mostly because they were located in funny positions and the bits of metal they were using to trigger the endstops were so long they were knocking off about 50mm of usable space on both axes so I changed them out for some basic mechanical end stops. Replaced the ribbon cable for some decent wiring that is a bit more shielded and got all of that back into the electronics bay to the arduino and ramps board which is now powered off 24V from the PSU.

All checked with Pronterface and sure enough I have some CNC action interpreting GCode! Milestone achieved :)

Next thing to do is get the laser firing.

Question:

Laser has 2 inputs, both of which attach to PWM outputs on the RAMPS board. Am I right in assuming that these lines are *just* 5V signal lines so they don't need crazy heavy duty wiring for them?

[Andrew Fisher]

Success!! Wired everything back up again tonight and everything moves around. Wired up the laser and all of that worked a treat (forgot about 5V from laser supply to I2C though so laser fire on start - yay for not messing around without protective glasses.

Question I have now though is I'm using the LMN version of Marlin and I can't figure out how to set the PWM value of D6 so I can turn down the firing power to do some testing with.

Dos anyone know what the gcode command for that is?

Cheers

Andrew

[ajf]

Excellent! 

I actually haven't got a very good cooling setup, just a bucket of distilled water in an ice bath.  My understanding is that you basically want to keep the tube at room temperature to avoid having the sealant used from degrading and allowing the gas to escape. LightObject has recently expanded their component offerings and there are various flow sensors available, all quite inexpensive - though shipping cost may be an issue.  When I get around to upgrading my cooling, think I will use thermistors to monitor temps.

I'm not sure if Marlin supports setting intensity manually for test firing, if that's the question. You might want to post a new thread with this question.  FWIW, I think Turnkey's Marlin fork is the current goto...

regards,

aj