Omtech k40 with Monport Nano too powerful

Hey all,
I just upgraded my omtech k40 with a monport nano control board but I’m having problems with the machine being over powered. in lightburn when I run at 0.1% power, the machine is pulling about 4mW and is a lot more powerful than it was with the previous controller. And as far as I’m aware, it shouldn’t even fire at such a low setting.

I’ve verified that the $30 value and the S-value max value are aligned and that seems to have not corrected the issue. I suspect the problem lies in my other $ values such as $31 $32 or $33, but I don’t have a reference for what these should be with this setup. I’ve also heard reference to a potentiometer of some sort, but I don’t believe my machine has one to adjust.

These are my current values:

$0=10
$1=25
$2=0
$3=1
$4=0
$5=0
$6=0
$10=1
$11=0.010
$12=0.002
$13=0
$20=1
$21=1
$22=1
$23=3
$24=25.000
$25=3500.000
$26=250
$27=3.000
$30=1000
$31=0
$32=1
$33=1
$34=1
$35=0
$36=0
$37=1
$39=1
$100=160.000
$101=160.000
$102=160.000
$110=8000.000
$111=8000.000
$112=100.000
$120=500.000
$121=500.000
$122=20.000
$130=300.000
$131=200.000
$132=200.000

and this is a screenshot of my current settings in lightburn

and I’ve also attached a photo of the wiring of the control wire to the psu

Thanks for any help!

How do you measure 4mW?, do you mean mA?

Ah yes, my apologies, the analog meter on the machine is indicating about 4mA when active at 0.1% power.

Yes I thought so :wink: but nice that you have an ammeter. That percentage versus power ratio is very wrong
I’m a little surprised that you haven’t received an installation guide for your new controller on the K40. A few years ago I changed my K40 controller to a Mini Gerbil v2 from Awesome.Tech. On their site there is a fine explanation for setting up ver. 2 and ver.3. You might find some input there. You are also welcome to get a backup of my setup, but I don’t know if it can be used directly for your controller. (I no longer have my K40 but was very happy with my first CO2 laser)

If you still have it, that would be great, thank you! I’ve been playing with my $ settings and have noticed some strange behavior. Like I don’t seem to be able to modify $33 which should be the PWN dutycycle/frequency, it’s stuck at 1hz. I’ll do some more troubleshooting but probably reach out to monport to see if they can tell me what’s going on with that.

I couldn’t either, possibly part of the firmware itself.

Be careful with the values and remember they come from another controller.

K40_5_22.txt (1001 Bytes)

I guess I’m getting forgetful… I thought the nano was not supported by Lightburn… is this something new?

:smile_cat:

It’s this control board here. It’s able to connect and interact with lightburn, and aside from the power control it works perfectly. I’m new to this space but it wouldn’t surprise me if there are several boards called “nano” which can cause confusion.

The original nano board is simply an 8051 that receives all the pre-computed moves.

It isn’t like a grbl controller that has the smarts to draw a line from a to b without assistance from the pc.

As far as I know, this is the reason these boards are popular they are grbl based and don’t require a pc to drive them.


You can isolate the problem to the controller side or the lps by measuring the pwm output (IN on the lps) with a voltmeter.

At 50% power is should read 2.5V, 20% 1V…

:smile_cat:

oh good idea, I figured the duty cycle would vary, not the voltage, but sure enough it seems to be the voltage. After doing some tests I saw that I was getting 2.8 volts between in and g coming from the controller on the cable that goes to the board attached to the power supply when at 50%, then somewhere around 1.2 volts on 20%, an around .5 volts on 0.1%, I figure there could be a problem with my multi meter, or it’s not reading the correct voltage since i’m not getting those readings while the cable is actually attached to the power supply board and there’s no load. However this does tell me that it is varying the voltage in correspondence with the settings in lightburn.

I’m going to borrow this image from another forum since they took the time to take and label photos of the board that my machine has.

Right now I have the case open switch connecting p and g together. K+/- is still connected as it was from the factory, but of note is that only the K- wire is actually inserted into the connector, I think this just relates to the test fire button on the control panel, so I’m not particularly concerned with it. Finally I have G connected to the ground of my control wire harness, and IN connected to the positive, 5v is left empty on the connector I received from Monport.
I noticed while testing that if I left the G/IN/5V connector unplugged entirely while reading voltages coming from the control board, the laser would still fire, I’m unsure if this could be an indicator that there’s actually something else controlling the laser firing, or if it is just because there was a floating voltage on the in pin since nothing was connected.

When P is pulled to ground, it is considered active. You have it inverted.

This is actually the water protect input… it’s normally connected to a flow switch. When the flow switch is active it pulls P to ground.

What does this mean?

IN controls the lase current. If this is wired to 5V, that’s probably the issue.

From my information, the K+ being grounded is used for the test button…

Dose the test button from the console work?

The 5V is usually where the control board gets it power… when the IN pin is unconnected it’s usually at ground level so the it should not lase.

Does any of this help?

:smile_cat:

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.