Hi Tweakie,

I agree again. I was just trying to say it was not theoretically correct a way of description. There will always be the slightest offset between the two dots. It might be small, but it's there. And without it you wouldn't again be able to produce shades of grey

Dan

Hi Tweakie,

I agree with this. It's just that I think that using the word "power" to describe this process is misleading as the laser is being turned on to the same (predefined by some other means) power each time.

And to be more correct, I think that the laser never really does several pulses on top of one another (well if it's focused fine enough that is), as the axis is in a constant move. I think the correct way of thinking about it is the dots intensity. And the other thing to consider while at it is that the exposure time is determined by the axis feed rate.

Yes, I had read through your PDF and I must say you've done great job there!! Thanks for that!

Dan

Well, I think the plugin doesn't control the laser power whatsoever. All it does is trigger it on and off.

Dan

On it's own, the plugin will not always smooth a gradient of tone without producing 'steps' as can be seen in the background of this image which is a smooth transition from light to dark.

Tweakie, just out of curiosity I would try engraving the same image rotated by 90°.

John, I think the plugin outputs equal length pulses for each pixel. As far as I understand it's the spacing which creates the toning.

Dan

EDIT: Found original Art's description:

http://groups.yahoo.com/group/mach1mach2cnc/message/74912

Hi Tweakie,

So how does it vary from the Impact plugin which still fires the laser at the same power each time? Roughly speaking, if I understand correctly the way the plugin works, it does internally the same thing as that program does, but following a different algorithm.

Dan

Can you explain please, Tweakie? The power supply still operates in digital mode so how does the image conversion help?

In the image the program outputs, are all the dots equal in size?

Dan

Yes, using the bolts against the turret would damage it.

You're lucky to have clean air

Dan

Thanks for the description, Hood. It is a nice solution. Just a thought here - looks like it might be better to have the screws' heads touching against the slot wall rather than the tool, as when tightened they would tend to drag the tool out of position.

So you have to use shims for height adjustments?

Interesting.. the sea here is a few kilometers away and everything picks up rust pretty quickly unless it is a good alloyed steel. Think it has to do with the air pollution levels as it's soaring here.

Dan

Hi Hood,

Can you please show how the tool is fixed in the slot?

What steel is it? Aren't you going to have it plated? Your weather is the worst one could imagine for machine tools.

Dan

In any case, this doesn't change anything about the segmented arcs I'm getting on this part. I tried exact stop this afternoon, and it didn't make a difference. Still getting the same behavior.   

I think the answer lies right there, in the numbers of your test. X axis is pretty much obvious, heaving an absolute deviation of 0.002", which is enough for those segments to be seen. I would guess that if you did the test over a longer distance you would see a distinct repetitive nature of the numbers. Entering the numbers in Excel would make it easier to visualize.

Since you're not using a belt drive, you have to start looking elsewhere in the driver mechanism. For instance, what is the motor? Could be a stepper motor with poor accuracy - unevenly spaced poles or magnets on the rotor. Since your Y axis is much better than the X, you could try swapping the motors.

Dan