I can't achieve upper then 3kpps

[fatihak]

First off all. I found here very late. it's the first time and thanks all labour.

I bought a galvo kit from ebay Meierlight Pt-20k (20k@20 optical angle,35k@5 optical angle)

I used microController with 8bit DAC and first draw a square in square for testing. The factory setting is 20 degrees and my image is 1/5 of full frame this means i scaned at 4 degrees (does it right ?). if I pumped the ponints less then 300 microseconds it didn't achieve draw corners also upper than 300 microseconds the drawing not better.

the other test is writing some words test : FATx6 = 18 letter. Again it achieve the letters point to point 300 microseconds. I got 36 fps (testing red laser is <1miliwatt)

the dac is tlc7524 and setting time is 100 nanosecond, MicroCotroller's command cycle time is 100ns and for x,y point sending time is lower then 2 micro seconds.

Finally does it means 1/300 microseconds aproximitly 3kpps ? What is wrong ? Is the problem mechanical as bearings or electrical as tunning or me as me ?

[JohnYayas]

It could be a mechanical issue but my first guess would be that the images don't have enough points.

[mixedgas]

Not enough points. Also No blanking, anchor or guide points.

Steve

[fatihak]

I've tried with more points, image is better than the first one. but my problem is point to point delay time.factory setting is 20 degrees and >20kpps. capability in 5 degrees >35kpps.

I draw this image with 60 points, point to point delay time is 250 microsecond,
square height and width 15cm. distance is 220 cm. does it mean scan angle smaller than 5 degree?

when point to point delay time 50microseconds image is below.

is the point to point distance still too much ?

[JohnYayas]

If those green dots are the points then you have enough to get a better image than what you are seeing.

Either your galvo setup really sucks or you are sending the points faster than you think you are sending. That's my guess. I bet your code has a bug and you are sending them at a faster rate than you intend. Maybe you are off by a factor of 10?

[mixedgas]

Time = 1/Kpps.

1/30,000 = .000033 seconds between points
1/20,000 = .000050 seconds between points
1/12,000 = .000083 seconds between points


Steve

[fatihak]

thanks JohnYayas for your fast reply.the green dots are points, but the image up side I wrote wrong, dot to dot is 1.875cm. others are true square=15 cm. distance 220 cm.

I use a delay time for waiting and it's correct. first i use 2seconds delay time and see the dots besides timing, after than reduce the delay time value. Distance is 220 cm and dot to dot is 1.85cm. when I used 50 us delay time it couldn't reach the corners.can it about damping adjustment?

-/+5v I measure (scan screeen/ galvo distance)=0.41 =>(90cm/220cm) . I think it means scan angle smaller than 15 degree. besides I'm using only 15 cm of 90 cm

thanks Steve, but I haven't reached the answer yet. I have calculated,too. but can you explain about distances dot to dot. does mine true? 220cm distance to scan screen, delaying time is 50us for 1.85 cm dot to dot. is it enough delay ?

[JohnYayas]

If your timing is correct then your image is what it is. Add some dwell points to the corners and more along the lines. Do you have a DMX board or anything else that you can use to see what it looks like from a known good source?

[fatihak]

I didn't reply quickly. I'm using only PIC microcontroller, dac and opamp(amp and 0/+5 v to +/-5v convert).
I found ild file to txt file then add to my code, after than tested for timing I had an ilda patern 18kpps the circle and the square were same width.


I need some tuning for circle and speed up but I'm afraid if it goes wrong I can't turn back.
when I studied on ilda I learned dwell point are important thanks JohnYavas. I tried Elm-chan's yuki at 20k


now I'm searching chracter table.

has someone like this fonts and share the ilda file