RDWorks Learning Lab 106 Going Dotty Part 1

great video

[@philiprioux7767]

I don't comment on videos often, but I just had to say. It is the long road with detail that makes your videos so good. In a world of 7 sec. videos that have lost all value...your work is encouraging. Keep it up.

[@conlinepilot]

A fantastic series Russ, I look forward to every one.

[@fredflintstone2621]

Hello Russ Densitometry and reflectance of your target material is your problem. Not a simple to explain topic. So 100% dot coverage is not black on wood for example, and 0% is not white. And to make matters worse you can't Laser 99% Grey or at the other end 1%.due to Pixies running together or just not engraving at all. What you need to do is adjust your picture to match the available tonal property's of the material and process. Process being machine speed dot size, scan width. To achieve this you will use the curves feature in your photo editing software to stretch and compress the grey values of each pixel to suit the reflectance property's of your material. What you will end up with is a picture on your screen that is very ugly. So for your laser to engrave what looks like 80% you may need to engrave 50%. to engrave 10% you may need to laser at 15% this is not a linear function hence the curves feature. Different curve for different material. Do this simple test create three squares and fill them with 100%, 90%, and 50% grey. you will probably not be able to tell the 100 from the 90 and the 50 will not be half way grey looking. Now some flatbed scanas have the ability to give you a % of grey in an area, or you may be able to get that from your photo software. Just scan your lasered test piece. There plenty of people who can explain curves better than I, you tube is your friend. Have funPr

[@SarbarMultimedia]

Thanks for your great tech comment and as I mentioned in the video I have had success with getting photos onto white card,wood and slate. I have also come to realize that a suppressed tonal range works best and, to use your words, an UGLY picture comes out best because of the non linear power output of the tube. However, in the pure world of dithered graphics there is no need to worry about the tonal range of the output because dithered graphics by default is a BINARY world, not a grey scale world. The perception of shade and tone arise in the brain from the different density clusters with in the picture comprising single dots that are all of the same tone. I agree that different power will produce a different sepia shade to the overall picture but ideally every dot will have the same " colour" (for an organic material such as wood leather or card) I have played with 3D engraving where it works with continuously varying power that etches different depths into the material and at the same time applies the tonal range you describe. The 3 D nature of dithered engraving is a completely different mechanism where it is the overlapping dots that are giiving multiple burns to the same area and thus creating deeper and darker areas.. That is the mechanism that I think every one works with and takes no notice of because it gives a pleasing result. That is exactly what I am trying to avoid. I am trying to get right back to basics even if it means having to use a 50ppi resolution picture to avoid the 3 D effect. When I get to that point and probably find that the results are rubbish, I can start adding factors back into the mix to try and isolate those bits of magic that make a good picture. I feel I still have a long way to go on my journey of discovery and just as this first session was a bit of a surprise, I anticipate finding many more unexpected results Thanks again for your helpful comments Best wishes Russ

[@fredflintstone2621]

Hello Russ So after you have worked out the effective resolution of your machine you can then find the maximum line ruling you can do. Not my words - “A very common doubt in the (pre) press world is how to know what screen ruling to choose. It seems obvious to opt for a high ruling, since high rulings means a lot of detail, and that is what everybody wants. It is not that simple, however. There is a simple way to calculate the maximum screen ruling: To do the calculation, let's refer to a simple formula to calculate the maximum number of gray levels: (dpi/lpi)2 = number of grayscale levels = the number of pixels in one screen dot (Dpi refers to the output resolution of the image setter). In your case engraver. We need at least 256 grayscale levels to make sure vignettes will look smooth (no banding), so the outcome of this formula must be at least 256. Taking this into account, the formula shows that the upper limit for the screen ruling is determined by the resolution of the image setter. The higher the resolution of the image setter, the higher the screen ruling can be set. E.g. suppose the image setter resolution is 2400 dpi, the maximum output screening ruling would be 150 lpi. …......” Another topic to edify your self on is Dot gain. Also you may run into problems with a moiré pattern if your chosen line per inch collides with the step rate / belt teeth ratio of your machine. I use my machine to image PCB's and only use single direction engraving. I have tried backlash compensation on a CNC with leed screws and that works, but the same algorithm will not work with Belt drive it generates an accumulative error. Happy ablation Pr

[@SarbarMultimedia]

Hi Pr I wish this was as simple as printing where the dot size is under control and a real dot. In the case of a laser dot ,it appears that best case will be a 2:1 sausage. However, this will produce a very weak (light) picture on organic materials ( it may work on glass) and may require settings that result in 3:1 sausages ( a new term for the laser engraving lexicon!!). I'm not yet considering how I play with all these parameters just yet as there is a lot more to explore. Thanks for the great technical extract Best regards Russ

[@h4z4rd42]

Great and really informative video again, Russ. Thank you, hardly waiting for the next one.

[@jameslamb4573]

Interesting and useful, the foundations of a good video, couple that with your avuncular presentation style and it's winner. Thank you.

[@ruralireland726]

Another great video Russ. On the subject of backlash I am thinking (maybe wrongly) that it could in fact be mechanical backlash. something I have dealt with on my CNC machines. As the belt and drive wheels move the carriage to the right the teeth in both will be compressed against each other in the direction of travel. When the direction of travel is reversed then there will be a tiny amount of slack between the teeth of the belt and the drive wheel before it they are compressed against each other in that direction. As both of your machines use the same type of belt and gear drives perhaps that is why it is evident on both machines? I think that toothed gear drives are more prone to backlash than the ball screw type drives, I wonder if using a ball screw drive would make much difference on a laser machine? Just thinking aloud on that one really. Looking forward to part 2.

[@SarbarMultimedia]

I can understand the possibility of tooth compressability or even a little bit of belt "climb" as it takes up a different positions on entry and exit. My two machines are so vastly different with their pulley arrangements that it would be a huge coincidence to see the same backlash on both machines. When you look at the two sets of results that "backlash" aspect is uncannily similar. I can also see that the backlash setting will be unique to a scan speed. From that point of view I do not understand why this setting is not a feature in the advanced speed setting parameters. That way it would be unique to each layer and not an overall setting. I will be examining this feature in more detail as I have never had need to play with it before. I think it represents a very small error in the overall scheme of what I'm finding so far. Thanks for the comment and insights Best regards Russ

[@iantard5065]

what I don't understand is, if the height of the dot is twice the theoretical then thete should be no white space between the lines if you used the bitmap that you drew. did you modify the bitmap to improve the visual result?

[@SarbarMultimedia]

Hi Ian The resolution of the pattern I constructed was 100ppi That converts to a pixel every 0.254mm. My pattern was black pixel, white pixel, black pixel etc and puts the black pixel centres 0..508mm apart. With worst case, the 2" FL lens theoretical spot was 0.1mm so when you double that to 0.2 there will still be 0.308mm of white space between lines. You can see that in the horizontal direction when the sausage length exceeds 0.5mm , touching occurs Hope that now makes sense. Best wishes Russ

[@iantard5065]

Thank you Russ. i was confused because your dxf looked like 1 pixel black 1 pixel white etc. followed by a one pixel wide blank line then rinse repeat. I wrongly assumed that the pixel size was the theoretical dot size. Thank you for explaining

[@1010tesla]

I like the new scale. Sausages / mm

[@SarbarMultimedia]

Hi Dwayne A Freudian slip i guess after watching too many cooking programs. When Britain leaves the EU and we are no longer bound by Brussels red tape, perhaps I'll use the term more often and add it to the laser printing lexicon. Best wishes Russ

[@andrewpavlovic5871]

cant seem to locate the files associated with this video, where have you uploaded them to?

[@SarbarMultimedia]

Hi Andrew I don't think I uploaded any files from this particular video. If I did they will be at www.rdworkslab.com. If you need any specific files I am happy to supply them. Just use you tube personal messaging to send me your email address and the files your require. Best wishes Russ

[@soheylamohtadi484]

@@SarbarMultimedia Hello Russ, very new in all this, I was wondering if any one can help; I am trying to build a Co2 LASER cutter and am gathering bits here and there before going to actually building it yet , now I am looking for a way to control the Laser module, I was thinking of using ARDUINO and how to connect it to the power supply, I would appreciate a helping hand on this. Been watching these videos and those I watched were very interesting and have taught me a lot of things, your explanations are very simple yet very comprehensive, I don't know if I can put my e-mail her but I will any way if one can reach me without bothering others, [e-mail]. Many thanks in advance guy really appreciate. Siamak

[@ElectronicSupersonic]

I was fighting this laser scan alignment issue when scanning text this weekend. After going through all of my alignment and mechanical linkages for slop, I still couldn't fix it. I switched scan to X-Unilateralism and the problem went away. I was told it's some nasty mechanical backlash, but I was pretty sure that wasn't the whole story. I'm glad to see that I'm not the only one with this problem imgur.com/a/FopSe

[@SarbarMultimedia]

Hi Jesse I have seen this problem before and it has been issues with connections between controller and HV power supply . This is not a backlash problem it's too clean and too large. Private message me with your email address and we will talk off line to try and solve your problem. Best regards Russ