Re: [diy_3d_printing_and_fabrication] Re: Projector Calibration/ Lens Distortion
- Well, for high accuracy printing with a DMD chip using a concentrated light source, DTGSMOMO may be the only way to assure a uniform curing.
This, by the way, is something that Etec printers have inbuilt: they do a claibratin procedure every-so-often
On 05/20/2013 02:26 PM, Bhatt Himanshu wrote:
BTW for DTGSMOMO did you use and specific pattern/file or Brute Force(Check each and every portion)method?? any Suggestion of using simpler and faster method(STL file etc)??
--- In firstname.lastname@example.org, "afogassa" <afogassa@...> wrote:
> Benq lens sucks, not good for what we do, try to project the grid and adjust your projector to get a uniform line thickness, no gost lines, no out of focus lines across the build area. if it´s impossible than you will need a better projector or replace the stock lenses(not easy to find).
> Adjusting the lamp position can solve your light uneveness, but you always end up with diferences in light across the build area.
> one way to solve that is to use a gray scale mask or extend the projection time on places that are not getting enough light.
> i have solved that by using this super hyper DTGSMOMO, what a heck is that? (digital transparent gray scale mask of my own :-)
> --- In email@example.com, "Bhatt Himanshu" <bhatthimanshu@> wrote:
> > Hello,
> > I tried several prints and came to conclusion that in my setup i get good prints in the middle area of my Build platform.
> > for testing i tried rows and colums of cylindrical support structure. from 0.9mm to 0.4 mm in diameter. first few prints, it dint print support structures bellow 0.7mm. then i increase cure time upto 30 sec. now i got 0.4 but only in the middle same structures on the left or right side of projectors are broken or not printed. at the Cure time of 30 sec i can use 25mm Build platform for supports up to 0.4mm but now the problem is in HOLES i didnt get the small holes of 0.4mm diameter which i was gettting earlier @ 18-20 sec cure time..
> > after searching on google i found this
> > http://www-bcf.usc.edu/~yongchen/Research/SFF_09_DLPCalbration.pdf
> > they resolve the problem using hitech algorithms .
> > I can print Idols, Yoda, Coins easily with great details but fine jewellary makes the problem coz in that i need both holes and small support structure (easily removable).
> > Here is my setup details.
> > 2800 Lumen Benqmx518
> > 50um XY resolution
> > 38um Z resolution
> > Software: B9C
> > Resin : B9R Red
> > I am thinking to use grey scale in problematic area. but for that i need extensive trials. if you guys have clever calibration process please share..
> > if you guys faced same problem than please share with me.
> > Thanks,
> > Regards,
> > Himanshu Bhatt
- On Wed, May 22, 2013 at 11:32 PM, Bhatt Himanshu <bhatthimanshu@...> wrote:
I read about ND filters and i think i can get it from local camera shop. Do i need diffuser if i use ND filter/filters properly.The reason I suggested a diffuser is because I was talking about converting a camera into a sensor so the "image" you get is actually an average of the light hitting the sensor. The idea of an ND filter is just to reduce the light levels though as I mention on another thread you can do this by projecting darker images anyway I think.I want to make it clear that I am not suggesting trying to measure the output of a specific pixel on the projector but rather to get the average intensity over a fairly small area (because crappy webcam sensors are tiny) and then use a grid of these measurements to produce a correction which should be fairly smooth if the aberration is only caused by the optics.Cheers,Graham