Welcome to Our Community

Some features disabled for guests. Register Today.

LCD Based SLA Resin Printer

Discussion in '3D printers' started by evilc66, May 3, 2016.

  1. Tosch

    Tosch New
    Builder

    Joined:
    Nov 2, 2018
    Messages:
    2
    Likes Received:
    0
    Did anybody try out using two carefully aligned displays on top of each other?

    Two displays should focus the beams from the light source a bit better so the spreading of the beams is a little bit canceled and it should be possible to get finer lines. But I am not an light expert.

    Does anybody know if there is some optical material which aligns the beams parallel to each other with up to no spreading at leaving the medium?

    Here the optical filters to put on top of a display to ensure privacy of the displayed data comes to mind. It prevents reading the data from any angel exept diectly from the top.
    But then the light source has to be more powerful and perhaps it cancels the wavelenght we need to cure the resin.

    Addition: Did some research:
    3M has that type of filters. They block all above +-30 degree of visibilty which is good for focusing and they block 35% of the blue light.
    But (as always) they gave lamellars.
    There is no „resolution“ given how many of them there are in a cm/inch. But it shurely will have some moire-effect and so this solution is not practical. The display and lamellars would have the same or a n-times of each other to work properly.
    —End edit—

    Does anybody know a research facility or university which is in this field of science? Would be interesting to get in contact with them here in this thread.

    Greetings, Tom
     
    #421 Tosch, Dec 11, 2018
    Last edited: Dec 11, 2018
  2. Manoloti

    Manoloti New
    Builder

    Joined:
    Sep 27, 2018
    Messages:
    4
    Likes Received:
    2
    @Tosch

    That “filter” over the screen sounds as an interesting thread.
    As far as I can tell, the screen by itself, and mostly its polarizer (not removable, other wise you won’t be able to form the image), reduces the light outcome in an important degree. I don’t have the means to measure it, but...I know it happens. That’s why is important to take all possible beneficiary variables in account in order to get the most “effective curing light” possible

    What I’ve done (that I didn’t mention before) is to use some tiny lenses (plastic ones) specific for the LED modules, that condense the light beam. From the factory setting of 120° of these LEDs to....15° in the 1w array, and right know I change them for some 30° ones (for the on going 5w array), because 15° in the 7cm of distance wasn’t enough to properly cover the surface of the screen. They are reaaaaaally useful and do a great lot to boost the cd (candelas) of the light source.
    On top of that I performed several test playing with the digital settings of the screen, such as contrast, brightness and gamma (in individual channels, in duos and all the RGB combined). From those experiences, I got away with the conclusion that by mainly boosting the gamma setting you can get a better performance in curing times. Not a whole lot, but I was able to reduce the layer time by 1s, that....in this “super not sensible” resin is remarkable. I should also remark that this change in gamma didn’t affect the “agglutination” of the resin, which remain even when cured.

    Concerning you question about research facilities, not so ling ago, I came across an article of an English university from...2015 (I’m not sure, I have the link to check it if you want) that I believe was the cornerstone for the Photocentric people to develop their enterprise. I....don’t believe that by any mean they will share any of their investigation, they are as closed as f*** when it comes to information. They just...don’t share the vision of this technology we might have. That is concerning “daylight” resins, now....UV ones....they more widely used and for sure there are some other people there on the Internet that can explain better how they work and their capacities. But the bottom line is that the DLP printing is not....so popular and...hasn’t found enough “grip” in actual industry or....activities to be further developed. By trying to be successful in my project I hope for at least my area of interest (designed visual experiences) open a new door, that designers can make use of, and further develop.

    Yours, Manuel.
     
  3. Kyle1988

    Kyle1988 New
    Builder

    Joined:
    Feb 18, 2019
    Messages:
    2
    Likes Received:
    1
  4. WirelessguyNY

    Builder

    Joined:
    Dec 19, 2016
    Messages:
    8
    Likes Received:
    1
    Do they offer a display driver?

    9 times out of 10 when I find a decent display there is nothing available for it. Given the dimensions and resolution the PPI would be pretty nice. If you plan on daylight resins this might be an easy choice. UV might need some testing.

     
    Kyle1988 likes this.
  5. Manoloti

    Manoloti New
    Builder

    Joined:
    Sep 27, 2018
    Messages:
    4
    Likes Received:
    2
    What you say is totally true. Most of times either there’s no availability or is a scam (I nearly got scammed by some website called LCDsource if a recall properly). I would suggest do it through Ali-express or Ali-baba, is the most secure option. Concerning the panel itself, it’s great. I originally was looking for that model but after many conversations the supplier didn’t have any in proper conditions, guess is no longer produced.
    @Kyle1988 if you can actually adquire it safely, those are would be great news.
     
  6. Kyle1988

    Kyle1988 New
    Builder

    Joined:
    Feb 18, 2019
    Messages:
    2
    Likes Received:
    1
    Well the last two things I am going to buy to assemble my build will be LCD and the LEDs accordingly. If I end up getting ahold of one I'll let yall know
     
    Manoloti likes this.
  7. sblaszak

    sblaszak New
    Builder

    Joined:
    Jun 30, 2019
    Messages:
    1
    Likes Received:
    0
    Hi All,

    I stumbled across this thread in the process of searching the Internet for design plans/ideas for LCD SLA 3d printers and have to say that evilc66's progress looks awesome. While I haven't, yet, had the chance to go through every post, a quick skim combined with Tosch's recent mention of doubling up LCD screens motivated me to register and make a post.

    A few times, in my Internet searching, I came across some DIY LCD SLA experimenters that were discussing the UV transmissivity of various types of LCD panels. Specifically, the discussion was about factors that make the displays less transmissive of UV light. There seemed to be a pretty strong implication that every time you increase the resolution of a display a very large decrease in UV transmissivity is seen. The numbers thrown around seemed to suggest that, for example, going from the average 2K display to a 4K display would mean a decrease from 1000s of prints between LCD burnout to only 100's or even 10's of prints between LCD burnout. Basically, it seemed like the supposed cause of this rapid decrease in panel life was a combination of increased panel heat-up caused by more light being absorbed instead of transmitted along with the resulting need to significantly increase the power of the UV light source to punch through the more opaque panel (there is also the fact that electronics in general, and opto-electronics in particular, have a tendency to fail at a non-linearly increasing rate as the heat and/or luminous flux increases linearly. This is something I know from many years of profession experience working in the engineer section of a lighting manufacturer.). While removing all filter layers on the front of the panel, if that's even possible, would certainly help, I believe that even with that the general consensus seemed to be that it would still shrink the panel life to something absurdly short and to, instead, focus on using a slightly lower resolution panel paired with anti-aliasing to help compensate for the loss in resolution.

    This ties in with Tosch's suggestion because I can see doubling up the panels as causing a similar situation. On the other hand, in this case, you now have the possibility of swapping the panels over time (like rotating the tires on a car) to even out the wear rate and/or just having to replace the top panel less often due to it being shielded from a good amount of the heat build-up and luminous flux by the lower panel.

    I apologize for not remembering exactly where I saw this discussion, otherwise I would provide a link. While I know there is a good chance you are already familiar with these concepts, I figured I would throw them out there just in case. While they're not, necessarily, a reason to throw out your present plans to experiment with using ultra-high resolution panels they may, at least, help give an avenue to explore in the event that you start seeing unexpectedly rapid panel failures and help avoid some unnecessary extra component expenses. I, also, apologize if this has all been discussed in this thread already and I just missed it in my limited skim.

    Regards,

    sblaszak
     
  8. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    I'm back!!!!

    First things first, I suck. It's been a long time since I posted any updates on this, and for that I'm sorry. I know a lot of you have been following along hoping to duplicate what I'm doing. Hopefully, this brief update will help some.

    I'm just going to get into some points about the progress I've made, and I'll make some more detailed posts a little later if needed (pictures and what not).

    First up is the issue I was having with getting a good image on the 8.9" display and NanoDLP. This was a serious head scratcher. On the Raspbian desktop, the image was perfect with no artifacts, tearing, random glitches. The same could not be said for NanoDLP. I played around with changing the display settings in the config file, as well as the resolution settings in NanoDLP. Sometimes I could get the projector calibration image to be stable, but once I went to print something, the image would jump around all over the place. Not good. None of the research I did found any solution, but I did find others with similar issues. It was finally suggested by Shahin (NanoDLP developer) to use the frame buffer option. Most people pass over this option, as it only shows up in the advanced settings, but enabling that worked a treat. So now that's out of the way.

    Build chamber air filtering is done, with two chambers at the back of the machine with activated carbon, and 60mm blowers forcing air through. Works pretty well, but I have to get some better quality blowers as these ones are noisy (vibration, not just airflow noise)

    Getting the RAMPS board working correctly took a minute. Looks like there are multiple variants of the mUVe3D Marlin firmware. I swear I downloaded the one that is linked on the NanoDLP site, but it would not cooperate with z-axis moves. I found a different one on Github and that works just fine.

    The biggest problem I'm having right now is getting the resin to cure. I'm using Siraya Tech's Easy resin (I also have Fast to test with), as it seems to be a decent resin for a decent price. Can't get it to stick to the build plate. My build plate is just plain old 6061-T6 aluminum (not cast MIC-6 like I have used in other projects), and it would not stick. Roughed it up with some 120-grit sand paper, still won't stick. Cranked up my exposure times, still won't stick. Sticks great to the FEP film though! It pops off easy enough with a little pressure from underneath.

    So, seeing as all the usual tricks to get it to work haven't, I'm enlisting heavy weaponry to see if I can figure this out. I have in my possession an Ocean Optics USB4000 spectrometer. For those of you that don't know what a spectrometer is, it's a device used for measuring light in both it's spectral content, and its relative intensity (you need other, very expensive equipment to measure absolute intensity). I set it up so that the measurement range for the spectrometer was maxed out with the display moved out of the way, which maxes out at about 65,000 counts (remember, this is relative intensity, so there aren't units involved). With the LCD in place, and the 'full white' calibration image turned on in NanoDLP, the counts dropped to ~2300 counts. ~3.5% of the UV light is making it through :( I still have to compare this to the D7 sitting on my bench, but 3.5% transmission ratio seems really low. I did verify that the LEDs are 405nm, as I thought that might be an issue (I'm pretty sure I ordered these as 395nm LEDs). I'm going to try and get a measurement on the D7 in the next few days.

    I have messaged the seller/manufacturer of the display to see if there are any UV filters that can be removed, either by me, or by them if it's not feasible for me to do it. I did find out that the film on the front of the display is not a neutral density filter like I assumed, and is in fact one of the polarizers. Removing that film renders the display useless, so take note from my experiment ;) Fortunately, these displays have become very cheap (paid $73 shipped for LCD and controller this last time, was over $225 when I first bought them).

    Anyway, some progress. More to follow
     
  9. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    After a little more digging, it seems like this LCD has a particularly aggressive polarizer (at least on the front) that blocks almost all of the light below 450nm. Conveniently, Amazon stocks linear polarizers that claim to have high transmittance down to 400nm. Got some on order that will be here later today. I also found out that the rear polarizer is a really thin, and quite well hidden film on the back of the LCD. That got removed last night in preparation of the inbound polarizer films.

    Maybe, just maybe, I'll have a working printer tonight.

    I'll post details later
     
    Paul Stoller likes this.
  10. Theatricalmayhem

    Builder

    Joined:
    Dec 3, 2019
    Messages:
    2
    Likes Received:
    0
    I have a related question about the Sharp LS055R1SX04 2K LCD screen. Does anyone know if this panel will pass 365nm light or does the polarizer/filter prevent this? It seems that it is widely used at 405nm. Additionally, what is the best software for direct pixel control of these types of setups? NanoDLP with an mipi to hdmi board or something else?
     
  11. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    From the reading I've been doing lately, you will have a hard time getting any appreciable amount of light through the screen at 365nm. I had originally hoped to use shorter wavelength LEDs for faster curing time and wider resin availability (at the time), but I'm even struggling right now with 405nm on my current screen.

    I ordered a Phrozen Shuffle XL screen to see if there were any differences in the way the screen is built. For all I know, they specified different polarizers for their screen versus the ones that are widely available on Aliexpress. I'll find out Thursday.

    As for what to use to control it, NanoDLP really is the best option for all aspects of the printer.
     
  12. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    Phrozen Shuffle XL screen is no different than what you can buy on Aliexpress for a third of the cost :(

    Looks like this is going to be a brute force approach, unless I can source good polarizers (which I might have). If I can get good polarizers, then the next challenge is going to be removing that third polarizer from the back of the LCD. I've been looking around at LOCA (liquid optically clear adhesive) removers to see if that will do the trick, but the most recommended one is out of stock most everywhere. Goo Gone Pro supposedly works well for the adhesive removal once the polarizer is removed (which is a giant pain in the ***), but I have no idea if it's damaging if left in contact with the LCD for extended periods.

    The Phrozen Shuffle XL uses a 150W array of LEDs with secondary optics. With the array much closer to the LCD than mine, combined with higher wattage and secondary optics, the throughput through the screen is going to be much higher than I'm experiencing. I'm hoping the pile of LEDs that I ordered close to 3 weeks ago will be here soon so that I can replace what I currently have and move everything closer to the screen.

    My woes continue :(
     
  13. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    IMG_20200101_170201999_HDR.jpg

    One 1/2" slab of aluminum, 135 LED's, and 105 solder joints later, we have the beginnings of the new LED array. These are verified at 405nm with the spectrometer, and have a 60 degree emission angle, so the light distribution should be pretty even. I have a new LED driver coming in tomorrow to power the whole thing. The array will be 27 LEDs in series, and 5 groups of these in parallel at 350ma per series string. That should put me at about 150W. That will be 30W more than the original array, and I'll be mounting it about 1" from the LCD instead of 3". With those changes, I should see a massive change in throughput through the LCD.
     
  14. SoapyShazbot

    Builder

    Joined:
    Jan 5, 2020
    Messages:
    1
    Likes Received:
    0
    Holy **** I've been dreaming about something like this thread for a while, first post on the forum!
     
  15. artag

    artag New
    Builder

    Joined:
    Jan 9, 2020
    Messages:
    5
    Likes Received:
    0
    Good to see there's still discussion here on SLA printers. I've been planning to build an LCD or DLP printer for a while and slowly accumulated some components when I was given an old 3d systems printer. This is a UV DLP printer with a fairly decent chassis and should make a good candidate for rebuilding as a hackable device that I could use to experiment with new ideas and resin products.

    But all the places I'd previously found for DIY SLA builds seem to have gone quiet - is this the only place left to compare notes with other builders or is there an active forum somewhere ?

    It seems odd that cheap commercial SLA printers are now quite numerous but software (such as nanoDLP) is mostly closed source and development has disappeared.
     
  16. evilc66

    evilc66 Journeyman
    Builder

    Joined:
    Jan 9, 2015
    Messages:
    392
    Likes Received:
    153
    There's certainly less discussion than I thought there would be for something like this. Buildyourownsla.com is almost a ghost town at this point. There are some Facebook groups, but that's not the format that I enjoy talking about and sharing ideas. We have some pretty sharp people here though, even though there isn't a lot of discussion on SLA printers. I'm trying to share as much as I can, when I can, but we aren't finished yet, so we don't really have an A-Z guide yet on how to do this.

    As for how I'm progressing, I have the new LED array fully wired to the driver now. I made a small interfacing circuit to allow the Arduino to control the driver (it uses a 10v PWM input, Arduino is only 5v), and that will be installed this weekend. I have yet to do a direct comparison between the old array and the new one as far as how much light is allowed to pass. I'll hopefully do that this weekend too.

    I do have an additional trick up my sleeve though ;) I ordered the best polarizing film that I could find to see if I can swap out the polarizers on one of the screens that I have been taking apart. API Optics has a large selection of polarizers that are readily available to the public, which is nice.

    So, a polarizer like this one is intended for LCD use. If you look at some of the values, like single pass transmission, it looks pretty good. The one I linked to is 42% single pass. When you cross them, it blocks out almost all light. It's down to 0.007%. This is great for a display that needs high contrast. We don't need that (at least not in the same way that a traditional LCD would). This polarizer, which is the one I ordered has a single pass transmission of 50%, but crossed transmission of 10.2%. When it says "crossed", that means at 90 degrees to each other, which normally would block virtually all light. In an LCD like the one I'm using, the polarizers aren't perpendicular to each other. The front and back are parallel, while the middle is at 45 degrees. At least, that's what I've been able to figure out so far. At 45 degrees the transmission rate will be even higher than 10.2% I'm hoping that if the new LED array isn't quite up to the task, then replacing the polarizers will get me to the point I want. I'm hoping I don't have to resort to that, as it's not easy to do. Getting the front and back polarizers off wasn't too bad, but the middle one is a right pain to remove.

    Anyway, hopefully some more updates soon.
     
  17. Theatricalmayhem

    Builder

    Joined:
    Dec 3, 2019
    Messages:
    2
    Likes Received:
    0
    Does anyone know what model number the phrozen sonic screen is? They claim its a mono screen and thus are getting faster cure times. I cant seem to find anyone who will sell it to me. These guys are out of stock Phrozen Sonic HD LCD Module | FEPshop - 3D printer experts. Seems as though this would be ideal for lower wavelength transmission.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice