Loading ...
Sorry, an error occurred while loading the content.
 

Re: [diy_3d_printing_and_fabrication] Re: wave lenght

Expand Messages
  • Graham Stabler
    Francis, Attached is a diagram showing the problem that you MAY face. Can t wait to see more from your experimentation, it is quite correct that you can get
    Message 1 of 35 , Jan 23, 2013
    Francis,

    Attached is a diagram showing the problem that you MAY face. 

    Can't wait to see more from your experimentation, it is quite correct that you can get away with less if it is the right wavelength.

    Graham



    On Wed, Jan 23, 2013 at 5:35 PM, Francis Leach <francisleach@...> wrote:
     

    Torben

    I have to assume the active element of the ORBIS projector is a LCD unit but I can not confirm the actual working principles . Perhaps it uses polarization like a laptop screen but it normally has to cope with 2300 Lumens of white light through tricoloured filters over each pixel site. In the narrow frequency in the UV range it will have a low energy density. The images of the slices are not multi coloured but Black or white,  our maybe  4 shades of grey between white and black. I have got to give it a try. It is a low cost solution that should not be dismissed because no one has tried it. The exposure times will be less that a white light system as UV photoinitator can be tweeked in the UV resin formulations.
      


    On 23 January 2013 15:05, torben.mogensen@... <torben.f.mogensen@...> wrote:
     

    Did you consider that LCD panels are polarized and will only let through 50% of the light.

    I did some investigation in to LCD panels - which is very tempting to use, but went the DLP way, because of the issue with polarization and also there is issues with UV degrading LCD panels.



    --- In diy_3d_printing_and_fabrication@yahoogroups.com, "Francis" wrote:
    >
    > I believe a rear back light panel can be constructed using a panel of acrylic and several 10W UV LEDS . They are much easier to drive and less costly.
    > The"as bought" ORBIS projector light system I am using( £100 on ebay ) is very simply. At present It uses a white visible light, high pressure lamp and a concave reflector behind the lamp to project forward onto a Fresnel lens which ensures the liquid crystal module is illuminated with parallel light. Having passed through the active LCD unit the light then passes through another Fresnel lens which focusses the image through the final projection lens.
    >
    > I intend to make a UV light source to replace the lamp. The UV source is initially to be 4 X 10W LED module shining into the edge of 5 thicknesses of acrylic panel stacked and mechanically clamped together. The acrylic panels were recovered from several dis-assembled defunct Laptop screens. A local laptop repairer was glad to give them away to me. The 4 LED modules have to be mounted on an aluminium heat sink which is cooled by the existing electric fans. Each LED is powered from a constant 10V voltage and 1Amp current module powered via a 12Vdc external source ( Computer power module ).
    >
    > I have done no testing to verify that 4 LED will provide sufficient power but I assume them to be 6 times more effective than a light light source when curing resin thus ( 10W X 6 X 4 = 240W equivalent compared with the 180w white lamp presently fitted.
    >
    > In this arrangement the number of 10W LEDS could be doubled or quadrupled using the same principle. I am intent on as short an exposure time as possible for a rapid build per layer. I have it in mind to explore the resin with multiple exposures of the same layer with different grey scale images in order to smooth the final surface finish.
    >
    > The Laptop acrylic panel is manufactured such that the entire rear face of the flat area is patterned with thousands of pin pricks which penetrate the surface. Normally Acrylic exhibits the phenomena of total internal light refraction, that is the light which enters via a thin edge, is reflected back into it's volume and will not emerge from the face perpendicular to the point of entry, unless a surface imperfection allows the light to do so. The light emerges only from the penetrating surface features. In effect it is a light panel with many thousands of light sources to spread the light over the entire surface. The acrylic panels act as a backward projecting light system using a a white plastic sheet as the rear reflector on the back face of the stack. The light is therefore projected forward back through the stacked acrylic panels onto the first Fresnel lens of the projectors system. The edges of the stack is encased in white plastic reflector and a housed in aluminium sheath ( reformed lager can) to block any other light escape. The thin aluminium is annealed and flattened. It can be marked out and cut with sturdy shears. The sheath can be formed by scribing it with a ball point pen and folding to the lines. The sheath is in the form of two slide together halves , matchbox style. The LED/heat sink is attached to one end and the second half of the sheath slides over the assembled acrylic panel.
    > If the multilayer stack does not liberate a successful solution a single thickness of 10mm will be prepared with a face drilled pattern of holes which will act as the illumination feature previously mentioned.
    >
    >
    > --- In diy_3d_printing_and_fabrication@yahoogroups.com, "torben.mogensen@" wrote:
    > >
    > > In LED projectors, concentrating the beam is already taken care of, as you are replacing one LED with another LED.
    > >
    > > If replacing a bulb, you will need a reflector. LedEngin has reflectors to fit their LED's - There is an 8 degree, I think that should do the trick.
    > >
    > > Yes, these LEDs are extremly expensive, but if you are replacing a bulb you would have to spend 100-150 anyway for the replacement.
    > >
    > > The smaller LED (4 elements) will properly also cure the resin, but I don't think you would get any speed improvement.
    > >
    > >
    > >
    > >
    > >
    > >
    > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler wrote:
    > > >
    > > > I don't think a 12 or 24 element LED without some sort of optics (light
    > > > guides or beam splitters etc) to get all of that light in one place will
    > > > not work well at all. Most of the light will go to illuminating the inside
    > > > of the projector and/or you will end up with a blurred image.
    > > >
    > > > That's what I think is the case if anyone want to discuss it. I certainly
    > > > would not spend $200 on one of those LEDs without testing the performance
    > > > with a low power extended (not point) source.
    > > >
    > > > Graham
    > > >
    > > >
    > > >
    > > >
    > > >
    > > > On Wed, Jan 23, 2013 at 1:13 PM, torben.mogensen@ <
    > > > torben.f.mogensen@> wrote:
    > > >
    > > > > **
    > > > >
    > > > >
    > > > > Hi, I'm going down that route.
    > > > > I found that LedEngin.com is the most powerful LED's.
    > > > > They have a LZP series (24 element) and a LZP (12 element)
    > > > > They are quite pricy .. $350/$200, mouser.com have them in stock.
    > > > >
    > > > > They require 21-23V, so you need a separate power supply and a rather big
    > > > > one, as the consume 4Amp/2Amp.
    > > > >
    > > > > If you are replacing a projector-bulb - you will need a reflector also.
    > > > >
    > > > > You need to consider cooling also - they get very hot and loose
    > > > > effectivness if they get over 70-80 degrees C. Specially if you leave it on
    > > > > all the time.
    > > > > I'm planing to control on/off with the microcontroller (my case Raspberry
    > > > > Pi) to reduce heat
    > > > >
    > > > > AFIK you need 400 Nm, 450Nm is just outside the useful spektrum for the
    > > > > photoinitiators I have seen.
    > > > >
    > > > > I made some tests on Fernando's resin with 400 Nm and it cures VERY fast.
    > > > > I have written about this in another post.
    > > > >
    > > > >
    > > > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler
    > > > > wrote:
    > > > > >
    > > > > > Ideally you want something that will match the way the original lamp gave
    > > > > > out it's light so that the optics can collect the light effectively and
    > > > > > project it well. I suspect a single source and no lens might be best but
    > > > > I
    > > > > > have never run a projector lamp outside of a projector, does the mirror
    > > > > of
    > > > > > the lamp collimate the light?
    > > > > >
    > > > > > If it does then the LED could be mounted in an old lamp facing backwards.
    > > > > > Of course if it is multi element then you will actually end up with
    > > > > > multiple collimated beams which may blur the image.
    > > > > >
    > > > > > It would be really nice to have a long lasting bulb replacement so good
    > > > > > luck with your quest.
    > > > > >
    > > > > > Graham
    > > > > >
    > > > > >
    > > > > >
    > > > > > On Wed, Jan 23, 2013 at 12:07 PM, afogassa wrote:
    > > > > >
    > > > > > > **
    > > > >
    > > > > > >
    > > > > > >
    > > > > > > I was just browsing ebay for it and found some multi element 405nm
    > > > > leds in
    > > > > > > the 10w, and 450nm up to 100w there is dome lens for those avaiable.
    > > > > > > just thinking since a replacement bulb is very expensive and sooner or
    > > > > > > later it will blow up anyway, i don´t think the light bulb on this
    > > > > > > projectors will outlast the life expectancy of 5000 hours before it
    > > > > blows
    > > > > > > up by some unknown reasons.
    > > > > > > well, acctually there is reason for mine to blow up, the first one the
    > > > > > > original one crack becouse of power failure, the second one becouse i
    > > > > > > bought on ebay made by chinese grade c manufacturer.
    > > > > > > So, yes I´m getting pissed and about to go back to the stone age i mean
    > > > > > > powder printer.
    > > > > > >
    > > > > > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler
    > > > > > > wrote:
    > > > > > > >
    > > > > > > > Alvaro,
    > > > > > > >
    > > > > > > > What is the most powerful LED source you can get with a single
    > > > > element?
    > > > > > > If
    > > > > > > > it has an array of 9 like some I have seen then the projection will
    > > > > not
    > > > > > > be
    > > > > > > > very good I think.
    > > > > > > >
    > > > > > > > Graham
    > > > > > > >
    > > > > > > > On Wed, Jan 23, 2013 at 10:18 AM, afogassa wrote:
    > > > > > > >
    > > > > > > > > **
    > > > > > >
    > > > > > > > >
    > > > > > > > >
    > > > > > > > > Since my lamp burned(craked) and the one i´ve bought from china
    > > > > didn´t
    > > > > > > > > last 10min before pop out i´m wondering about replacing the lamp by
    > > > > > > 20w uv
    > > > > > > > > leds lets say 405nm or 440nm.
    > > > > > > > > What wave lenght is being used to harden Fernandos resin? I mean
    > > > > what
    > > > > > > > > portion of the white light is being acctualy used to harden the
    > > > > resin?
    > > > > > > > > maybe 2600 lumens is over kill since we use just the 400 to 450nm
    > > > > waive
    > > > > > > > > lenght? maybe a 20,30w led with the right waive lenght is enaugh?
    > > > > > > > >
    > > > > > > > >
    > > > > > > > >
    > > > > > > >
    > > > > > >
    > > > > > >
    > > > > > >
    > > > > >
    > > > >
    > > > >
    > > > >
    > > >
    > >
    >




    --
    Regards

    Francis Leach


  • Graham Stabler
    It is not an easy problem, I ve played with this sort of thing before and always found I lose a lot of light because as the guide tapers you get more back
    Message 35 of 35 , Jan 26, 2013
      It is not an easy problem, I've played with this sort of thing before and always found I lose a lot of light because as the guide tapers you get more back reflection. I would start with simple experiments with a single LED to see how much light you end up keeping. I would think a square end would be fine.

      One concept that I was thinking about is a fused fibre optic bundle which is then tapered. These are used as image intensifiers, the light hits the wide end and ends up intensified at the narrow end. Note that generally using a lens to make a smaller version of an images does not actually make a brighter version only when all the light can be brought to a tight focus so if you make an image of the light from an LED only smaller it will not be any brighter. The tapered fiber bundle on the other hand literally squeezes it down smaller. But you need to get the light in the fibre, that might be best done by butt jointing the the emitter so go for an LED without lens if possible.

      Proper intensifiers are EXTREMELY expensive but you can buy "image conduit" for somewhat less, that could be heated and drawn to a taper and it can also be bent with heat.. 

      But really I am just throwing these ideas out there, I am not sure it would work and it might be rather expensive. A blue ray laser diode (£7) and a driver (£20) might be worth trying first. Not UV but 405nm and closer to the desired wavelengths.

      My gut feeling is that a cheap DLP projector IS the cheap solution. Mechanically scanning a laser diode is another cheap albeit slow solution. And last year we talked about hacking a spinning mirror from a laser printer, that also has potential. But in any case I would start simple and get a feel for what you can do before putting a lot of work into it.

      Cheers,

      Graham 



      On Sat, Jan 26, 2013 at 1:04 PM, Francis Leach <francisleach@...> wrote:
       

      Graham

      Having considered your recommendation regarding a point light source as the only suitable arrangement, I really appreciate this guidance as I could have been working toward a dead end situation.

      Following on from that realisation, I wondered if I could arrange for a light guide made of acrylic to be used to combine the output of 4 or 6 UV LED arrays, so the accumulated light could be made to emerge at the focal point of the condenser set/Fresnel lens before the LCD element. Would the end of the light guide be best as a small half spherical bead or simply polished smooth.
      I suggest using a brass shell to contain the acrylic light guide as It can be fashioned as a split shell with air cooling fins and a suitable bracket to ensure it is aligned correctly with the Fresnel lens 

      I use 3D modelling so it would be simple to model a investment cast brass shell that could be filled with acrylic resin.  If you think there is any potential in it, I will model the item and submit it for scrutiny and critical review by you and the group.






      On 23 January 2013 23:05, Francis Leach <francisleach@...> wrote:
      Graham

      Thank you for the diagrams. It does help to explain the optical system in the Orbis projector. There is a small difference and you may be able to explain it's effect..
      In the existing projector the Condensor system shown in the diagram is replaced with a fresnal lens presumably to reduce the cost of glass optics.
      If the Fresnel lens is retained and close coupled to the light off the back-light which therefore full impinges on the lens. Does it not go on to be made to move through the active slide unit as your depict without the lose of light?




      On 23 January 2013 18:52, Graham Stabler <grezmos@...> wrote:
       
      [Attachment(s) from Graham Stabler included below]

      Francis,


      Attached is a diagram showing the problem that you MAY face. 

      Can't wait to see more from your experimentation, it is quite correct that you can get away with less if it is the right wavelength.

      Graham



      On Wed, Jan 23, 2013 at 5:35 PM, Francis Leach <francisleach@...> wrote:
       

      Torben

      I have to assume the active element of the ORBIS projector is a LCD unit but I can not confirm the actual working principles . Perhaps it uses polarization like a laptop screen but it normally has to cope with 2300 Lumens of white light through tricoloured filters over each pixel site. In the narrow frequency in the UV range it will have a low energy density. The images of the slices are not multi coloured but Black or white,  our maybe  4 shades of grey between white and black. I have got to give it a try. It is a low cost solution that should not be dismissed because no one has tried it. The exposure times will be less that a white light system as UV photoinitator can be tweeked in the UV resin formulations.
        


      On 23 January 2013 15:05, torben.mogensen@... <torben.f.mogensen@...> wrote:
       

      Did you consider that LCD panels are polarized and will only let through 50% of the light.

      I did some investigation in to LCD panels - which is very tempting to use, but went the DLP way, because of the issue with polarization and also there is issues with UV degrading LCD panels.



      --- In diy_3d_printing_and_fabrication@yahoogroups.com, "Francis" wrote:
      >
      > I believe a rear back light panel can be constructed using a panel of acrylic and several 10W UV LEDS . They are much easier to drive and less costly.
      > The"as bought" ORBIS projector light system I am using( £100 on ebay ) is very simply. At present It uses a white visible light, high pressure lamp and a concave reflector behind the lamp to project forward onto a Fresnel lens which ensures the liquid crystal module is illuminated with parallel light. Having passed through the active LCD unit the light then passes through another Fresnel lens which focusses the image through the final projection lens.
      >
      > I intend to make a UV light source to replace the lamp. The UV source is initially to be 4 X 10W LED module shining into the edge of 5 thicknesses of acrylic panel stacked and mechanically clamped together. The acrylic panels were recovered from several dis-assembled defunct Laptop screens. A local laptop repairer was glad to give them away to me. The 4 LED modules have to be mounted on an aluminium heat sink which is cooled by the existing electric fans. Each LED is powered from a constant 10V voltage and 1Amp current module powered via a 12Vdc external source ( Computer power module ).
      >
      > I have done no testing to verify that 4 LED will provide sufficient power but I assume them to be 6 times more effective than a light light source when curing resin thus ( 10W X 6 X 4 = 240W equivalent compared with the 180w white lamp presently fitted.
      >
      > In this arrangement the number of 10W LEDS could be doubled or quadrupled using the same principle. I am intent on as short an exposure time as possible for a rapid build per layer. I have it in mind to explore the resin with multiple exposures of the same layer with different grey scale images in order to smooth the final surface finish.
      >
      > The Laptop acrylic panel is manufactured such that the entire rear face of the flat area is patterned with thousands of pin pricks which penetrate the surface. Normally Acrylic exhibits the phenomena of total internal light refraction, that is the light which enters via a thin edge, is reflected back into it's volume and will not emerge from the face perpendicular to the point of entry, unless a surface imperfection allows the light to do so. The light emerges only from the penetrating surface features. In effect it is a light panel with many thousands of light sources to spread the light over the entire surface. The acrylic panels act as a backward projecting light system using a a white plastic sheet as the rear reflector on the back face of the stack. The light is therefore projected forward back through the stacked acrylic panels onto the first Fresnel lens of the projectors system. The edges of the stack is encased in white plastic reflector and a housed in aluminium sheath ( reformed lager can) to block any other light escape. The thin aluminium is annealed and flattened. It can be marked out and cut with sturdy shears. The sheath can be formed by scribing it with a ball point pen and folding to the lines. The sheath is in the form of two slide together halves , matchbox style. The LED/heat sink is attached to one end and the second half of the sheath slides over the assembled acrylic panel.
      > If the multilayer stack does not liberate a successful solution a single thickness of 10mm will be prepared with a face drilled pattern of holes which will act as the illumination feature previously mentioned.
      >
      >
      > --- In diy_3d_printing_and_fabrication@yahoogroups.com, "torben.mogensen@" wrote:
      > >
      > > In LED projectors, concentrating the beam is already taken care of, as you are replacing one LED with another LED.
      > >
      > > If replacing a bulb, you will need a reflector. LedEngin has reflectors to fit their LED's - There is an 8 degree, I think that should do the trick.
      > >
      > > Yes, these LEDs are extremly expensive, but if you are replacing a bulb you would have to spend 100-150 anyway for the replacement.
      > >
      > > The smaller LED (4 elements) will properly also cure the resin, but I don't think you would get any speed improvement.
      > >
      > >
      > >
      > >
      > >
      > >
      > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler wrote:
      > > >
      > > > I don't think a 12 or 24 element LED without some sort of optics (light
      > > > guides or beam splitters etc) to get all of that light in one place will
      > > > not work well at all. Most of the light will go to illuminating the inside
      > > > of the projector and/or you will end up with a blurred image.
      > > >
      > > > That's what I think is the case if anyone want to discuss it. I certainly
      > > > would not spend $200 on one of those LEDs without testing the performance
      > > > with a low power extended (not point) source.
      > > >
      > > > Graham
      > > >
      > > >
      > > >
      > > >
      > > >
      > > > On Wed, Jan 23, 2013 at 1:13 PM, torben.mogensen@ <
      > > > torben.f.mogensen@> wrote:
      > > >
      > > > > **
      > > > >
      > > > >
      > > > > Hi, I'm going down that route.
      > > > > I found that LedEngin.com is the most powerful LED's.
      > > > > They have a LZP series (24 element) and a LZP (12 element)
      > > > > They are quite pricy .. $350/$200, mouser.com have them in stock.
      > > > >
      > > > > They require 21-23V, so you need a separate power supply and a rather big
      > > > > one, as the consume 4Amp/2Amp.
      > > > >
      > > > > If you are replacing a projector-bulb - you will need a reflector also.
      > > > >
      > > > > You need to consider cooling also - they get very hot and loose
      > > > > effectivness if they get over 70-80 degrees C. Specially if you leave it on
      > > > > all the time.
      > > > > I'm planing to control on/off with the microcontroller (my case Raspberry
      > > > > Pi) to reduce heat
      > > > >
      > > > > AFIK you need 400 Nm, 450Nm is just outside the useful spektrum for the
      > > > > photoinitiators I have seen.
      > > > >
      > > > > I made some tests on Fernando's resin with 400 Nm and it cures VERY fast.
      > > > > I have written about this in another post.
      > > > >
      > > > >
      > > > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler
      > > > > wrote:
      > > > > >
      > > > > > Ideally you want something that will match the way the original lamp gave
      > > > > > out it's light so that the optics can collect the light effectively and
      > > > > > project it well. I suspect a single source and no lens might be best but
      > > > > I
      > > > > > have never run a projector lamp outside of a projector, does the mirror
      > > > > of
      > > > > > the lamp collimate the light?
      > > > > >
      > > > > > If it does then the LED could be mounted in an old lamp facing backwards.
      > > > > > Of course if it is multi element then you will actually end up with
      > > > > > multiple collimated beams which may blur the image.
      > > > > >
      > > > > > It would be really nice to have a long lasting bulb replacement so good
      > > > > > luck with your quest.
      > > > > >
      > > > > > Graham
      > > > > >
      > > > > >
      > > > > >
      > > > > > On Wed, Jan 23, 2013 at 12:07 PM, afogassa wrote:
      > > > > >
      > > > > > > **
      > > > >
      > > > > > >
      > > > > > >
      > > > > > > I was just browsing ebay for it and found some multi element 405nm
      > > > > leds in
      > > > > > > the 10w, and 450nm up to 100w there is dome lens for those avaiable.
      > > > > > > just thinking since a replacement bulb is very expensive and sooner or
      > > > > > > later it will blow up anyway, i don´t think the light bulb on this
      > > > > > > projectors will outlast the life expectancy of 5000 hours before it
      > > > > blows
      > > > > > > up by some unknown reasons.
      > > > > > > well, acctually there is reason for mine to blow up, the first one the
      > > > > > > original one crack becouse of power failure, the second one becouse i
      > > > > > > bought on ebay made by chinese grade c manufacturer.
      > > > > > > So, yes I´m getting pissed and about to go back to the stone age i mean
      > > > > > > powder printer.
      > > > > > >
      > > > > > > --- In diy_3d_printing_and_fabrication@yahoogroups.com, Graham Stabler
      > > > > > > wrote:
      > > > > > > >
      > > > > > > > Alvaro,
      > > > > > > >
      > > > > > > > What is the most powerful LED source you can get with a single
      > > > > element?
      > > > > > > If
      > > > > > > > it has an array of 9 like some I have seen then the projection will
      > > > > not
      > > > > > > be
      > > > > > > > very good I think.
      > > > > > > >
      > > > > > > > Graham
      > > > > > > >
      > > > > > > > On Wed, Jan 23, 2013 at 10:18 AM, afogassa wrote:
      > > > > > > >
      > > > > > > > > **
      > > > > > >
      > > > > > > > >
      > > > > > > > >
      > > > > > > > > Since my lamp burned(craked) and the one i´ve bought from china
      > > > > didn´t
      > > > > > > > > last 10min before pop out i´m wondering about replacing the lamp by
      > > > > > > 20w uv
      > > > > > > > > leds lets say 405nm or 440nm.
      > > > > > > > > What wave lenght is being used to harden Fernandos resin? I mean
      > > > > what
      > > > > > > > > portion of the white light is being acctualy used to harden the
      > > > > resin?
      > > > > > > > > maybe 2600 lumens is over kill since we use just the 400 to 450nm
      > > > > waive
      > > > > > > > > lenght? maybe a 20,30w led with the right waive lenght is enaugh?
      > > > > > > > >
      > > > > > > > >
      > > > > > > > >
      > > > > > > >
      > > > > > >
      > > > > > >
      > > > > > >
      > > > > >
      > > > >
      > > > >
      > > > >
      > > >
      > >
      >




      --
      Regards

      Francis Leach





      --
      Regards

      Francis Leach




      --
      Regards

      Francis Leach


    Your message has been successfully submitted and would be delivered to recipients shortly.