Fiber optic Lanterns ca. 1845, 1/96

[lambsbk]

The high power 600 lumen light source was heat-sinked today and the chip stayed cool while running. Tonight I will enclose and test it. In principle it should work but as a back-up I bought an 1100 lumen bulb on Ebay a few days ago.

 

The heat sink after drilling and tapping (I got this from a computer store - it is an old heat sink for a CPU - which they gave to me).

 

I used brass 4-40 3/8" machine screws for the positive side of the chip and silver for the negative side.

 

It lit up very well with power from the wall transformer.

Edited April 19, 2014 by lambsbk

[mtaylor]

Wow.... that's an impressive heat sink. Must be sinking what, 10 W or so?   And it looks bright.

[lambsbk]

10.8 watts Mark. 600 lumens. The chip gets hot so the heat sink is needed to dissipate the heat. It was not that hard to make it from the CPU computer heat sink and it was free. It worked a lot better than the can type LEDs on my test tonight with the optics but it is still not quite what I wanted. The next LED to test is the 1100 lumen 10 watt chip I bought on Ebay for about $2. It is still in the mail. If I can find a simple convex lens I am going to try to focus the light to the optic but that may be a little to far to go. The light the LED produces is cool (no infrared). Very little transference of heat to the optics. It all goes into the heat sink. I'll get a pic up tomorrow of the simple lamp enclosure I made for testing -

 

Mark: Thanks for edging me toward this part of the project. I was going to do it last in the build. But I can see that would have been a mistake as bugs really need to be worked out at this stage.

[mtaylor]

Dave,

Are you enclosing the LED in something?  Maybe Aluminum (aluminium) foil (shiny side in) to help?  The few times I've worked with fiber optics, we had a dedicated LED for each strand and special fixture that connected the two. A royal pain to say the least.  Maybe a plate with holes over each LED in the array might work and help with the focus..  dunno'.

 

I hope you'll turn this into tutorial for the articles section. Excellent work and research.

[lambsbk]

I am enclosing the LED in a housing that focuses the light toward the optic fiber. I made it from some scraps around the house - a little brass and copper. I had some brass micro-tubing left from the camboose floor so I put that to use in 3cm sections to guide the optic fiber straight from the 'muzzle end' of the housing. It is just a mock up right now but the principals should not change.

 

The experimental mock up. Crude but effective.

 

The housing from the bottom.

 

The brass micro-tubing on the optic fibers.

 

Together for now with rubber bands! Obviously - still a work in progress.

 

[lambsbk]

This pic is with the lower power light. I am still waiting for my 1100 lumen bulb in the mail.

 

 

[mtaylor]

The low power looks good.  I was surprised on my Constellation how little light it takes to bring things to life.

[WackoWolf]

I agree with Mark, the lower power lights look real good. This is an interesting tread and I am enjoying the way you are testing everything and showing us the how they work out. I will be waiting for the next test to be shown.

[lambsbk]

I got the 12V, 10W, 1050mA, 1100 lumen bulb today and I am MUCH HAPPIER with it. Not super bright but the lantern light is well simulated at this lumen level. Cost - $2 on Ebay. So far the heat sink is draining all of the excess heat from the chip so no over-heating is occurring. As expected on my Connie build there is light bleed through between the Spar deck and the bulwark but I have already figured out how to eliminate that problem.

 

http://modelshipworld.com/index.php?/topic/444-uss-constitution-by-lambsbk-–-revell-–-plastic-–-196-a-fresh-build/page-1

 

The fiber optic I have been using is still at full length (i.e. I have not cut them back yet - and won't until the ship is permanently mounted) but as they are cut back so will light loss (which is proportional to the length). Adjusting the voltage down will decrease the lumen output if it becomes too bright after trimming them.

 

It is time to move from the experimental housing to something more permanent. Over the next few weeks I will probably have that solution posted. For those of you following this thread thanks for your encouragement and comments.

Edited September 11, 2013 by lambsbk

[vacotton]

Hi Dave,

What are the differences between EKG and vernier calipers?  I am considering getting a caliper for the lamp project.

Verne

[lambsbk]

The Vernier caliper is good for sizing objects. To me it has not been as helpful on my build as I thought - although my pair is not the finest quality. The EKG caliper is similar to a compass in action spreading equidistance as they move about a common central pivot. I suppose they are similar to a ship's navigation caliper with a pin on each leg for charting distances. Mine are pictured in the beginning of this thread. It is in the items needed pic as item no. 1. 

[vacotton]

Dave, that makes perfect sense.  In close quarters, verniers probably wouldn't work to well.  I will pick up a pair of EKG's (sans chart  ) soon.  The square clear stock is on its way.

 

Now, I followed you fine until you got to the array on the heat sink and then you lost me.  I get the part about your power source:  110 AC from the wall, through the transformer/rectifier with 12V DC output.  But how did you hook up the LED to the chip?  Then it looks like you fastened a shower curtain housing on a board but that could be anything that allows the bundle of Fiber Optics to be connected securely and perpendicularly to the LED light source.  Is this amenable to battery power with 9VDC?  And finally, I read that LED's require resistors in order to vary the color and flicker.  Is all of that taken care of in the IC on the chip or would that be another step that you have not covered yet?

 

I am getting pretty excited about mastering this approach and I want to understand it completely.

Verne

Edited September 15, 2013 by vacotton

[vacotton]

Aha moment:  The yellow field in the middle of the GG array actually illuminates.  All you have to do is capture that light without allowing it to escape.  Very elegant.

Verne

[lambsbk]

It is a beads LED array. 9v probably would drive it but it is cranking out 10 watts - a battery will not last very long at that power.

 

The 'housing' is temporary and is being used to experiment. Think of it as a projector focusing toward the bundle of optic cables. They are 2 to 3 inches away from the actual LED chip since the chip generates heat and will melt the optic fibers if they get too close.

[vacotton]

Okay, I understand now.  Now to Ebay for some shopping. 

Verne

[CaptainSteve]

Dave !!!

 

A brilliant thread, seeing as I'm only just embarking on my own cruise with Connie and - from the beginning - had planned to light her up.

You clearly know what you are talking about, whereas I lack any technical training in electronics. I'm anxiously waiting to see your finished work.

 

Initially, I had planned on using LEDs, along with a Flickr3 circuit-kit which I had purchased, for my build. However, now I am seriously considering switching to optic fibre. This brings up a question: Will my Flickr3 still work with optic fibre ??

 

One concern.

Seein' as I am working with wood for my own build, then I must confess to being a tad TERRIFIED of any significant increase in heat !!

 

Finally, a suggestion which you may/may not wish to consider for your overall design ...

Primarily to reduce power usage, you could possibly incorporate a light-level detector which would switch the rest of the circuitry (lights/flickering/etc) on only when it was night-time.

[lambsbk]

Thanks for the Kudos Steve. I also looked at the flickr circuits and almost bought a 6 channel version...but I did not want to mess with 6 different LED sources. I also realized the flickr effect was probably too strong for enclosed lanterns so a steady light output was used instead. The transmitted fiber optic light is not bright...yet. It is also cool at the lantern so heat is not an issue. However if you need a brighter light source the 'in ship' mounted LED could work. The LED's will need to be reshaped and tested but it is feasible. The fiber optic is nice because it will never fail since light is transmitted along the fiber and the source is external to the build. But it is challenging hiding the fibers and, at least the plastic fiber I used, is not 100% efficient at light transmission - in fact I would say it is only about 50% efficient - but I hope to improve that ratio by shortening the cable at the end of my build and by improving the housing (incorporating a focusing element - hopefully cheaply). That now seems more than a year away though. 

[lambsbk]

Steve: I also wish that I had incorporated a few more deck lanterns - I used 5 per side but really 7 per side would have been better - but it is a lot of cables to pull through the keel.

[Crowler]

Dave

 

I am amazed at your skill and creativity on your build. I am on my 3rd 1:96 USS Constitution build (Slow learner) and put rice lights in my 2nd one. Worked pretty good, I placed 10 rice lights on the ceiling of the Spare Deck to light up the Gun Deck, hope that makes sense, new at this. I love your idea and the accuracy of the lanterns and how you put the fiber optic cable in so it won't break! I am looking forward to the rest of your build and thank you for posting all your hard work for us/me starters!

 

Steve