Creating a community for DIY LED lighting stuff (DIY Quasars inside)

T

TimTom

Member
Hey guys,

lighting can be expensive, so I started with Yuji LEDs, 12V Power supplies and controllers to get flickerfree, dimmable, color temp adjustable, high CRI lights.

I found a factory in china that sells housings that can contain 2 rows of LED strips and voila: DIY Quasar tubes with a high CRI.

Also I convinced a buddy of mine to build me a controller that is basically like one of these smaller Litegear dimmers, variable color with one dial while keeping a consistent output and all sorts of modes like fire mode, paparazzi etc. It is based on an Arduino, we thought about giving this thing to the public.

Have a look here, it is a quick demo: https://www.youtube.com/watch?v=RNrEO48Zalc

The attached pic shows a test I did with my DIY lights. the key is a 100W Yuji chip in a bowens mount housing with 90cm softbox (basically a super cheap Aputure 120D with Dome alternative), the scratch light on the camera right is a 2ft bicolor DIY tube light set to roughly 6000K.

insta_1.2.1.jpg

What I would love to do is create a community for DIY loving cinematographers, gaffers, filmmakers. For example, I just discovered the Ikea SPÄNST tube lights and asked myself if they could be modded to make them dimmable. They could be dope, cheap practicals. I imagine a databse of LED chips and strips with information like output and color rendition so people can go to ebay, alibaba or whatever and just get going.

Id love a database of links, guides and solutions surrounding lighting, LEDs, circuits, creative solutions, hacks.

Any ideas on how to get such a community going?
 
Last edited:
If you build it, they will come! I would love to see something like this get started. For now, I think this thread is as good a place as any? I feel like there are a number of us lurking out there.

I've been experimenting with the 100w Yuji's myself. This is a prototype for a permanent install that I built using a cheap PAR can as an enclosure with a riveted mount that supports a large 180cm umbrella. Looks absolutely ridiculous but it's small (less than 9" long) and hidden behind the umbrella so no one ever sees it.

IMG_3967.jpg

I only used simple dimming control via a potentiometer -- would be interested in your controller box if you decide to share. Similarly I'm happy to share a parts list if anyone wants to get started experimenting with building these.
 
Yep, I've got every item listed from that video sitting in a box waiting for some down time to play. The little round 19mm LED is most interesting to me, but I've yet to see anyone actually MOUNT it to something.

I want to build a relatively "hard" light, because there's way too many super soft sources nowadays, and as a DP friend of mine was fond of saying;

"You can make a hard light soft, but you can't make a soft light hard"....
 
Yes, I like this topic. I think building the soft lights using LED tape are pretty straight forward. But when it comes to hard lights where heatsinks, cooling, and packaging are more of a challenge, sharing information would be great. Someone posted their DIY 3D printed housing sometime back. That offers a lot of possibilities. I don't know how many watts you can stick in a plastic housing, but if people start designing housings and we can just send the files to a 3D printer, we could have some nice things. Maybe plastic isn't practical, but if it is, I can make 3D drawings suitable for 3D printing. What I don't have, is the inclination to experiment with the 50-150w COBs. If the community comes up with the electronics and ideas they have for packaging, I'd be willing to work on drawings for them.
 
I've purchased 3 borosilicate fresnel lenses - listed on Ebay as for the Arri 300. I also purchased 3 ready made barn door assemblies for the same. I hope to purchase some Cooler Master Heat Sink Fan units, the ones with the copper heat pipes. Then at some point I'll be purchasing the drivers, and then yes, the small diameter Yuji 100W leds. The cost is too much to purchase everything at once, so I'm piecing things together slowly, as funds and time allow. I've followed all the above mentioned threads and links, and look forward to any new DIY information that becomes available.
 
Paul F said:
I don't know how many watts you can stick in a plastic housing, but if people start designing housings and we can just send the files to a 3D printer, we could have some nice things. Maybe plastic isn't practical, but if it is, I can make 3D drawings suitable for 3D printing
Click to expand...

If properly cooled/ ventilated, I think plastic would be ok from a heat perspective. For reference I ran one of the 100W Yujis at full strength for 2 hours and the hottest point on the housing was only warm to the touch. My thermopen registered 98F. Cooling via a Noctua CPU fan. Looks like PLA gets soft around 140F so seems ok.

Having never worked with 3D printed materials before my questions of practicality would also extend to durability / tolerances / ease of modification. Can you screw/rivet into it? I think what the Hive guys are doing with their C series lights (simple enclosures with a lot of options for accessories and modifications) is excellent, and if those sort of things are possible via printing, that would be great.

Another useful application would be building a custom chassis that could hold a PSU / arduino controller / small LED readout screen.
 
Quiet Cooling

Would love to hear from folks who have built these things about what they're using for cooling, with a particular aim toward finding some quiet active cooling options out there. I used Noctua L9is in my prototypes and found them to not be as quiet as I wanted.

Here's a link to a reference stress test in a fairly noisy room: https://cloudup.com/ccS0i_hm9So (you'll hear hum from the jump -- the outside AC unit was cranked, noise floor was around 50db to start with). I narrate what you're listening to -- the first :40s are with no correction applied, then repeated but with de-noise running. Lav is a DPA 4060. IMO even after correction this isn't quiet enough for an interview setting.

I picked the L9i because it was small, but if I was doing it over again, I would step up to the DH15 (I use one of these for an audio PC and it is significantly quieter). The idea of using water AIOs around video equipment makes me nervous, but maybe I should get over that. Anyone have success with those?
 
dcoughla said:
Having never worked with 3D printed materials before my questions of practicality would also extend to durability / tolerances / ease of modification. Can you screw/rivet into it?
Click to expand...

The sky is the limit on tolerances and materials since they are now printing with metal. But of course, the question is what is the cost at a give tolerance and material. I'd suggest if there is some first housing people are interested in, I can draw it and submit it for a quote and learn if this is a practical approach.

Screws, rivets should be no issue with many materials.

What do you mean by modification? Redesign and re-print after a prototype or cut/sand an existing part?
 
Paul F said:
What do you mean by modification? Redesign and re-print after a prototype or cut/sand an existing part?
Click to expand...

Modification was the wrong word, I should have said "accessorize". (Was thinking of "light modifiers.")

For me the ideal enclosure design would be similar to the Hive units,

https://www.youtube.com/watch?v=rc_ALi1k6-c

Minimalist enclosure with enough space for the fan/LED, hole for a female XLR on the back, vents, some kind of permanent mount to attach accessories (Profoto/Bowens), and the ability to attach a reasonably strong mount that allows for c-stand / truss mounting.

IMO the big Q is what the best price:performance:noise cooling solution is as that will pretty much dictate the size and shape of the enclosure. Secondary Q is whether to put the controls on the back or have them on a separate breakout box. Pros and cons to both approaches, some of that depending on whether TimTom wants to share his controller tech!
 
Yes, I like that package. If we had a set of guts, we could work on the package. Let's have all the controls on our phone. There is a grade school kid on youtube showing his light dimmer he is running from his phone so the pieces are there and maybe we can figure it out.
 
I'm a little cautious about some of the techniques being discussed here.

The sticky strip approach works to a degree, but there's a couple of quite big downsides associated with the resistive current limiting. Current is controlled essentially by dumping a lot of power as heat, which compromises efficiency to a point where you're not getting nearly as much out of LED tech as you'd hope. Second, if we put lots of rows side by side, it can overheat, due to both the resistive losses and normal dissipation in the LEDS, to the point of being a genuine safety hazard with the higher-power stuff. This can massively shorten the life of your (expensive, high-CRI) LED strip. It isn't easy to tell when this is starting to happen without a colour meter. Eventually it will become obvious that the light is drifting toward the purple as the heat destroys the organic phosphors in the white LEDs. This can happen in a matter of a few hours. Putting three strips side by side risks causing this problem. Filling up a whole board with them, as in the Mazzuca video, practically guarantees trouble.

The other major problem I notice here is that he's using pulse width modulating dimmers which risk causing flicker at any frame rate with rolling-shutter cameras (which is almost all modern cameras) and especially at high frame rates.

I'm sorry to be a downer and I don't mean to clobber Matt as he's a smart and inventive guy, but there are some genuine issues with the approach that's being proposed and people should be aware of that.

It is possible to homebuild simple LED lights which don't have these problems and I have worked out an aliexpress shopping list to do it if anyone's interested.
 
Last edited:
You must log in or register to reply here.
Back
Top