Light Panel Personal Project

Hi guys, I’m starting a new personal project and I’d like to share my development on it and hopefully get some feedback as I go along. I’ve been working on this for the past 2 months at a very slow pace but thinking a lot of the details through during that time. In this first post I’ll be dumping my thoughts and work so far and flesh it out properly later.

First note of caution, I’m a recent mechanical engineering graduate and I’m currently trying to learn as much as possible about industrial design (sketching being the focus right now).

This project started from a problem I encountered. Over the summer, I was doing consultancy work and I was working in a windowless basement. After a few weeks, my colleagues and I came to the realization that we would start feeling groggy in the afternoon and felt a bit disoriented when finishing our work day and going outside on a hot bright shinny day. I also found that a lot of the lighting was coming from point source which would cast harsh shadows making tasks like drawing and soldering harder.

There have been a few studies that would seem to indicate that light might play a significant role in various systems. Some suggesting that blue light in the evening might hamper melatonine production as well.

This prompted me to start thinking of ways to ameliorate interior lighting.

This lead me to some of the reasons why windows are great:

  • Light intensity and colour change through the day.
    These changes are affected by which way the windows is pointing. (Sun rises East)
    The direction in which the light comes out of the window is affected by the time of day.
    Windows are large and typically diffuse the light over a large surface.

From this, my initial thought is that I would need to create a lamp that changes colour and intensity with respect to the time of day. Then this lamp would need to be large in order to provide a nice soft diffuse light. For this reason, LEDs seem like the ideal light source.

As for the form, I had three main ideas:

Something resembling a paper lantern. It’s a known form and would probably have less resistance when coming to market. They are also quite cheap to manufacture. The main problem with these is they need to be suspended and they also have a very specific aesthetic which doesn’t work in all rooms. I also though of a foot lamp with a large shade might work. However, I want my project to look into primary sources of lighting more than accent lights.

A projector like device which would point on a wall to bounce light off of it. The main advantage of this is that it’s a small unit. Other than that, it would be hard to have it work in every room. It would be unsightly if fixed to a ceiling and would be hard to hide there. It might be kind of playful if sitting on a table or shelf but it would be hard to have the light be unobstructed by the time it hits the wall. Another problem with this is that it would be a very bright spot if you happened to look directly towards it.

The solution I settled on is a panel. It would be roughly the size of a windows. Not only would it provide light similar to a window, it would have the vague form of one. I think this familiar form would suit more spaces than the other solutions (it could also be mounted on ceilings). Also since the light would come from walls, it would help with a problem I frequently encounter when sitting at a desk, I cast a big shadow right over my work.

So far, I’ve started building my light panel. I’ve soldered and attached LED strips to board. Half my lights are daylight balanced (as blue as I need to go) and the other half is tungsten balanced for a more orangy colour. I’ve hooked up a simple circuit to drive my LEDs from an Arduino. Right now, I can fade between the two LED colours as shown in the link below. The next big step is to figure out what colour and intensity the Sun is throughout a cloudless day and interface that with code I found to determine the position of the Sun based on time and date.

I know I’ve been light on process in this post but I really just wanted to get the ball rolling. Hopefully, my next update will be on background and process hopefully covering intent, possible solutions, inspiration and ideation for the panel - all done in a visual manner.

I’m not sure how much research you’ve done behind why light does the things that it does. But I am not sold on simply replicating your experiences with window lighting.

In the end of the project I would ask you why it does the things it does. I would require sufficient scientific evidence as answers.
Otherwise it’s just a light that does some things you HOPE will do.

From the general knowledge I have of lighting. The wavelength of light is key. If you want the same effects of day light you need to have the same wavelengths of light that the sun produces. Also diffuse as you had mentioned is great as harsh shadows are hard on the eyes.

Good luck.

Wow, it’s been a while without an update. School and job hunting got a bit in the way (ok, and a fair share of problems). I’m sorry as I haven’t shared much of the process but I finished putting together my first draft of the project for my portfolio.

Presentation wise, I know I’m missing a view that shows the profile of the frame. I’ve stepped away from making health claims as they are hard to back and ultimately controversial. Hopefully the purpose of the product is still conveyed but I might have been looking at this from so close that it’s become too obvious to me.

Anyhow, I’d love to hear your thoughts.

If you prefer to see it in pdf here.

There is a similar product that you might wanna take a look at.

[ Deleted ]

There are some potential comercial uses. I’ve been to a lot of trade shows, big box stores, Vegas casinos conference rooms and offices deep within corporate buildings, no windows anywhere.

I’m not getting what’s the difference between this and normal flat led panels. I gave those in my loft hallway and looks the same.


This is a good resource.

That’s a pretty neat product. The lighting colour might be a bit weird if you don’t have white walls though I guess it could come with some sort of white backing.

The idea would be to have it follow roughly the light pattern of the Sun but take some liberties to lower the difference between the brightest and and darkest settings so it still stays practical. And instead of going to a very blueish twilight colour at night, stay orangy past sunset to promote melatonin production.

I might need to add a page on ways it can be used/programmed. Ultimately, it’s a large variable light source that’s connected to the Internet. It could be useful in different scenarios. There’s a general purpose setting I mentioned above. In a gallery/studio it could avoid mixed lighting colour reproduction problems by accurately matching the colour temperature and intensity of windows. It could also be used in many light therapy applications. Provided, the LED provide a wide spectrum, it could possibly be used for seasonal affective disorder by providing light closer to a light box during treatment hours. It could also be used for circadian rhythm treatment (including jet-lag and shift changes) by providing bright light on wake. And lastly, it could sync with your phone’s alarm to provide a dawn simulation.

Yeah, there’s something to be said about good lighting. Just yesterday, I shopped at a few big box stores and one of them had a big skylight in the middle of the store. It made a nice difference on the mood compared to the others.

Well there’s not all that much, this light changes colour and intensity and is connected to the Internet. I guess adding that page I mentioned earlier on use cases would help with differentiation. Going with a panel seems like the logical choice. It has most of the advantageous characteristics and makes sense from a feasibility perspective. It just doesn’t make for a very sexy portfolio piece as it’s the obvious solution and it’s very similar to existing products.

I think I’ll need to add some mention of light therapy, I was afraid of falling into snake oil territory. Wiki provides a nice trail of reputable references to include. :slight_smile:

Thanks for the feedback! I’ll be adding a page on different modes and their purpose.

So similar to what f.lux does.

I’ve appended a mode of operation page to my earlier post. Does this make things clearer? Any notes on the presentation?

“The panel is framed in light maple.”

Why light maple?

Why does the frame need to be visible to hold the Tyvek diffusion panel?

Why are the frame joints so visible?

What is the power supply?

What does it look like when it is off versus when it is on?

Any lightly coloured wood would work, I happened to have access to maple. I thought darker wood would make a big contrast next to the bright light and draw a bit too much attention. The light colour also makes it feel contemporary. Similarly with a larger frame, it would also start to look like a picture frame with a blank canvas, I guess there’s something depressing about that. :laughing:

I tried putting the diffusion over the frame and it just looks weird when the light is on because you can see the outline of the frame. I guess a clear plastic frame would solve that problem. But I personally like the frame as it gives the product a bit of character, it might be too inconspicuous without.

My average woodcrafting skills and tools made the frame joints visible. My miter saw doesn’t cut true when at an angle. I had originally mitigated the problem when assembling the frame by choosing which piece would mate with which piece. That way, one side of the frame has less visible joint than the other. Sadly, I got distracted when installing the tyvek and I put it on the worse side…

Right now, on the prototype, the power supply is a 12V 3A switched-mode power supply. I believe it used to drive network equipment in it’s former life. In a production model, I think a similar power supply could be used and integrated inside the panel. I see two ways the panel would be installed. One where the light is connected to a wall switch, so the light turns on as a standard lamp. A very small rechargeable battery could be added to allow the microcontroller to keep up to date for a few days when the light hasn’t been on. A second mode of operation were the panel would always have access to electricity and the panel would be programmed to automatically go on and off or accessed via a smartphone app or even integrate with a larger home automation system.

The first image next to the door gives an idea of what it looks like when it’s off (it’s on in that picture but next to a very large window). The tyvek is a bit reflective, kind of like white satin sheets but with a bit of texture to it. I’m not sure tyvek would be the best choice for a production model, the type I have shows the fibres in the fabric as light passes through it.

Don’t let your personal prototyping skills limit/impact your design choices. Design it exactly how you think it should be. If you cant make it, then photoshop it so it shows your true design intent.

Design first. Model second.

(PS. I never noticed that was a real model. looks like a render.)

Are you designing the installation process/experience too?

Could be fun as a system of circles . That way it could feel more fun, more like light itself vs an imitation of a window.

quick sketch, maybe more like this.

I haven’t fully designed the installation process. I think it would mainly depend on some market research. Also, would there be several models available for different situations like ceiling mounting, built-in installations. I’ve thought of the general installation but I haven’t focused my energy there. I’d also need to design the smartphone app to make the installation experience complete.

That’s roughly what I did. I didn’t use wood simply because it convenient to prototype with. I think it has a place in home decor. I’m a bit ashamed but here’s one of the original pictures, prototype board, light leaks and crooked frame included. :laughing:

On my last projects, I’ve focused on bringing the ideas to a working proof of concept prototype. I think I leave little stones unturned by doing so. I also enjoy all parts of bringing a product to life, and I can put it in my engineering portfolio where I’m currently seeking employment. I think I’ll be separating both a bit more in future projects, or possibly stick to models. Focusing on styling, brand identity, design research, sketching, surface modelling and rendering on one side and nitty gritty calcs/FEA, parametric modelling and embedded programming on the other.

I completely agree. But I also completely agree with Sain, don’t let your limited production capabilities drive the design.

In your presentation you should clearly state the model is a proof of principle and in addition to that, show images of the final design that are not influenced by your “skill” of manufacturing.

I asked those questions because I thought your choices were quite arbitrary. Your explanations help, but I am not fully convinced of your choices. Especially with the wood frame. I find it distracts from the concept. I am focusing on the frame, not the light. Make me focus on the light. Same deal with a cord coming out of it, total distraction.

Thanks for taking the time Michael! :slight_smile:

I had considered going with a circular shape. Part of my reason not to go with it is that I didn’t want to have something too similar to Rich Brilliant Willing 's circular light which got quite a bit of hype recently. But I have to say I like the shape. It feels a bit more like the Sun and makes it standout and fun. I could really see the constellation idea work in a large board room or entrance desk area.

While a bit more out there, a skylight that could pulse to simulate clouds going over head would make for a cool architectural feature.

you could do an interesting system of polygons that nest as well. Lots of options. Now that you have selected the wall form factor blow out all the iterations and explore. You prototype is still a good proof of concept.