Eight Simple Under $20 Lightboxes

This thread is about eight simple, easy-to-make, under $20 lightboxes for estimating relative outputs and lumens. I'll start with the Clorox bottle and post more if there's interest. Am doing this project anyway but thought I'd share because your posts have been so helpful to me. So thank you all for that.

I've always liked flashlights but never spent more than $30 for one until a few weeks ago when I bought two unbranded lights that were advertised to be 1200 lumens. Unfortunately when I compared them to my Dorcy they didn't seem any brighter. So did some research and found this site which gave me the knowledge to make more informed decisions. Now my problem is not overdoing it with all the holiday sales!

Nothing new here... just applying what I've learned as a lurker and trying to make it work. All the typical disclaimers about these not being in the same league as a calibrated integrating sphere apply.

Here are eight quick and dirty lightbox "models" and the flashlights I own so far to compare. Bought a Mastech LX1010B digital 0 - 50,000 Lux Luxmeter on Amazon for $15.57 total, which leaves $4.42 for remaining supplies to keep each lightbox under $20.

Lightbox Models

1. Clorox Bottle
2. 2" 90 Degree PVC Elbow
3. 2" 45 Degree PVC Elbow & Shoebox
4. Room (ceiling bounce)
5. White Plastic 50 lb Icemelt Pail
6. Styrofoam Cooler
7. Milk Carton
8. Paper Mache Sphere

Flashlights

1. EagleTac D25A Mini
2. Thrunite TN12
3. Maelstrom X10
4. Fenix TK35
5. Dorcy 41-4289 200 Lumen Cree LED
6. Maglite XL200

First up is the Clorox bottle, 1.42 gallon size. Thought this would be interesting because it's white (with a slight blue tint) and the sensor can be mounted in the higher neck eliminating the need for a baffle. Might eventually coat the inside with flat white housepaint but for now will go with it as is. Shined my TN12 into the bottle and saw it needed to be wrapped in aluminum foil which I then wrapped in duct tape for durability. Used painters tape to secure the sensor because it comes off easier.

For now will limit my analyses to linear (first order) regression and scatterplots. Here's what I have on this one:

Next up is the 2" 90 degree PVC elbow. I cut out a 2" cardboard disk and in the center of it cut another hole the size of the sensor. Then put the disk on one end of a 2" fitting, centered the sensor on the hole and taped it all together with blue painters tape. Then attached a 2" 90 degree PVC elbow to the other end of the fitting and shined the flashlights into the open end of the 90 degree elbow.

Two big surprises. First, the 50,000 Lux meter can measure over 60,000 Lux. Second, this smaller and simpler model gave similar results to the clorox bottle, but with an even higher R-squared.

Comparing measurements from the 90 degree PVC elbow to the clorox bottle gave the top regression and scatterplot. I noticed two obvious outliers from the Maglite XL200 and Maelstrom X10 which I've labeled 1 and 2. Didn't know what to make of it at first but then remembered how much more sensitive to the position of the flashlight the PVC elbow lightbox was, so decided to redo the measurements for those two WITHOUT looking at the old numbers. In other words the goal was to simply get the best measurements I could, regardless of how they would plot. Which gave the bottom two results. The XL200 measurement still seems high but less so, and the lesson learned is to be more careful going forward with light positioning.

Interestingly, redoing the measurements for the Maglite XL200 and Maelstrom X10 with the 90 degree PVC elbow lightbox gave a better fit with the measurements from the clorox bottle lightbox, but the XL200 measurement still seemed high and the R-squared for the new measurements from the 90 degree PVC elbow lightbox as a predictor of reported lumens dropped slightly. So went back to redo the measurements for the XL200 and X10 with the clorox bottle lightbox too -- again, without looking at the old measurements to bias me.

The regressions in the first column are all with the original measurements and the regressions in the second column are all with the redone measurements for the XL200 and X10 from BOTH lightboxes.

Bottom line is both lightboxes gave measurements consistent with each other and seem to be equally accurate, although the 90 degree PVC elbow lightbox was more sensitive to the position of the light.

Hello luvlites, welcome to CPF. :wave:
That's some very interesting data. I was wondering if you might be able to post a few pics of the setups?

I'm thinking that all the Homemade/Modified folks that would get a kick out of this might be still away for the Holidays, LOL ...

Thanks for the excellent lightbox info and for info on the luxmeter at a low price. I take it you had no problems using it?

Kestrel said:

Hello luvlites, welcome to CPF. :wave:
That's some very interesting data. I was wondering if you might be able to post a few pics of the setups?

I'm thinking that all the Homemade/Modified folks that would get a kick out of this might be still away for the Holidays, LOL ...

Click to expand...

Thanks Kestrel!
Here are pics of the two I've done so far. Please excuse the amateur photography :thumbsdow

Blindasabat said:

Thanks for the excellent lightbox info and for info on the luxmeter at a low price. I take it you had no problems using it?

Click to expand...

You're welcome.

No problems at all, very easy. And I barely knew what a luxmeter was last month.

P.S. My readings were taken by centering the flashlight on the "input hole" (IH) of each lightbox, with the bezel at the IH, making sure the lens is parallel to the plane of the IH so the light shines directly in. Have found it useful with the 90 degree PVC elbow lightbox to make sure the center of the beam hits the same point on the elbow which I'd verify for all but the dimmest beams by looking at the outside of the elbow.

Question:

I got reported lumens numbers for the ThruNite TN12 from their website which gave 705 for the max output.
http://www.thrunite.com/en/ProductView.aspx?id=62

That point is an outlier on the graphs and I just noticed on the flashlight box it says max = 605. That makes the fits better which is not necessarily my goal but it would be nice if that's the correct number to use. Any thoughts?

luvlites said:

Here are pics of the two I've done so far. Please excuse the amateur photography

Click to expand...

Those are totally cool, where's Conte when you need him, lol.