Mirror Experiments

**mechman** · **03-08-09, 06:13 PM**

Mirror experiments is something else that I'm going to attempt to get started this week as well. We all know that using a mirror as a substrate will help reflect more light back. But how much light? And how much paint negates the reflection of a mirror? Also what are negatives of using a mirror as a substrate?

I have a bunch of 12" square mirrors that I will be spraying and rolling. These are 3mm thin mirror tiles. One of the things that has come out and all experts agree on is that using a mirror as a substrate will cause a slightly distorted image. I can believe that this will be worse for people using a 1080P projector. Why does this happen? The light is diffused through the paint and then reflected back to a different location than where it entered. Make sense? Hopefully the experiment will yield some useful information.

Digital Caliper

Scale

**mechman** · **03-08-09, 06:13 PM**

The paint used for both this and the translucency experiments was Valspar Ultra Premium Flat Enamel White or VUPE White.

Rolled

The tray, roller, roller cover, and paint weighed 988 grams before starting. After one coat on a mirror it weighed 982 grams. Meaning 6 grams of paint is used per coat on a 12X12" tile.

The thickness of the mirror without paint on it was 2.72mm. It was supposed to be 3mm.

After one rolled coat it measured 2.93mm for a difference of .21mm. Meaning one coat of paint gives you roughly .2mm thickness of paint.

I used no water with the rolled paint. I used a 1/4" nap roller.

After curing for 5+ days the thickness of the painted coats has settled some.

Paint Thickness

Miiror with no paint 2.72mm
w/1 rolled coat 2.83mm
w/2 rolled coats 2.91mm
w/3 rolled coats 2.99mm

Sprayed

Paint Thickness

mirror no paint 2.72mm
w/1 sprayed coat 2.76mm
w/2 sprayed coats 2.78mm
w/3 sprayed coats 2.80mm
w/4 sprayed coats 2.83mm
w/5 sprayed coats 2.86mm
w/6 sprayed coats 2.89mm

Keep in mind that I did a spray an initial frost coating to help the paint stick better to the glass. Here's what that looked like:

Data coming soon.

**mechman** · **03-08-09, 06:13 PM**

Data


		The amount of light measured (incident) at the screen coming from the projector was 15fL for 100IRE and 3fL for 50IRE. I used both a 100IRE and a 50IRE windowed image to get measurements.

Some photos

On the left is the reference white hardboard panel - to the right is the 1 coat sprayed mirror

shuffled them around so you could see the tags

A Black Widow mirror tile up front!

Conclusions

These readings initially shocked me as I really thought that a mirror would produce a brighter image. That's why I had Harp recreate what I've done and asked him to confirm what I've found. Well, he did! Now the question is why. The obvious answer is that paint is nowhere near translucent enough to allow light through to hit the mirror. That's confirmed by the translucency experiments. The answer to this elsewhere is to add polyurethane. we all know this would be a terrible decision as polyurethane yellows rather quickly requiring new paint on a yearly basis. And you don't want to keep adding coats to that $$$ mirror you bought!

In the past I did some work with a clear protector that showed promise. Harpmaker and custard are also working with other mediums that may offer themselves up as a 'translucency helper'.

Light fusion (mirrors) as it stands now though is busted by the screen mythbusters!

**Harpmaker** · **03-08-09, 07:04 PM**

Quote:

mechman wrote:

One of the things that has come out and all experts agree on is that using a mirror as a substrate will cause a slightly distorted image. I can believe that this will be worse for people using a 1080P projector. Why does this happen? The light is diffused through the paint and then reflected back to a different location than where it entered. Make sense? Hopefully the experiment will yield some useful information.

Right, the thicker the mirror the worse this effect will be. It could look very similar to "ghosting" on analog TV's. In theory, a front-surface mirror would not have this effect since the actual paint layer would probably not be thick enough to cause it; but almost all common mirrors are rear-surface so the light must go through anywhere from 1/16 to 1/4 of an inch of glass before it is reflected back, and it must again go through all that glass before it leaves the mirror on the way to your eye. The glass forms a sandwich between the paint and the reflective surface.

**Mike P.** · **03-08-09, 07:04 PM**

Quote:

The light is diffused through the paint and then reflected back to a different location than where it entered.

The thinner the mirror the less the image is distorted, correct?

Edit: Don posted the same time I did and answered my question.

**mechman** · **03-08-09, 07:43 PM**

Yep thinner would be less distortion. How I'm going to show this will be interesting. I'm planning on just taking some macro shots with side by side materials - one with mirror as a substrate and the other using the foam board I use as test panels. The thing is with mirrors is that they're so expensive.

**Mike P.** · **03-08-09, 07:48 PM**

Expensive, yes, but the results can be awesome from some of the "test shots" I've seen.

**Harpmaker** · **03-08-09, 09:39 PM**

It may be possible to reuse the mirrors for other test mixes by removing the dried paint with Xylol/Xylene. I discovered it worked well for removing paint from PVC and even tempered hardboard without damaging the substrate. I don't believe it affects plexiglass either, but I'm not sure.

We'll know more after mech does his tests, but as I currently understand it, painting on mirrors is a very tight balancing act; too little paint and the image is too bright and has a very narrow viewing cone; too much paint and the effect diminishes to the point where it looks like a painted wall.

**robert_1967** · **03-09-09, 03:52 AM**

Nice experiments.

I have also experimented with mirrors as substrate. As you said the picture gets "smeared" the thicker the mirror is. But, as in my case with my SONY VPL-HS60, I have rather exstencive screen door effect, so it actually improved my pic from that point of view.

What I did was to use a plastic laminate used for obtaining "frosted glass" look. I would say that this might be a very interseting way to go, if it weren't for one thing. Hotspotting. We need to increase the thickness of the diffusing layer untill the hotspotting disappears. I believe that a mirror substrate screen is easier to get to work if you limit the size of the screen, that is from a hot spot point of view.

I curious thing I noticed. The 3M invisible sticky tape I use for fastening paper and stuff on the glass generated by it selfe a nice picture ! Imagine that!

**lcaillo** · **03-09-09, 07:51 AM**

Has anyone tried using first surface glass mirrors or mylar mirrors like they use in many of the better RP sets?

**mechman** · **03-09-09, 09:35 AM**

Quote:

lcaillo wrote:

Has anyone tried using first surface glass mirrors or mylar mirrors like they use in many of the better RP sets?

My guess is yes, that someone has tried it. They've been playing with mirrors for quite some time at avs. I think the big problem with them is that you can get a commercial solution, that is more than likely a better solution, for a cheaper price. When I was pricing out a mirror 3+ years ago for a 100" screen they were running around $180.

I looked for first surface mirrors at a couple of home improvement stores. But I couldn't find any.

That would be the best solution though.

**Harpmaker** · **03-09-09, 01:19 PM**

I found some very interesting sites that sell mirrored mylar. The stuff is cheaper than I thought, but since it is thin it would have to be mounted in some manner.

The first site only sells 1 mil film, but it is 54" wide and seems very reasonable in cost.
http://www.planetnatural.com/site/reflective-mylar.html

The second site is a bit more expensive, but offers thicker mylar and also has instructions and fasteners for making a portable mirror out of the film. They sell 1, 2, 5 and 7 mil mylar.
http://www.mirrorsheeting.com/

I just searched for "hydroponic reflective mylar" (without the double quotes) and these sites were on the first page of results (I used the Clusty.com search engine).

**Prof.** · **03-09-09, 07:27 PM**

Quote:

robert_1967 wrote:

Nice experiments.

I have also experimented with mirrors as substrate. As you said the picture gets "smeared" the thicker the mirror is. But, as in my case with my SONY VPL-HS60, I have rather exstencive screen door effect, so it actually improved my pic from that point of view.

What I did was to use a plastic laminate used for obtaining "frosted glass" look. I would say that this might be a very interseting way to go, if it weren't for one thing. Hotspotting. We need to increase the thickness of the diffusing layer untill the hotspotting disappears. I believe that a mirror substrate screen is easier to get to work if you limit the size of the screen, that is from a hot spot point of view.

This is how I made my original ultra high gain screen..Acrylic mirror with a clear Matt Acrylic overlay..
The problem with this type of screen (as you mentioned) is the severe hotspotting..
Increasing the thickness of the "frosted glass" type Acrylic, only adds to the image distortion..
The only way to control the hotspotting is to curve the screen, so the hotspot is spread across the screen..But then you will still have some light fall off at the sides..
The other thing is that this type of screen produces very high gains..in the order of 8.0+, which is way too high for most modern digital projectors..
I originally designed this type of screen, to be used with low lumen CRT projectors..

**Prof.** · **03-09-09, 07:35 PM**

Quote:

Harpmaker wrote:

I found some very interesting sites that sell mirrored mylar. The stuff is cheaper than I thought, but since it is thin it would have to be mounted in some manner.

The first site only sells 1 mil film, but it is 54" wide and seems very reasonable in cost.
http://www.planetnatural.com/site/reflective-mylar.html

The second site is a bit more expensive, but offers thicker mylar and also has instructions and fasteners for making a portable mirror out of the film. They sell 1, 2, 5 and 7 mil mylar.
http://www.mirrorsheeting.com/

I just searched for "hydroponic reflective mylar" (without the double quotes) and these sites were on the first page of results (I used the Clusty.com search engine).

Harp..Whilst these films are excellent as a first surface reflector, they are almost impossible to laminate to a substrate without some creases or irregularities to the finished surface..and any flaws in the surface will be clearly seen in the finished product..
And as mentioned above, they cannot be used in a flat screen with the right amount of translucency to the top surface material, as they hotspot terribly..

**robert_1967** · **03-11-09, 03:21 AM**

Hi Prof. Nice to meet you.

I see youv'e done some research. Curved surface would spread the spot, yes, I understand.

But when I say that one could increase the "frosted" thickness, I meant in the sence of increasing the optical difraction. For example I took a sheet of printer paper and saw the same "smearing" effect once I put it on top of the mirror, ableit it was significantly reduced. On the other hand the "gain-effect" almost dissapeared. So I was thinking one could use thinner paper.

Or you are sure that the mirror principle is a dead end?

**wbassett** · **03-11-09, 10:51 AM**

Quote:

robert_1967 wrote:

Nice experiments.

I have also experimented with mirrors as substrate. As you said the picture gets "smeared" the thicker the mirror is. But, as in my case with my SONY VPL-HS60, I have rather exstencive screen door effect, so it actually improved my pic from that point of view.

What I did was to use a plastic laminate used for obtaining "frosted glass" look. I would say that this might be a very interseting way to go, if it weren't for one thing. Hotspotting. We need to increase the thickness of the diffusing layer untill the hotspotting disappears. I believe that a mirror substrate screen is easier to get to work if you limit the size of the screen, that is from a hot spot point of view.

I curious thing I noticed. The 3M invisible sticky tape I use for fastening paper and stuff on the glass generated by it selfe a nice picture ! Imagine that!

I've been smack dab in the middle of this debate many times!

Robert what you just said is probably what was the biggest benefit of using mirrors back then. Also in the day, projectors back then were not nearly as bright as they are now unless a person is using a presentation projector and not a dedicated HT projector.

SDE can be greatly reduced with this method because the light is being diffused and refracted and coming back at different angles and with lower energy. This causes a slight image blur, but usually not so much that it becomes disconcerting to view. It does though cause the pixels to fuzz out and blend together which reduces or even eliminates SDE (depending on the projector and screen size of course). Also is the brighter image on axis, and again back in the day with the lower lumen projectors it was an acceptable trade off for some.

Too much refraction though and we start to get color shifting. This mainly shows up in white areas or when the screen is mostly one color and there are panning shots. The cost vs actual performance has always been the debate. It is by far the most exotic type of screen out there, but also the hardest to make as well as the most expensive DIY option.

**wbassett** · **03-11-09, 11:21 AM**

Quote:

lcaillo wrote:

Has anyone tried using first surface glass mirrors or mylar mirrors like they use in many of the better RP sets?

I know mylar has been tried. It is definitely much cheaper than buying 4'x8' mirror, but it can be hard to work with and get stretched tight. It's also pretty thin and spraying is the only way to go, rolling flexes too much.

Another alternative is aluminum. I have five gallons of Black Jack 5168 and just found a source for more in my area so I'll be up to twenty gallons (hopefully by the end of this weekend!). Aluminum is often used to make mirrors, and it is used as the mirror coating inside CRTs. What intrigues me with this over something like mylar is it's just as inexpensive (maybe even moreso) and can be applied to anything. The big problem though... I seem to be the only person that can find Black Jack 5168!

Rose Brand sells mylar mirrors premade, but they are pretty expensive in my opinion.

I never really was a 'fusion' guy myself, mainly because it just doesn't work the way some try to explain it does. Then again, some people are still stuck in a rut over neutral grays and some are in a big hole trying to make a neutral gray out of red, blue, and green. It can be done, but the real question is why do it? Anyway, those last two were off topic in this thread.

The big question mech and I have mulled over for a couple of years now is when does the effect of the mirror diminish as the layers of paint are applied. I've seen demonstrations of mirrors and yes they can produce incredible on axis gain, but the cone is so narrow that even moving one seat over and the brightness starts to noticeably drop off. It is a very tricky balancing act that's for sure. If it truly is the best of the best, I always wondered why we've never seen this featured in any HT mags out there. They always love the exotic and expensive and I highly doubt that screen companies haven't thought about this themselves as a screen option, but ultimately dropped the idea.

I did run this whole concept past the profs at RPI and RIT and they thought it was interesting but ultimately said there were too many negatives to it and they disagreed with the forum version of how and why it is supposed to work and if it even works in a practical sense.

**mechman** · **03-11-09, 11:29 AM**

Quote:

wbassett wrote:

Then again, some people are still stuck in a rut over neutral grays and some are in a big hole trying to make a neutral gray out of red, blue, and green. It can be done, but the real question is why do it?

I think you should have said that 'it can be done, just not the way they are trying to do it'!

I believe Harp said it would take 2 weekends to fix that mix. I think it would take much less. And don't forget the yellow which none of the Michaels stores in my area carries.

Back on topic. I've got a few things to purchase after work today and then tomorrow will be a day of rolling/measuring/spraying. Hopefully I'll have a start on the results Friday.

**wbassett** · **03-11-09, 11:54 AM**

Mech if I pick up those two five gallon containters of Black Jack... now that the weather is warming up I'll send you a quart. If you spray it on the back of a piece of glass or plexi you'll have yourself an inexpensive second surface mirror! Plus it's a great thing to test out first surface mirror theory with.

Or... people could just use BW, it is loaded with aluminum so it basically has a mirror right in the mix

**mechman** · **03-11-09, 05:21 PM**

Sounds good Bill!

So here's the plan...

Three mirrors for three rolled coats.

Six mirrors for the six sprayed coats. The initial sprayed coat will be a dusting and not be measured.

Thickness measured for each.
Quantity of paint used measured for each.
Reflected light measured for each - both on axis and at 30 degrees. The constant will be a ~10X10" hardboard panel with the same paint on it.
Image quality determination via macro camera shot

Am I missing anything?

**Prof.** · **03-11-09, 07:35 PM**

Quote:

robert_1967 wrote:

Hi Prof. Nice to meet you.

I see youv'e done some research. Curved surface would spread the spot, yes, I understand.

But when I say that one could increase the "frosted" thickness, I meant in the sence of increasing the optical difraction. For example I took a sheet of printer paper and saw the same "smearing" effect once I put it on top of the mirror, ableit it was significantly reduced. On the other hand the "gain-effect" almost dissapeared. So I was thinking one could use thinner paper.

Or you are sure that the mirror principle is a dead end?

Hi robert...

The most effective "frosted" material I ever found was a self adhesive vinyl film..
It's less than 1.0ml. thick so diffraction is kept to a minimum, and you can usually get it in varying densities or translucency's ..
The down side is you cant apply yourself..it has to run through a laminating machine and on an 8'x4' sheet can be quite expensive to apply..

There was a Company many years ago that did make a commercial mirror screen, which was used for external daylight viewing, but there was limited application for it, and I think it died out..

In reality, silver or polished aluminium is a better reflector material, and if you want to go down this path, then these first surface reflector materials is a better way to go..and if you cover these materials with a frosted film..you might have an interesting workable screen..

**Richum** · **03-12-09, 09:27 AM**

How well would Auto Air Aluminum (fine) perform as a mirror?

Perhaps it could be thinly top coated with Bermuda Beige. Since they are proven when mixed (BW), it would be interesting to see the results of the combination in layers.

Unless someone has already tried this I might. I have a little of both left from my BW screen. I should have enough to spray something 4'x4'.

**wbassett** · **03-12-09, 09:41 AM**

Seeing that aluminum is often used to make mirrors, it should work, but would be very expensive. That's why I like Black Jack, I can get gallons of it at next to nothing.

**mechman** · **03-12-09, 10:35 AM**

Quote:

Richum wrote:

How well would Auto Air Aluminum (fine) perform as a mirror?

Perhaps it could be thinly top coated with Bermuda Beige. Since they are proven when mixed (BW), it would be interesting to see the results of the combination in layers.

Unless someone has already tried this I might. I have a little of both left from my BW screen. I should have enough to spray something 4'x4'.

Welcome to the Shack richum!

I wouldn't get too hung up on mirrors. My hypothesis has been that they will do little if any to help vs the problems they create. And from the two individuals above, we're well on our way to proving that. They've already discussed the hot spot problem, which makes sense as the pj's I'm familiar with cannot give you an even brightness across the full image - foot candle wise. And the brighter areas would be more polarized than the less bright areas. Then there's the whole image sharpness issue with second surface mirrors and the translucency of a given mix (which can be a big difference brand to brand).

Hopefully these two experiments will shed a bit more light upon the use of mirrors as a screen substrate.

Before experimenting first surface-wise it may be a good idea to wait until the translucency experiment is finished. I'm planning on painting everything today with prelim measurements tomorrow and final measurements next week after a week of curing.

**mechman** · **03-12-09, 03:44 PM**

The single and double layer rolled mirrors are done. I have only the third coat to roll on the last tile to do and then rolling is finished and on to spraying.

The rolled coats on a 12X12" mirror use 6 grams of paint. For the poly panels, 10X12", the amount was the same - 6 grams.

The mirror thickness pre-painting was 2.72mm. After one coat it was 2.92mm.

When I finish up I'll have all of this in post number two and three above.

Advertisement