Mega Infinity Mirror

I’m hoping to create a mega infinity mirror (~1.8m x ~1m).

Will this work with a width/height so large?

Any thoughts on how I could cost effectively do this + have it ‘tamper’ proof if left out (i.e. not glass that can easily be smashed)?

The lights will only need to be visible at dusk/night.


Hi Scott,

Definitely would - I’ve seen some pretty large ones. If you look up infinity mirror tables there’s heaps, here’s one:

If you want it really tough, you could make it out of Polycarbonate sheet - might be a bit pricey though. If you go 1200x1800 this is a common standard material size for sheets of anything so you might even be able to save yourself from having to do any cutting.


Thanks @oliver33

Seems like Polycarbonate should fit the bill with any non-delibrate damage (short of scratching it up with a knife). But yes, expensive, at least online. I’ll make some calls.

Could I just use a Bunnings mirror film and apply it to some wood and it will act as a mirror for the backing? Might be more economical than sourcing an actual mirror to then be backed anyway?

Also, do you get a better effect if the two mirrors are closer or further apart from each other?


Hi Scott,
Welcome to the forum :smiley:

I’ve put far too much thought into infinity mirrors during my build of an infinity mirror hypercube, I’ll have to grab some photos of it when I get home.

@Tim has also gone deep with the infinity mirror effect on his LuMa 1 disco ball, so it seems we at Core have a weakness for infinity mirrors.

Regarding materials, you should be aiming to maximise reflection while also minimising distortion.
Reflection doesn’t have to be perfect, mirror window tint from Bunnings over some 3mm clear acrylic panels was enough for me to get an illusion about 8ish reflections deep. This is with cheap mirror film on both sides, for a 2D infinity mirror panel one side doesn’t need to be partially transparent so you can get even better reflections.

Distortion is a little bit trickier to manage, if your mirrors aren’t perfectly flat (they won’t be) then you get a slight warping of the reflection, sometimes it’s called the funhouse effect.
Tempered glass will introduce a lot of funhouse warping to your reflections whereas standard glass does not. Polycarbonate and Acrylic sheets are probably somewhere between those two extremes.
Given you are trying to go as large as possible this will likely be your biggest hurdle to overcome as smaller setups like mine and the LuMa 1 are too small to notice this warping.

For the cost of the window mirror tint, it is absolutely worth experimenting with to see if you can get the results you need. Applying it without bubbles is a bit of an art form in itself but once the sun goes down and the lights come on a lot of your mistakes are hidden anyway.

If the warping with the mirror tint is too much, then you will need to look at sheets that come fabricated with mirror tint already on them, this is about where I got distracted by other new and shiny projects and haven’t yet returned to my next infinity mirror upgrade.

This just changes the effect, you will be somewhat limited to how many reflections you can get, so a bit of spacing does make it look deeper with the same number of internal bounces.

PS. If you decide to go 3D one day there are some very cool builds out there.


Funny you should mention cubes/3D shapes… I’m considering one now!

I was thinking laser cutting with finger joins with 30cm sides as that seems economical with Bunnings 600x900x3mm acrylic mirror sheet. Would that be strong enough or should I print a structure for the acrylic to slot in to?

Full disclosure: it’s for a Christmas light display :joy::+1:


Hi Scott,

I got home and totally forgot to get some photos of my cube, I’ll set a reminder to take some more.

I’m not sure what you mean here, acrylic does an ok job of supporting its own weight when it’s oriented straight up and down, if it’s mounted parallel to the ground though you may find the middle bows or sags and warps your reflections a bit leading to a bit of a fish-eye distortion.

Again the biggest acrylic I’ve used is 165mm squares hot glued into a frame so I’m not sure about supporting a larger sheet as acrylic is pretty heavy.


Something like this?

I’ve made large finger jointed plywood boxes but they were 9mm thick rather than 3mm or 4.5mm so I’m not sure if it’ll work with those thicknesses.


Ahh I see, thanks for the photo, again I don’t have much experience on that scale so I’m not sure how well it would work.
I’m not sure how you would estimate or measure the amount of bowing and sagging that occurs outside of building it and checking after the fact.


Hi Scott,

With 30cm spans bowing shouldnt be an issue. I found an offcut laying around in my material pile and it bowed about 1mm.

The peice was 230mm wide(into the box) and spanned 300mm.
For one of my projects I had 3 fastening points (M4 bolts and there was quite a bit bowing when you put a force on it, keeping it supported on all 4 edges will help a ton though!

EDIT: Excellent point! Take a look at @Oliver 's calculator instead for a much more accurate calculation - left this here for any other keen makers :smiley:
As for estimating you can use a deflection calculator much like the ones civvies use on beams: Beam Deflection Calculator
Most calculators will spit out a pretty low number though!

I’m keen to see it done!


Actually for a plate, you’ll be better off with a plate deflection calculation:

For Cast Acrylic, the Young’s Modulus is about 3.0GPa, Poisson’s ratio is 0.37 (MatWeb - The Online Materials Information Resource)

For a just point load of 1kgf, I get 0.67mm for Liam’s setup.

Here’s one for a uniformly distributed load (Like accounting for the mass of the plate)
Though it looks like they charge for it :frowning:
But the formula for deflection is here: Loaded Flat Plates - Roy Mech (y_m = …)

Density of plate is 1.185g/cm^3, for a 3mm plate is 0.3555 g/cm^2.
Under gravity of 9.81 m/s^2 gives a load of

(0.3555 g/cm^2 * 9.81 m/s^2 * 0.001 kg/g * 10,000 cm^2/m^2) = 34.9 Kg*m/(s^2*m^2)
= 34.9 N/m^2
= 34.9 Pa

Pro-tip, for small deflections it’s accurate to just calculate things separately and add them up, so you can calculate your plate under it’s own weight, then add the deflection for something on top of it - assuming the stress is less then what it would take to break it.