AVS Forum banner
1 - 6 of 6 Posts

1,484 Posts
Discussion Starter · #1 ·
Those of you who have followed me on my build thread know that I just dive into new things head-first and hope I learn as I go. Building my screen was no different. And yeah, I made more than a few mistakes along the way but it ended up being pretty darn awesome! :D

It's a 10' x 5'-8" (138" diagonal) 16:9 spandex screen. This is my journey.

Frame Orientation

I've subscribed to a few spandex screen threads for awhile and from them, I assumed I was going to go with a frame built "flat". It wasn't until a week before I built my frame that I realized that most of the "flat" designs were from a few years ago and the majority of the newer builds have the frame built "on edge", similar to a canvas portrait frame. I made a last minute change to my plan to be on edge for no other reason than that's what was the current predominant design.

I'm not 100% convinced that it's the best way to build a frame. The stresses on the frame tend to be on the X and Y axes more than on a "curling" stress. As such, the thicker pieces in a flat configuration would provide much more bending resistance than one on edge. The on edge design does have much better curling resistance, but I don't think that's a major factor with a spandex screen.

All that said, having a thinner piece of wood directly behind the screen can only be a good thing. I could go both ways if I did it again... but yeah, I'm not really convinced that either design is inherently better.

Woodworking, Not Construction

A number of frames these days are built with 2x4s. All y'all doing that clearly have a much better lumber yard than I have access too since the chances of me getting two straight 10" 2x4s is approximately zero. Instead, I decided to try and wake up some of my latent woodworking skills, which had gone fully dormant in the past few years of me doing almost exclusively construction. As such, I designed the frame to be made out of 4/4 hardwood, with no mechanical fasteners at all -- just proper joints and glue:

I normally get unfaced 4/4 lumber and mill it myself, but because of time constraints, I just got some s4s poplar from Home Depot. I debated getting Oak, since it has significantly stronger bending strength... but it's also twice the cost at my local HD and I don't need bending resistance that much.

I started out by ripping all of the pieces to 2-1/2" widths so that everything is exactly the same:

I then stacked up the similar pieces and cut them all at the same time to ensure the exact same length:

And... that's pretty much where I stopped doing things right. Any experienced woodworkers reading this might want to avert their eyes from the carnage that is to follow.

Not Like Riding a Bike

My very next step was to cut out the notches in the top and bottom rails to accept the cross member supports. It was at this point that I realized that woodworking skills are not like riding a bike. If you neglect them over time, then you will absolutely lose the skills.

Case in point, here's how I made the notches:

That might look fine (top and bottom pieces notched at the same time; router with pattern bit on template) but what is not visible is that I didn't use the physical cross member to set the width of the template. Instead, I did it the "construction" way and just measured it off with a tape measure and marked it with a pencil.

And truth be told, this was the second set of notches -- for the first set, I did it using a straight-edge guide rather than a pattern bit. That was far worse.

The end result is that the notches are too big by just over 1/16". That might not sound like much, but remember that my goal was to have only glue and no mechanical fasteners at all. With that much of a gap, that's not possible:

So I made my first concession and attached the piece with some 1-1/2" brads.

And then I got worried that that wasn't enough and put a screw into each joint as well. Bleah.


Next up was the corners. I wanted to create a half-lap joint, of sorts, only on edge instead of flat. There would be a large amount of glue surface in a joint like that, so it should be plenty strong if done right. The pieces look something like this:

What's maybe not obvious in the picture is that I forgot another cardinal rule of woodworking -- use precision tools when making precision joints. In this case, I used my jig-saw... absolutely nobody's idea of a precision tool. And yeah, as a result the cut isn't perfectly square or flat or anything that is really needed for a quality joint.

The top two joints at least didn't look terrible, if not very good:

I conceded to use some brads to hold it together since I don't have any 10' wide or 6' wide clamps and didn't feel like coming up with a workaround if I had already introduced some mechanical fasteners into the mix.

I didn't like how uneven the joint was, but there was enough glue and enough surface area to think that maybe it would be okay anyway.

Then I saw my next fundamental mistake -- forgot to dry fit before gluing!!

I accidently flipped the bottom rail around and ended up having two "posts" in the same position rather than one being on top and one being on bottom. To get them to fit at all, I had to cut off one of the posts.


That's not even a little bit strong enough. A simple end grain butt joint covering only half of the piece? Terrible.

So I completely surrendered my notion of woodworking purity and went out and bought four corner braces:

My failure was complete.

At least this showed me that I really needed to relearn a lot of my lost skills and it pushed me into building my screen wall panels with more care. Some of it is slowly coming back to me.


I primed the entire frame with a tinted Killz PVA primer.

After that, I painted the frame a flat black... of which I have no good pictures. All of the pictures I took of the frame turned out blurry and with the various blacks of that plus my theater smeared together. So picture the above, but in black.

On that note, it turns out that getting a pure black from Home Depot is pretty easy now. I heard about the old-school "Mouse Ears" black, which is no longer carried, but that's not needed. I just told the guy at the paint station that I wanted the blackest black he could make and he instantly got me a sample card labeled just "Black", which is 12 oz of black colorant in a deep base and nothing else. It's in the computer so no convincing was needed. Perfect.


1,484 Posts
Discussion Starter · #2 · (Edited)
Building the Spandex Screen

I have a Panasonic PT-AE8000U and the standard recommendation for that projector is a layer of Silver Matte Milliskin under a layer of White Matte Milliskin. I ordered both from Spandex World.

Now, the conventional wisdom is that if you order 4 yards or less of spandex from Spandex World and do it online, then they'll ship it to you in a box. A box that will wrinkle the fabric. I didn't want to deal with wrinkles, but I couldn't seem to find a time to call them. They have standard business hours on the East Coast and I'm in AZ so that three hour difference was killing me.

I eventually just ordered online but I did two things in an attempt to convince them to ship it on a roll. First, I ordered 4.5 yards of each rather than the needed 4 yards. I figured that the extra $9 would be money well spent if I got it on a roll. Second, I added a note to the order roughly like so:

"Please ship the spandex on a roll. If the shipping cost is extra for a roll, then please call me and I'll authorize the extra shipping charge! This is very important to me!!"

I can imagine the puzzled looks on the packagers there wondering just what could be so important about a few yards of spandex, but maybe something in all that worked, because they did ship it on a roll!


Here's what the two colors look like together:

Yeah, it's pretty startling. The white almost looks luminescent even in person, compared to the muted silver. And yes, that's matte white.

The spandex is 13'-6" long, which is plenty to cover my 10' wide screen, but only 60" high. That meant that in order to stretch it over my 68" high screen plus 2-1/2" on each side would require the spandex to stretch a full 12", or 20% of its height.

I decided to staple the spandex to the frame rather than use something like screen tight, if for no other reason than no local HD or Lowes carries it. Plus staples are cheaper and I have an upholstery stapler.

For the stapling pattern, I went off of a few YouTube videos I found on building a canvas frame. The idea is to alternate staples in a top-bottom / side-side pattern.

It starts with the top and bottom center staples:

I'll admit that I was very worried at this point at those very defined creases due to the stretching. I hoped that they would erase themselves when I stretched in the other direction.

Sure enough, after the initial side staples, the first set of creases went away:

I then kept following that pattern, where I'd put a couple staples on the bottom; then a couple on top; then on each side. The staples were spaced roughly 3" to 4" apart, but I didn't measure.

I also followed the advice to pull the fabric slightly to the right (or left) in order to take out any slack as I went. This is as opposed to just pulling straight down. I found that I needed to stretch the fabric more and more as I got to the ends.

Cursed Corners

I was worried about the corners so prior to making the screen, I made a few acoustical panels to practice my corners on. I used the same canvas frame tutorials on how to do that.

It turns out that I needed have bothered. The extreme stretchy nature of spandex makes doing corners completely and utterly different than using fabric that doesn't really stretch. For spandex, you sort of just stretch it around the corner and staple it in place fairly roughly.

In theory.

In practice, I found that I had stretched the spandex too much as I went and ended up with lots of bunching in each corner:

Only one of the corners worked like expected. The other three all had minor to massive amounts of bunching that couldn't be easily smoothed out:

I didn't want to have to take out most of the staples to fix it, so I took a nuclear option -- I just started pulling the spandex in tiny little increments just to smooth out a very small part of a wrinkle and then stapled it in place. And then did it again. And again. And again. I spit staples at it as fast as I could pull the trigger.

It worked. The corners were nice and smooth.

But the sides were an unholy mess as a result:

That's actually the better side. Anyway, you can see how massively some of the spandex is stretched in my zeal to get everything smooth. I didn't want to leave it like that, especially as a first layer.

So what I did was put a layer of staples up higher, maybe 3/4" from the top of the frame, and froze the spandex in place that way. That meant that all of the lower staples were no longer load bearing. That "allowed" me to laboriously track down each of those staples and pop them out. I say "laboriously" because those staples didn't want to come out. I ended up using a combination of a pair of dikes plus a knife and mostly cut them out of there.

In the end, the massively stretched out parts were eased, so it looked fine and wasn't taking up so much bunched up space. I don't have a picture of that, though.

Lesson's Learned

I learned some lessons from that first layer and one of the main ones is that I didn't want to follow a pattern designed for non-stretchable fabric if I'm using very stretchy fabric!

So for the white (visible) layer, I tried to put it on with a little foresight:

I started by just tacking the spandex in place on the two sides with one staple each and then leaving the sides until the end! Instead, I concentrated on a stapling pattern along the length of the frame, on the top and bottom.

I also made a conscious effort to not pull the spandex more than absolutely needed.

It worked! The corners were trivial and there was plenty of slack on the ends to leisurely staple the very very lightly stretched spandex into place.


And here's what it looks like hanging up:

I did have one big concern when looking from behind...

Yeah, that's a distinct moire pattern from the extreme stretching I had to do for the silver layer. I was really hoping that the lessened stretching I did with the white layer, plus the combination of the two layers, would mitigate that problem.

Thankfully yes, it did end up not being a problem at all. Whew! I cannot see any pattern at all in the spandex from the front, even if I examine the screen up close while an image is projected.


1,484 Posts
Discussion Starter · #3 · (Edited)
Pics or It Didn't Happen

It was a bumpy road creating the screen, so was it worth it in the end?

Oh yeah! In a word, that screen plus the Panny projector is AWESOME :D

I did a quick waveform contrast calibration but otherwise haven't calibrated the projector. The following pictures are all done with the lamp on eco mode and in REC709 configuration... by which I mean that the lumens are supposedly quite low at this point. With the stated gain of 0.8 for the screen, one might expect a dim image.


The colors just pop and it's plenty bright, even in contrast heavy scenes:

It doesn't do it at the expense of fine details, either. Here's a very up close picture of a test image:

That looks pretty spot on to me!

And here's a composite image of what the Avengers looks like playing on that screen in context:

Pardon my crappy photoshop skills. That's just one picture of the theater with lights on, with the camera on a tripod, and then another picture of the screen with a movie playing (and theater darkened). I then combined the two pictures.

All in all, I'm absolutely loving it :D


4,081 Posts
That spandex stretching exercise seems a lot more painful than I'd envisioned.

Were the dimensions of the spandex square when you got it ? If so, could you not have just stapled the entire top edge by stretching it in only one direction a predetermined amount, then doing the entire bottom by corners first to establish the horizontal stretch followed by pulling vertical to the known edge of the frame at each point ? I guess what I am getting at is that starting with a rectangular piece of spandex and stretching it evenly over a perfect rectangle frame shouldn't involve pulling in both directions at once.

I have not done it, is why I'm asking. I have seen that diagram of bouncing from side to side for stapling and it never seemed like a good idea to me for exactly the issues you had with the silver layer.

PS. The silver looks beige in your photos. Is that just camera or my monitor or does it really not look silver ?

1,484 Posts
Discussion Starter · #5 ·
That spandex stretching exercise seems a lot more painful than I'd envisioned.

Were the dimensions of the spandex square when you got it ? If so, could you not have just stapled the entire top edge by stretching it in only one direction a predetermined amount, then doing the entire bottom by corners first to establish the horizontal stretch followed by pulling vertical to the known edge of the frame at each point ? I guess what I am getting at is that starting with a rectangular piece of spandex and stretching it evenly over a perfect rectangle frame shouldn't involve pulling in both directions at once.

I have not done it, is why I'm asking. I have seen that diagram of bouncing from side to side for stapling and it never seemed like a good idea to me for exactly the issues you had with the silver layer.

PS. The silver looks beige in your photos. Is that just camera or my monitor or does it really not look silver ?
The spandex was "square-ish" as it arrives. It's hard to characterize it as really square, since the stretchiness of it tends to make the edges a little loose and wavy.

I suppose it might be worth a try to staple one side at a time. That's counter-intuitive to the way I approached it, since I was consciously trying to keep the spandex taut the entire time. Doing it one side at a time would mean that it would be very limp while doing that first side and I'd be afraid that I'd be making my later stapling job harder.

I've only done one screen, though, so I can't speak to that definitively. That's the type of thing that somebody who is using screen tight could really experiment with, since it would be trivial to undo and retry different patterns.

The color is definitely silver in person. On my monitor, the closeup picture of the screen with the egregiously bunched up sides is reasonably close to the actual color. It mostly looks beige in the first few pictures and that's due to the auto white balance of my camera getting thrown off by the extreme blacks and whites.

DIY Granddad (w/help)
24,838 Posts
Stretching Spandex should not be that difficult, but experience works for much, far more so than trial be error. It's hard to have "experience" when it's a first go-around...so explicitly following known directions and not improvising will almost always yield more consistent results and successes. Corners are exercises in stretching, but that must depend upon how equally the rest of the material is stretched around the perimeter. Over streching seems to be the most common mistake made, done so in the hope it will itself correct problems when in fact it was unequal stretching that creates most problems in the first place. This is borne out by the issues Granroth had with his corners.

I don't see what is so difficult or strange in the process where you first do a Top / Bottom center at a specific tension, then match that at Center Length positions. Doing so and then repeating that concentrically around the frame creates an even pull, and when a Corner is Finally reached, one simply pulls out diagonally, on 3 axis from the Corner to "Form-Fit" the Corner, staple it in place,then Trim off the excess material around the Frame.

Really...it's almost always the case that when someone has difficulty in doing something as directed, then the first thing that comes to mind is the the directions are at fault. Everyone has experienced those feelings, (...I know I have...) but in truth, and upon most cases of retrospect reflection, that is seldom the case. All that has to be done is to compare the many successful attempts against the few "OMG...what a mess!" ones and it becomes apparent that the issues lie with the individual, not with the application as a whole.

Now while I am NOT singling out Granroth here, his example is the only one at hand being discussed so it must be used. And...comments that seem to want to dismiss or discredit known methods that have found success always attract like comments...and that sort of thing can lead to many being both confused and misdirected.

Spandex is a forgiving material. If one does not make it work against itself. A good example is using a Pneumatic Stapler. If the Gun is not set to the right depth, and the staples are driven in excessively deep "...completely flat or recessed at all...) then yes...absolutely removal will be exceedingly difficult. Every Staple driven becomes a "commitment" to the process, and if an error is made, it adds difficulty to the equation.

The Over Stretching syndrome is something that must be experienced...and usually a "eyeball" judgement of having reached a sufficient pull to achieve a flattened layer doesn't happen until both Top and Sides have received equal amounts of staples. The solution never requires excessive tension...that only adds to the issue of unequal tension-ing.

Now...that is not to say one cannot find an alternative way to apply Spandex. But one must be careful not to create issues where there should not be any. The Morie pattern seen on the Silver showed exactly what happens when one over compensates. The fabric spaces open up...allowing for the detailing of individual Threads and the spaces between them. But as also noted, the top layer mitigated that issue to a full extent. Had the surfaces been reversed there would have been visual hell to pay.

In any case, to err by being slightly less taunt is better than trying to pull Spandex too tightly...either to compensate for having too little material, or in the attempt to equalize imperfect tension-ing. And lastly, Granroths eventual success still shows how well Spandex can adapt to almost any circumstance short of wholesale ineptitude. And Granroth's effort was certainly far, far elevated above that level. Determination to succeed counts for everything in DIY......but sadly, for all too many the primary emphasis seems to be "easy and cheap" only. Almost in every case where excellence in end results is desired, one gets out of one's efforts what they put into it.

And again, that is evidenced by Granroth's efforts. :cool:
1 - 6 of 6 Posts