DIY GFRP Sandwich Panel 1x15PR400

[astack]

Your drawing makes total sense to me. Going aroud like that and havig the seams meet sort of half-way past the previous seams should be strong. I assume you'd figure out something similair for the baffle and back, either by layering, or by usimg different folding schemes with a larger piece of cloth/fb, which would I guess be another sort of layering?

I just like the idea how after it's all said and done, it's sort of a one-piece "shell" without any weak spots that could separate or break away easier than the rest.

Would you guys do something similair on the interior? Again more curiosity. No glass experience here.

Doing the inside like that, I have no effin clue. That's part of the reason I did the panel method.

Another way to think about it, is it's like tolexing an SVT. Do a panel on the back, and then wrap the sides, seam on the bottom.

what will you do about handles?? no tweeter??

Handle's a comin. No tweeter. I'm still learning these things, but I think this has enough high end for me.

Don't know if I'd want to take on a FRP build, but I enjoy watching build logs like this one. Thanks for taking the time to document and post.

Since you're an engineer, here's an armchair quarterback question:
One thing I've noticed in every FRP build is internal fillets with filled epoxy, as you're doing. I understand that it's difficult to wrap the glass around a sharp internal radius, but... do you think it's necessary to carry the fiberglass to the internal corners? It seems that would be a low-stress area, and it would save time to simply glass slightly short of the internal corners and edges. That could be done in the flat, before assembly.

Zac2944, nice to see you here. Chime in if you like.

Hmm. If you did the glassing after making the box, that would help I would think. Hadn't thought of it, but maybe that's the way to do it. But doing it before, it would save having to do the trimming, for sure... One issue I see, is the fiberglass cloth inevitably distorts and will have some frays. Given the fillets are only 3/8" or so, that's not a lot of overlap/tolerance. Maybe I don't understand your question?

Speaking of "time savers," at this point, I feel like I have to make a second composite cab in the future just to prove that it doesn't have to take 4 months to build a cab.

 

[astack]

Btw, just did a quick weight forecasting calc.

At the end of yesterday, the box weighed in at 6lbs 8.5oz. I'm estimating another ~3,200 sq. in. of glass, which will be ~3lbs when epoxied. So, the raw box will be under 10lbs, maybe 9.5lbs.

With everything else, it should still be under 25lbs. No bad.

 

[1n3]

My question wasn't put very well. Here's a picture to clarify.

[Invalid or Expired Link Removed]

At top is what everyone seems to do - make a fillet, and lay glass around internal corners. This can only be done after the cab is assembled, so you're working in an confined space.

At bottom is the scheme I asked about, where the glass stops short of the internal corners. This could be done before or after assembly, but I think it'd be easier to do before. There would still be epoxy in the corners, but no glass. My suggested rationale for this is that stress drops as you approach the edge, so the added strength of the glass is not needed there.

I'm not picking on you... this is something I've wondered for a while. Do you think this makes sense structurally?

 

[Passinwind]

At bottom is the scheme I asked about, where the glass stops short of the internal corners. This could be done before or after assembly, but I think it'd be easier to do before. There would still be epoxy in the corners, but no glass. My suggested rationale for this is that stress drops as you approach the edge, so the added strength of the glass is not needed there.

I'm not picking on you... this is something I've wondered for a while. Do you think this makes sense structurally?

IME, the corners often receive by far the most stress in the event of the cab being dropped. Personally, I always reinforce all the bottom edges internally in my cab builds.

 

[astack]

OK, that's more or less what I thought. Great picture, too

In terms of what stresses are happening, it'd be too easy to over think it, for what this is. Like Charlie said, if you think about dropping this thing, or somebody jumping off the top, or any other number of crazy things it would see, there's not a good way to analyze this box.

That said, let me think about this ... If the glass didn't go all the way, I'd be concerned mainly about the panel just shearing straight through, if somebody jumped on it (as if it cookie-cuttered itself, if that makes sense).

Also, yes the corners are much stiffer than the middle, but that also means they tend to attract higher forces.

Lastly, there can be stresses acting in any of 6 directions (3 push/pull pairs and 3 shear or sliding pairs). Often they tend to trade off. So, yes there may be low stress in the direction you're thinking of, but the total force has to get through somehow, and that's usually through a stress in a different direction.

And, having done it the other way, I'm not sure it would be much of a time saver. The panel that mates on the outside, should have the glass go through at least, too.

 

[Passinwind]

OK, that's more or less what I thought. Great picture, too

In terms of what stresses are happening, it'd be too easy to over think it, for what this is. Like Charlie said, if you think about dropping this thing, or somebody jumping off the top, or any other number of crazy things it would see, there's not a good way to analyze this box.

That said, let me think about this ... If the glass didn't go all the way, I'd be concerned mainly about the panel just shearing straight through, if somebody jumped on it (as if it cookie-cuttered itself, if that makes sense).

I've dealt with scary shearing forces and mechanical hinges a lot in sailboards, where your body can easily be the weakest link. At some point with composites you have to settle on the most acceptable failure mode, and that often means dialing in some flex somewhere in the system. Deformable corners are not necessarily a terrible thing.

 

[zac2944]

For mounting try bonding 1/2" ply on the inside to distribute load and accept mounting hardware. I use a small ring behind the speakers, handles etc. You can use regular t-nut or just wood screws. Much easier than bonding tnuts to epoxy/glass and and adds very little weight. Can be used in corners too.

 

[astack]

Great points guys.

I may end up doing the plywood thing. Seems sacrilegious for a composite build though... I know Robby did it... You're right, weight-wise it's not adding a lot, but does earn it's keep in the added strength and reliability.

The fancy way would have been to do steel embeds like on surfboards, etc. But that's easily heavier than ply and a bit overkill here, imo...

I'm also concerned about the free edge at the speaker cut out crushing when I tighten or if I over tighten the mounting bolts...

 

[zac2944]

Great points guys.

I may end up doing the plywood thing. Seems sacrilegious for a composite build though... I know Robby did it... You're right, weight-wise it's not adding a lot, but does earn it's keep in the added strength and reliability.

The fancy way would have been to do steel embeds like on surfboards, etc. But that's easily heavier than ply and a bit overkill here, imo...

I'm also concerned about the free edge at the speaker cut out crushing when I tighten or if I over tighten the mounting bolts...

It won't crush. If you glass over the exposed core at the cutout it will be very strong. The speaker frame does a good job of distributing the load. Just don't go crazy when you tighten it down.

 

[1n3]

Thanks for the comments. I was thinking of normal operation, where strain should be greatest in panel centers. Of course, transport is another matter. There, I'd think the external skin strength would be more important.

 

[astack]

It won't crush. If you glass over the exposed core at the cutout it will be very strong. The speaker frame does a good job of distributing the load. Just don't go crazy when you tighten it down.

Hrm. Was hoping not to have to, but it would be an improvement to glass that. I might need to build up that area again, thinking of something like Great Stuf, because the Dremel overcut/over sublimated the foam on those cuts...

Now I'm thinking the most bulletproof method would be to put a 1" cube wood block @ each speaker bolt location between the skins to act as a spacer. Then put a 1/2" ply ring on the inside as you suggested and use t-nuts or sim. That would lock it all together really well, distribute the load to the inside skin and act to stiffen the baffle from both sides. I actually like this idea a lot. But, shh don't tell anyone there's wood in it

Next time, BTW, I think I'd also use the jigsaw at an angle to give more area on the inside to grip to.

 

[zac2944]

Hrm. Was hoping not to have to, but it would be an improvement to glass that. I might need to build up that area again, thinking of something like Great Stuf, because the Dremel overcut/over sublimated the foam on those cuts..

Just use some filled epoxy (micro beads, cabosil, etc) to replace the core that got chewed up. Fitting wood blocks after layup with be a p i t a. Make a really light filler by cutting it with solvent and add more micro. It gets like whipped cream and sands very easily.

 

[Passinwind]

Hrm. Was hoping not to have to, but it would be an improvement to glass that. I might need to build up that area again, thinking of something like Great Stuf, because the Dremel overcut/over sublimated the foam on those cuts...

Now I'm thinking the most bulletproof method would be to put a 1" cube wood block @ each speaker bolt location between the skins to act as a spacer. Then put a 1/2" ply ring on the inside as you suggested and use t-nuts or sim. That would lock it all together really well, distribute the load to the inside skin and act to stiffen the baffle from both sides. I actually like this idea a lot. But, shh don't tell anyone there's wood in it

Just for giggles, take a look at the various plastic Link Removed we use for sailboards. They're not too hard to retrofit, although of course they'll be stronger if done during the skin layup. High density two part foam can be pretty useful in these situations too, IME. The vendor in my link can fill in the details or suggest other methods, as they've been in the biz for many years. They used to be within walking distance of my house...

 

[astack]

Just use some filled epoxy (micro beads, cabosil, etc) to replace the core that got chewed up. Fitting wood blocks after layup with be a p i t a. Make a really light filler by cutting it with solvent and add more micro. It gets like whipped cream and sands very easily.

Hadn't heard of the solvent thing. West Systems, as usual, has a good primer: http://www.seqair.com/skunkworks/Glues/WestSystem/Thinning/Thinning.html

That actually sounds decent. But I don't think the blocks would be as much of a PITA. If I put a wedge shape on the side, I could probably just shove them in without having to remove foam (maybe this is bad idea, but we're brainstorming here, OK? ) More likely, I'd go at the foam with some dremel attachment that won't melt it, but also would be too soft to damage the glass skins.

Just for giggles, take a look at the various plastic Link Removed we use for sailboards. They're not too hard to retrofit, although of course they'll be stronger if done during the skin layup. High density two part foam can be pretty useful in these situations too, IME. The vendor in my link can fill in the details or suggest other methods, as they've been in the biz for many years. They used to be within walking distance of my house...

Ha awesome. I should befriend a surfer here and pick his brain on all this.

Speaking of, I've gotten to the point when describing to friends, etc. what I'm doing, I just say it's a speaker box made like a surfboard. It's a lot clearer than that mumbo-jumbo I put in the title

 

[Passinwind]

Ha awesome. I should befriend a surfer here and pick his brain on all this.

Speaking of, I've gotten to the point when describing to friends, etc. what I'm doing, I just say it's a speaker box made like a surfboard. It's a lot clearer than that mumbo-jumbo I put in the title

I might be able to hook you up, there are some awesome board builders in the Bay Area who also hang out up here a lot.

Using wood is definitely valid too though:

Link Removed

 

[astack]

Phew long day today, but I ended up with a fully glassed box. Oh yeah.

I cut up all the glass to begin, as usual. Set up shop. Then went through the sequence as planned. Mostly.

After the first two pieces:

Turned the corner:

Wrapping the edges was easier than I had anticipated:

Again, the 1" foam is working out great. A thinner foam wouldn't have as much to stick to when you fold over. It'd be doable, but I'll take the easier way. But the additional work of the wrap onto the front and back easily doubled the time to do the layup. Usually, it's 30 min +/- for a layup for me, and these were closer to taking an hour. Thankfully, it was in the 60's here today and I had plenty of pot life.

I used a bit of silica on the vertical surfaces and all that.

After the epoxy had a couple hours (it's a fast cure epoxy), I was able to use a cutter to remove the excess glass at the front edges. Great call, Zac, on cutting it green. And using the glassed inside surface of the side panels was perfect as a guide. I only finished up a bit ago, so the rest of the glass around the front is going to have to be trimmed in the morning. Hopefully it's not fully cured by then...

There was a bit of a screw up/change in plans in the ordering, but it didn't really affect the outcome. Still ended up with 2 layers with no overlap. For the final piece on each layer, they both ended up long because I overcut everything. What worked well was lay up the full piece, spread epoxy and then cut. Having the glass wetted made for less fraying and cleaner seems.

I did 4 5/8" +/- overlap (that's just what worked out to maximize overlap and minimize wasted glass - of course a spreadsheet was involved). Next time, I'd strike a line at each seam on the foam to use as a guide. I think I'd also note somehow the start time for each piece so I can do better job with pacing -- I've been meaning to buy a cheap shop clock so I can know how long I mix the epoxy for and what the pot life actually is, etc.

So, all in all, not bad. Just took longer than anticipated and I used a ton of epoxy. Plus gonna have to float quite a bit. But I feel that I'm starting to get the hang of this more and more. I'm glad I started with laying up panels as an easy intro. There was some tricky stuff today that I would've botched otherwise.

Speaking of epoxy, I'm using the epoxy weigh out spreadsheet less and less. Now, I'm just guessing at how much is needed from experience and simply weighing out the required 1:4 ratio (for my epoxy). The densities are close enough that the error between doing 1:4 by weight and 1:4 by volume is only a couple percent. The precision on my scale is much worse (1g), so no need to get to caught up in that.

Tomorrow, probably won't have time to do much of anything on this, but next up is cleaning, fairing, and making a decision on mounting.

Also on the to-do list for tomorrow, improve the storage situation in our practice space:
Parts:

Assembled:

Grossly overdesigned -- they should hold about a dozen SVT heads. Don't forget quake straps!

 

[astack]

I might be able to hook you up, there are some awesome board builders in the Bay Area who also hang out up here a lot.

Mayhaps. Thanks for the offer.

 

[astack]

Well just couldn't sleep and had to take care of the glass trimming

Sweet. Done.

Time for another weigh in:

Probably will be close to 9 lbs after fairing.

Side note: After yesterday, I'm thinking a slight improvement in my sequencing would be to not glass the back's outside before assembly, everything else the same though (so only the baffle gets both sides prior). It'd be just as easy plus I could then lap the outside glass around the seam.

 

[zac2944]

Nice. All of my cabs seem to come out 9lbs +/-1

 

[astack]

Solid as a rock...

OK, back to the cab...