Vacuum Infusion - Decks

[FLexpat]

I decided to use vacuum infusion to make replacement decks for my 23 Sceptre but I couldn’t find a whole lot of useful info how to do it so I figured I’d start a separate thread on vacuum infusion. I will probably be going back and editing my posts to make them more understandable and fill in missed details. Assuming this is successful, this thread will only get into how I’m teaching myself, what I learn, and how I use vac infusion to make the decks (or what abject failure looks like). Either way, I’ve already learned a lot. Hopefully I get decks done and back to my rebuild thread soon.

Vacuum infusion (VI) seems to go by a lot of different names like Vacuum Assisted Resin Transfer (VART), Vacuum Infusion Process (VIP), Vacuum Injection Molding, etc. – seems like a bunch of folks are going acronym happy about it, but it all seems generally the same to me. Lay your glass and core out on an impermeable surface, seal the other side (flexible plastic sheet or mating mold surface), pull a hard vacuum on it with a vacuum pump, hook up a resin supply, let resin fill the glass/core up, and let it cure. It’s a bit beyond vacuum bagging where you lay everything up with resin by hand, seal it up and pull a vacuum to have atmospheric pressure compress it all with an air-free cure. Vacuum Infusion uses atmospheric pressure to ‘inject’ the resin, keep bubbles out, compress everything and make a good layup. Like everything else there are some details that can turn into nightmares and wreck the project and/or tools. One of the best threads I’ve read is at boatdesign.net; https://www.boatdesign.net/threads/infusion-plan.43380/.

My 23 currently has the decks cut out and is partially covered on a trailer in Park City, UT – that is kinda important since the atmospheric pressure at ~6800’ is less than at sea level; 11.5 psia instead of 14.7 psia so it limits what my vacuum pumps can do and limits how much force atmospheric pressure can apply on my layups. And I’m trying to get stuff done before it gets too cold and the snow starts.

I am using vinyl ester instead of epoxy since I want to be able to gelcoat it without worrying about bonding to epoxy. Since this is infusion, I am using ultra-low viscosity infusion resin instead of laminating resin.
The basic plan for the decks is (top to bottom):
7725 2x2 twill (to minimize print-through)
1808 (0/90, +csm)
1400 (+/-45, no mat)
1400
1808
¾” double cut Corecell (A-500) – Wanted M-80 but I got a bunch of A-500 really cheap on ebay
1808
1400
1808

I am doing the layup upside down so that I get a smooth surface on top (hopefully). I am beveling the edges of the Corecell at 45 deg so that the bottom and top layers of glass tie together for >2” all the way around the core. There is a ~1.5” lip all the way around the cockpit for the new decks to rest on. I left lips around the fishbox and fuel tank hatches too. Details on that go into the rebuild thread when I get a chance. The original deck had a 3/8” core and this one has ¾” so that affects the stringer clearance – more on that later.

There is a bunch of contradicting info out there on resin uptake for glass and not much info on resin uptake for the Corecell. I ended up making a stupidly complex excel spreadsheet to estimate high and low for uptake, then started experimenting.

I am running 2 vac pumps in series behind a resin trap. I can get to ~22” Hg, ambient here is ~23.2” so that works out to something like 30mm Hg which is ok since the styrene in the resin doesn’t boil til 5mm. For reference, ambient at SL is 29.9”.

Trial 1; 12x12 layup of all glass, no core
Goal – figure out better resin uptake numbers and get a quick peek at what I’m planning wrong – unfortunately I only took 1 pic of this test

First I cut all the glass and weighed it with a digital food scale – it was 6.4% more than the spec weights.
I taped a sheet of polyethylene sheet plastic to a table top since I don’t have a good layup table set up yet. (Don’t do this if you want a good part – the poly wrinkles and seems to be affected by the resin)
I put a layer of peel ply down, the stack of glass, then another layer of peelply over the top of everything.
I just used a piece of slit tube (instead of spiral wrap) for the resin inlet – won’t do that again. The inlet was full length on one side of the glass stack.
I used MTI (gas permeable) tubing for the vac line – along the opposite end of the glass.
I covered it all with Strechlon 200 plastic and stuck that to the board with tacky tape.
I pulled a vacuum on it and used an ultrasonic leak detector (ebay) to find leaks and patch them with tacky tape. Did a leakdown test for ~15 min and it held vac.
It was late afternoon, ~77F in garage and I decided to run the test… spec gel time for 1.2% MEKP at 77 is 35 min.
My calcs for resin included the tubing volume and some residual but the total range was a high of 30oz and a low of 17oz; the wide range was due to the crazy estimates of uptake ranging from a ‘general’ 1.5 times glass weight to 0.45 times glass weight (infusion specific). I mixed up 20 oz and rolled the dice.

Results and lessons:
My glass weighs 6.4% more than spec.
Polyethylene sheeting is good for some things, but not this.
Remember to have a clamp or clothespin handy to hold the infusion tube in the resin tub.
Resin took too long to get through/out of the lousy inlet tube – when it was starting to gel about 4” into the panel I had to use a piece of pipe to roll it into the glass towards the MTI tube – that almost worked but I should have used ½” OD spiral wrap as the inlet in the first place. Also start the infusion when it is cooler and let air temp increase.
The strechlon sealed the top half of the MTI vac tube – I love the MTI, but it needs breather on top of it.
Short version on resin uptake – RESIN UPTAKE FOR INFUSION IS WAY LESS THAN A HAND LAYUP/CURE. I will now use about 0.5-0.6 oz resin/oz glass (VAC INFUSION ONLY!), peel ply takes about 0.25 oz/ft2.
The glass stack was thinner than I expected (better numbers in Trial 2) and amazingly strong – even with a lousy infusion.

Next up – Trial 2; 12x35 glass with 9x30 core (it worked much better and I took more pics).

[flyingfrizzle]

Great info here. I wish I did more comparisons and weighed pieces when I had played around with the infusion VE drum I had. I would like to learn more and really get into doing layups this way as it is a much cleaner, stronger and better way of doing panel. Thanks for sharing the info.

[FLexpat]

As I think about having this thread make sense, there is a ton of background info I’ve skipped over, so I will be inserting that periodically. Here is some more background before I upload pics from my phone and babble about Trial 2…

An advantage of resin infusion over hand lamination is that much less resin is used; the compressed fibers take less resin and the process forces you to think about how much to mix so there is less waste. With a higher glass to resin ratio the parts are stronger and weigh less for the same laminate schedule. Unfortunately there is a lot more waste material like bags/films and peelply.

Another advantage of infusion is that there are way less fumes to worry with; there are some fumes from the resin container while mixing and infusing but almost nothing while the resin cures (I run a tube from my vac pump exhaust to the outside where it dissipates). When you peel the bag off after cure there are resin fumes, but not a whole lot.

From what I have read, it seems best to infuse from the lowest to the highest point – that way any air gets pulled out by the vac pumps and the resin does not ‘cut off’ dry areas that are then no longer connected to the vacuum. Gravity helps keep streaks of the resin from racing ahead on vertical stuff but not on horizontal stuff (like decks). With flat surfaces like decks some people infuse from the center outwards and some infuse from the edges in. The infusion plan is probably one of the more important things to consider – it drives set-up and materials as well as if you fail or not. I’ve come to the conclusion that for flat panels and ones with slightly elevated centers (like my decks with cores), it is better to infuse from the outside in since it is also going up. Another reason to infuse from the outside in on flat panels is that the area for the resin front keeps getting smaller as the resin infuses so the infusion does not stall as easily. With a boat hull in a mold they infuse it upward from the keel. If I had a dish shaped part I would infuse from the center. With an upside down hull or a cabin top they infuse it from the edges and then upwards towards the center. For now I’m just going to think about decks. There is a bunch of discussion on this in the boatdesign thread; everyone seems to have an opinion.

Controlling the resin flow pathways so that resin does not race ahead along lines in the fabric or core is important. Once an area is cut off by resin racing ahead and joining another resin area (think like huge ant pincers), the only way to get it to infuse is to puncture the ‘dry’ area with a large hypodermic syringe (I got syringes from Tractor Supply) connected to a T in the vac line with a shutoff valve, pull a vac on the void until resin saturates the area and gets into the syringe, then close the hole with tacky tape while withdrawing the needle. Seems like a giant pain that I want to avoid.

One source of air is bubbles or dissolved air in the resin; ya gotta be careful to not stir in air when the catalyst or hardener gets mixed in. Some guys put the resin in a vacuum chamber for a while before mixing and then let it stand for a couple of minutes after mixing before starting the infusion. Some put it under vac after mixing (before infusion). On my second test panel I got a small bubble – I think it was from mixing even though I thought I was being pretty careful and let the mixed resin stand for a minute or so before infusing. But I probably won’t mess with a vac chamber- thinking I’ll just let it stand a little longer before infusing.

[FLexpat]

Trial 2; 12.5x35 layup with core
Goals: (I had a bunch I wanted to learn from this test)

Find out how fast resin will move through the entire layup
Find out how far resin will go before gel
Make sure spiral inlet flows fast enough
Make sure double-cut core flows ok with glass layup
Verify resin uptake for glass with double-cut core
See if using self-adhesive Teflon film on melamine works with vacuum

I was struggling to figure out how to do a 13’ long layup of the decks with an air tight layout table – the best answer I could come up with was a couple of 4x8x3/4 sheets of melamine but the joint was a problem. There is a Teflon sheet (roll material, 1m wide) that works really well for hand layups that has a silicone self-adhesive backing and a nice slick surface. I decided to test it on a 2x4 sheet of melamine to see if it works with vac infusion since it would make sealing the joint between 4x8 sheets easy. I put the Teflon down on the melamine and then started doing the glass/core layup just like what I have planned for the decks:
peelply
7725 2x2 twill (top surface but on the bottom for the infusion)
1808 12.5x35 (CSM towards twill)
1400 12.5x35
1400 12.5x35
1808 12.5x35 (CSM towards core)
¾” double cut Corecell (A-500) 9x30 with beveled edges
1808 12.5x35 (CSM towards core)
1400 12.5x35
1808 12.5x35 (CSM towards peelply)
Peelply
It is all topped off with nylon bagging film

The twill is super hard to cut straight even with good glass scissors and tape, so I decided to try some electric cutters – THESE ARE AWESOME!!!! I used ‘em on all the glass, the peel ply, everything.



I laid everything out with spiral wrap for the inlet on one end and MTI tube for the vac on the other. I used a plastic fitting for the spiral wrap from Airtech to connect through the nylon bag to the Tygon infusion inlet tube – they work pretty well and resin does not stick to them. That was sealed through the nylon bag with tacky tape.


I did cut the spiral wrap a bit to let resin get in from the fitting easier (pics of that next time). I taped each end of the spiral down with flash tape to keep it from moving and to keep from getting flow out of the ends which could mess up this test. The test is also supposed to represent how resin would flow into the full deck layup from the perimeter. Having the spiral was on one end and the MTI tube was on the other end was like infusing one side of the deck but with a longer infusion distance (35” instead of ~15”). I had the glass extend 3” between the spiral and the core to mimic how the deck flanges need to be made. The peel ply covered all of the glass and the MTI tube was on top of the peelply – I covered The MTI with a strip breather material. I used tacky tape to stick the nylon bag film down over the whole stack – the tape was right on the edge of the Teflon sheet.



I pulled a vac, patched a leak with help from the leak detector, leak tested it for 15 min, then vac overnight to make sure any water was out.
Based on the first test results, I calculated the max resin uptake with 0.6 oz resin/glass for the 7725 and 1808 and 0.5 for the 1400. I increased the glass weight by 6.4% over spec based on that test too. The resin uptake of the Corecell was a lot harder to estimate. Gurit quotes resin uptake for A500 as 245g/m2 of surface area. I am using double cut core instead of perforated core only because I got a great deal. Perforated cor is better for flat panels and double cut is for curved surfaces. Double cut makes it hard to figure out the surface area but the short version is that ¾” double cut has ~7450 in2/yd2 of surface area including all the cut surfaces and both sides – that was a royal pain to figure out. That works out to a resin uptake of 41.5 oz/yd2. Including the volume for the layup, the feed line, the spiral, peelply on both sides, and 8 oz residual, I came up with 56 oz. My lower bound from using 0.45oz/oz for the glass and everything else the same was 46 oz.

It was about 70F at noon in the garage when I mixed 55 oz of the VE infusion resin and 1.2% MEKP using a drill mixer. I did it slow but could still see tiny bubbles in the resin. I let it stand for about a minute to let some of the entrained air out (it wasn’t enough time), put the tygon infusion tube in the resin and unclamped/unkinked it to start the infusion.

It took about 3 min to reach the core which was pretty close to my calc of 1.1 inches per minute.



The infusion rate through the core started out fast but got progressively slower.






The resin in the tub gelled at ~33 min with 14 oz left (~74F) so I clamped the tube then but it had not gelled completely in the part – resin continued to move for ~5 more minutes. It did a pretty good job of infusing but there were some dry corners at the end that did not infuse.

While it was infusing I noticed that the Teflon sheet was pulling slightly off the melamine around the edges of the layup – I kept pushing it down but was pretty sure the Teflon would not work as planned.

I let it cure under vac overnight (garage was ~80 for several hrs and dropped to 65ish), then removed the wrappings and trimmed with a metal cutoff wheel on an angle grinder.
Bottom (up during infusion)


Top (down during infusion)


I was amazed at how good it came out.
It weighs 6 lbs (2lbs/ft2)
Almost no flex with >300 lbs in the center when supported on the ends only.
The glass in the flange was ~0.175” thick as compared to an estimate of ~0.27” for hand laminating.
The resin moves through the glass flange lots slower than through glass+core.
For my layup schedule it took 15 minutes for 3” of glass flange and 18” of glass+core (21” total) – that should work fine for the decks but I might add some flow media on top of the flange just in case.
Although I was testing to see how far it would infuse before gelling, I think I should shut off the resin before gel since there was a lot of excess resin around the spiral and into the flange. Maybe even pull a vac on the infusion line after saturation.
My resin uptake numbers seem to have enough conservatism (slightly high) – I hope. I could probably get ballsy and use 0.5oz/oz for the glass and 41 oz/yd2 for the core and be fine but I don’t have enough experience yet to risk a $400 failure and do-over to save $20 in resin.
I need to let the resin stand longer after mixing for air to get out – there is a small bubble at the flange/core edge that I think is from entrained air in the resin.
The Teflon sheet does not work with vac for me – it lifted up from the melamine and left creases in the layup around the edge of the core.

For my deck layout table I will go old school and butt joint the melamine then wax it and mist with Partall.
The house did not smell like styrene – I might even be able to do this when the wife is home.

[flyingfrizzle]

Tons of good info there. I don't care what you do It is hard to keep the air bubbles out of the resin after mixing. I used to add cabosil to the thin voc infusion resin to thicken it some when I used it for vertical hand lay ups and I would mix it the day prior and let it settle over night and then try to be ez when adding the catalyst before use just to minimize the air bubbles from mixing in the cabosil.


33 mins is not bad, Im surprised it didn't gel sooner. I would also chill the ve resin in the fridge prior to infusion attempts and then pour it in large flat pan to keep it from building heat so fast. After the attempt I would turn on my 65 amp shop heater on and bring the shop up to 90 deg over night to help the cure. I use the word attempt due to I did not seem to have as good of a grip on the infusion as you.


I always used Melamine or Formica and have seemed the edges as you mentioned. It seems like you can get them fairly smooth but it dose show up in the part still slightly. My father in law has a plastic extrusion business and I have got a few pieces of scrap from him to try. It will do 60" wide material and long as you want to run it. His machine will do up to 6 or 8 mm If I remember correctly. I haven't tried this yet but was going to and hope it will work glued down to a backer so I could avoid seems.


One question. Your layup schedule? Just curious why you use the 1808 CSM vs layers of something like stright biax 1700 or 1800. Just wondering how the chop strand dose for the infusion. If it slows the draw rate or speeds it up. Im sure it adds some weight over standard biax in the over all glass to resin ratio but just curious what your take is on this?



Also, Like the Airtech hub. Looks like it would be easier to reuse than the 2 piece hubs.


Thanks for sharing all your info...

[castnetcracker]

are you using VE formulated for infusion? its much less viscous and wets out a lot faster. heres a video of Metan infusing a new sole for reference of how they located their inlets, hoses, etc. good luck on the decks, with that layup schedule it should be a tank!

https://www.youtube.com/watch?v=FE_bBFFNCK4

[FLexpat]

I am using VE Infusion resin from Composite Envisions - Can't remember who makes it for them but it does seem to have a nice balance of promoters and inhibitors - and the viscosity is 100 cp (pretty much like water). That’s much better than laminating (and most ‘thin’) resins which are in the 600-700 range. Since my garage gets fairly cool at night, I think my resin stays about 55-65 until I start messing with it. That probably helps on gel time but increases viscosity. Since the humidity is pretty low it really does not take much for me to crank the garage up to 90 this time of year. When it starts at 35 and is below 20 outside I can get it just over 70 in an hour or two. If I’m doing this too late in the fall I should probably get an LEL meter before cranking up the propane heater in there.

I picked that layup schedule for a few reasons:
I got a bunch of the materials (glass and core) cheap.
It is really strong in most directions and pretty light
The ¾ core takes up some of the potter putty space and gives the whole deck positive buoyancy; including core, resin, and glass the layup is about 36-37 lbs/ft3.

I am treating the top layers as almost sacrificial; I have seen a lot of things dropped on decks and destroy the top skin into cores and the twill and top 1808 should protect the other layers from damage. Also, with this layup schedule it shouldn’t matter if something damages it from a strength perspective and it should be easy to fix. Based on my numbers it actually weighs slightly less than the original decks new and lots less than the soggy mess I took out.

It would have been nice to get 1800 and some continuous fiber mat (CFM) instead of the 1808 but that didn’t fall into my lap. CFM is supposed to flow much better than CSM – I think it uses less or no binder and binder inhibits flow. I can definitely see that the CSM on the 1808 slows down the flow. I wanted some sort of mat to prevent the print through from the biax though.

I would love to have access to long extrusion pieces like that. Seems like if you used pegboard as a template for drilling some backer board you could glue it together without any air bubble high spots.

I took a pic of the surface of the cured layup and one of the matching corner I pulled it off of; you can really see how the wrinkles in the Teflon set the layup.




The 'bottom' of the layup (which was actually up during infusion) is much smoother - I think I could just gelcoat it and get an ok surface.

I re-watched the Metan videos today and noticed some things they do different than I do – some make sense and some seem like a Ford vs. Chevy thing. Gotta think about it some more but really good to watch again. Seems like they skipped over some really important details but they probably don’t want to educate the competition and they really did not have the time; promo video vs. step-by-step educational.