Efficient way to fabricate a mold ?

I am building a 46ft power cat. The design and the method are by Derek Kelsall (many thanks to Derek for both). I have built a multihull dinghy and have been vacuum infusing divinycell panels for a few years now and have some favourite techniques, accessories and established routine.

Given the current quarantine and everything related, I thought of using the already equipped workshop for laminating something commercially and partially employing myself.

If it is something that isn’t one-off, it calls for a mould, so I made my plugs and started on the mould following the typical, textbook route. My first moulding part has surface area of 3m2

In several days working on the mould, I realized that my new routine vastly differs from what I am used to from before. I had to get a new 5mm nozzle spray gun, had to build a ventilation moving 720cuf/min through carbon filter out of the shop, I quickly ran out gloves, brushes and 20kg of PE resin, I had to time my nearest 2 days to the mould layers, my shop became a mess and is no longer kinder friendly. And I am still trying to make something conceptually similar to what I was making before, e.i something strong, stiff, with controlled shape and surface, but somehow in what seems to be a so much different way.

This makes me wonder if the traditional mould construction with PE gelcoat and ~6 layers of CSM is still the best way to go.

My thought process is if the end result (i.e. mould) is physically and mechanically comparable to other composite parts, then the fabrication techniques should not be vastly different. Theoretically, the first layer of thickened epoxy reinforced with a light fabric can serve as both gelcoat and airtight surface, then would go proper skin, then some flexible core (Coremat?), then another skin, and then everything can be infused with a fraction of the typical resin estimate.

These thoughts were largely inspired by watching Andrés Chavarría build ()

I haven’t run a thorough estimate on the infusion way of building, but the rate I am going through the resin and consumables with the conventional mould approach is 3-5 times higher than what I normally have infusing a cored laminate of comparable footage, so it cannot be the most efficient way of doing it.

What is your mould-making experience? Any thoughts on a better or more efficient way? I would define “efficient” as one taking less time, less money and less special equipment

 

First. The way you are doing it is the slow way, there are much faster methods to build molds. None are really cheap though.

Polyester and VE tooling is far faster and cheaper than epoxy tooling, and for about 98% of what's built they're fine.

You can infuse the tooling, it works well, but there are some limitations, or I should say things you need to understand before going that route.

Using a tooling surface coat ( gel coat) is critical to the surface quality of the mold and parts being built, plus the longevity of the mold.

If you only desire an industrial quality surface there are many ways to cut time and money. If you want parts with a good cosmetic finish, the time, money and attention to detail go up significantly.

 

The way I am doing it now is spraying 20mil layer of PE tooling gelcoat, letting it cure for several hours and then spraying another layer. Just this one stage requires a special gun, special ventilation, compressor, full-body protection and even then it is not much fun to do in a garage. Then, I lay up 600g/sqm (20oz/sqft) CSM using 1.5kg PE resin/sqm (5kg resin per layer). Manuals say I need to build up thickness and put some 6 layers given my mat density, which translates into 30kg of resin for a mere 3 sq. meter mould. Because PE shrinks, I have to put one or few layers at a time. Each layer turns into a separate job stage consuming 3-5 mixing batches (~1kg at a time), altogether it takes me some 2 hours per layer including prep and cleanup. I have to let the previous layer cure and shrink, but not fully cure so I still can chemically bond to it later, hence I can't leave it for too long and work on something else. CSM needs to be torn, not cut, tearing leaves lots of fine strands all over the shop.
Yes, this way seems to be very slow, there must be a better way.

Infusion is a much cleaner and better manageable process of lamination. Epoxy doesn't shrink and is more family and neighbour friendly. Logically, infusing epoxy seems to be the direction to move, but it needs to be a different setup with a different surface coat, different skins and a new core material.

 

I can understand why you're frustrated or discouraged by the process you're using, but you aren't going to reinvent the wheel, there are reasons why certain steps can't be eliminated.

It's also why building tooling is so expensive, there are critical steps that can make or break the project. The longer you plan to use the mold, the better attention to detail you need to use.

Using actuall low shrink tooling resins will cut days out of the process.

Infused tools require special products and methods if you desire a great cosmetic surface.

 

I'm a total novice so pardon me if I'm talking rubbish here. Your main problem seems to be using a spray gun and having to mix, set up etc multiple times?

For the tooling gelcoat there are mold resins out of epoxy that are thicker and thixotropic. That at least should remove the problem of shrinking and make it easier to quickly build up a thicker tooling surface. If this is cheaper is another question... And you have to do it by hand.
Buy Mould resins online at R&G https://shop1.r-g.de/en/list/Resins/Mould-resins
EDIT: Andres also seems to be using this epoxy mold resin, but it's not resistant to styrol which might be a problem. The devil always seems to be in the details!

There is also a wiki but it's in german (translate) and they are talking mostly epoxy there:
Harze:Tempern/Warmhärten – R&G Wiki https://www.r-g.de/wiki/Harze:Tempern/Warmh%C3%A4rten
https://www.r-g.de/w/images/9/96/Formenbau_mit_GFK.pdf

This video also shows how you can throw fillers on top of the still wet tooling gelcoat to get a good bond to the rest of the backing if you want to continue at a later date. Or you use wet-in-wet.

So they do say you can use a cloth directly on the mold resin without CSM coupling layer. The video also shows this.

R-G also have instructions on integrating carbon into the form to build a heatable form for tempering. All very fancy.

I've also wondered if you could use wood "bulkheads" and PU spray can foam to stiffen up the backing of a mold to get away with less laminate thickness. But that might not be so durable. But you'd also have an insulation for a heated mold.

I'm watching "Life on the hulls" and he has a vinylester spray gun that mixes outside the spray nozzle so with the right tools it would seem to be very quick to build up gelcoat. Together with a chopper gun I figure making a mold becomes rather quick and easy. But lots of material. Maybe the best way would be to find a composite workshop that has those tools, prepare your plug properly and pay them to do it for you with the right tools.

Sorry for the long rambling and uncertain reply but I learn by regurgitating what I have seen before to better understand it myself

 

Epoxy tooling surface coats are very difficult to use, then tend to crack. They can be made to work on smaller parts, but big parts are almost impossible, especially for a rookie.

And fillers do a terrible and unreliable job of bonding gel coat and epoxy. It's far better to just use an epoxy compatible gel coat.

But its rare to make epoxy tooling for any marine project.

An estimate for a place to build tooling for a 46' boat could easily top $200,000

It could be less if it wasn't a complete set of tooling though.

 

Obviously, there is no one “best” method, just like with boat building, it is always a compromise between fast / good quality / economical and perhaps a few more axises where you get to pick only two and take a hit on the rest.

I guess, before asking about an efficient method, I should have defined the context and prioritized the compromise dimensions.

It would be nice to have a high-end glossy surface on the final parts, but I may as well use a textured polyurethane or polyurea paint (bedliner) and thus lower my chain of expectations to plug finish, mold surfacing coat and prints from the adjacent layers.

What’s more crucial to the project is fitting into the cost and time budgets and not being evicted from the house before I complete the moulds.

Eventhough working with polyester helps to contain the costs, the fumes from 3m2 surface multiplied by 6 applications (not to mention spraying 2 gelcoat layers) pose serious risks to sustainability of a garage-based workshop. Carbon filter and 720cuf/min ventilation only partially help.

Given these priorities, I believe that infusion will make a big help on the consumption (i.e. total cost) and the exposure of polyester. I have to bite the bullet on the mess and fumes from spraying gelcoat and on layering up the first layer, but the rest of the layers are better to be infused. But if I compress all the layers with vacuum, I will get it strong but not stiff. This means I need some flexible core like Lantor Soric (used by Andres) or alike to get the bulk without using a ton of resin.

Epoxy-based mould system gives advantages on absence of fumes, better resin properties, longer working time (less mixing batches), and it could be used over styrofoam-based plug without shielding layers, but it is pricier and not so clear on the surfacing (gel)coat options.

Wish there was a brushable epoxy gelcoat that could be covered by infused polyester, wouldn’t that be nice?

There are quite a few options, materials and techniques to consider. So far, I tried the most conventional method and now can see many areas where it could be optimized. The research and design are still in progress.

 

I had only heard of "3D core" which is a PET or XPS hexagonal foam with infusion channels so infusion is easier. Seems similar to the Lantor Soric.

PRODUCT RANGE - 3D|CORE - the structure in the core https://3d-core.com/en/products/product-range.html

 

You can brush on polyester tooling gel coat, its easier than epoxy to work with.

You want the molds to be stiff, this is one of the things you need to plan into an infused mold.

Infusion will also induce far more fiber and core print on the surface. So a skin coat needs to be planned with resin and fabric to help reduce it.

You can achieve very long gel times with polestr resins, far longer than you'll need for these parts.

Polyester can be infused far easier and faster than epoxy, so extra long gel times are rarely needed.

 

And don't confuse a lack of oder with a lack of toxicity.

If I was going to be exposed to one of these resins, the likelihood of health issues is far more serious with epoxy.
Polyester's terrible oder is a big plus, it creates a situation where you will avoid the oder if at all possible long before the fumes reach a level where they become a problem.

The annoying oder is a huge hassle when trying to work on a project where other people reside in the same place, or neighbors are close.

 

The toxicity of epoxy is a well-beaten topic, not looking to spin it another round, just sharing my experience.
When laminating 6sqm panel my exposure to epoxy lasts about 20min, which is to measure and mix a batch and to adjust the infusion intake playing with clamps and hoses. My exposure to stinky amine hardener lasts some seconds needed to scoop a cup and to weight it before mixing in. Such exposure is relatively easy to control given the short timeframe.
The same would be true for polyester IF it was also infused. But the gelcoat, whether it is brushed or sprayed, has to be laid as an open layer and hence makes a completely different story. Even though I have a forced ventilation system, my Organic Vapor 3M respirator is not keeping up with the styrene concentration so I am far more concerned about health issues working with open Polyester than with Epoxy. Besides, I don't expect my neighbours to be savvy enough where I could explain it is just a light smell of styrene and nothing hazardous. The next day after I started to work with open polyester I ordered smart home Air Quality Monitor measuring TVOC.
I agree with the theory, but in my practical experience epoxy is safer to handle than polyester.

 

50 years of hands on experience and the research to back it up, that's all I have to offer.

What I explained earlier is what you just shared.

The odor of polyester trumps the actual toxicity at the levels recorded in a workplace for people somewhat new to it.

Your nose can detect styrene at levels in the 1-3 parts per million range and even lower. The typical workplace limit is 100 ppm for an 8 hour shift. But going to 50 ppm.

The average person would think they were dying at 50 ppm. Although someone in the industry wouldn't even smell it.

At these levels even after decades in the industry its difficult to determine if there any real health effects. Don't get me wrong, I'm not saying its healthy, only that there aren't glaring health issues related to being around it for very long periods of time. Past research was done on people exposed to styrene at much higher levels for decade's and they did find some health related issues. Hearing loss was an odd one.

Epoxy on the other hand can be extremely toxic even from inconsequential contact, like uncured grinding dust, brief skin contact etc.

Contact sensitivity is a real thing with epoxy, and it can be quite serious. I know many people with this condition.

People that do DIY projects always react to the odor of polyester as if they're going to die before the end of the day. While it is obnoxious, the odds of any health issues are far outweighed by the meal and alcohol served for dinner.

 

For some reason it keeps changing the spelling on odor.

 

My process;


24 mils nominal VE tooling gelcoat , 4-5 passes via air assisted airless
~2 to 3 hr cure
35 mils nominal VE barrier coat , 2-3 passes via air assisted airless
~2 hr cure
90 mils chop, 0.75" nominal fiber length, 100% VE via internal mix chopper gun
overnight cure
grind out bubbles, voids, defects, sand entire surface with 32 grit, putty in ground-out spots in with VE resin + milled fibers at ~1.5% MEKP9, then immediately move to chopping
90 mils chop, 0.75" nominal fiber length, 100% VE via internal mix chopper gun
overnight cure
grind out bubbles, voids, defects, putty in with VE resin + milled fibers, allow cure, sand entire surface with 32 grit, blow off with air, wipe with acetone

Open mold continuation;
Apply low-shrink tooling resin per manufacturer specifications, typically Polynt Optiplus 040-8089 at about 180 mils at a time per manufacturer recommendations
Small molds ; ~0.3" nominal thickness (one bulk)
Large molds ; (~0.6" nominal thickness (two bulks, sometimes three)

Closed mold continuation;
Infuse low-shrink tooling resin per manufacturer specifications. I typically prefer Vectorply QXCFM-3515 or QXCFM-5615 for infusion. For better flow rates, 3515-6VC is utilized.

Material costs averages:
VE Tooling gelcoat ; $5/lb
VE Barrier coat ; $8/lb
100% VE resin ; $3/lb
Optiplus 8089 ; $4.5/lb
Infusion low-shrink ; $5.50 - $6/lb

Because of the resin savings, I can typically be below $10/sqft for infused tools. Open mold is usually about $15/sqft.

 

The OP has been back a couple of times, but hasn't commented in this thread, it's probably my fault for discussing health related issues.

This thread sort of followed a common theme here, a person joins and says, "I want to build X", there are parameters set on how they want to do it, but little info as to why. It doesn't matter if it's wood working or composites, it follows a similar pattern.

After some questions from longterm members the ideas of why become a bit clearer, unfortunately they may not be accurate assumptions of how to achieve the goal of building X.

The thread starter wants use materials or processes that they don't fully understand to cut costs and time.

When experienced members start to discuss details of why its done in certain steps or with specific materials to be successful, they don't like it and sometimes bail. Or argue why doing it their way will be far better even though it never has been in the past.

In this case building tooling for a 46' cat that will be used in production is an expensive and time consuming project, there is no way to cut corners. The methods used to build good tooling have been fairly well established over many decades.

There are reasons why epoxy tooling is extremely rare in the marine industry, it significantly increases the time, cost and hassle factor.

I fully understand why people dislike the odor of polyester, and its even worse for someone that isn't around it often.

With infusion the odor becomes almost a non-issue, but infusion requires more attention to detail and some special steps to achieve a good surface profile.

Also, stiffness needs to considered in the design.

The spraying or brushing of a tooling surface, whether it's polyester or epoxy is a required step. Polyester is by far much cheaper and easier to use, and for production use it meets or exceeds all of the requirements.
It does have the odor penalty, but its short term and not easily avoided with either resin type.

If you are building a 46' cat at a rented house and doing it on the lowdown so neighbors or the owners don't find out, then you're off to a very bad start.

Basing the materials used to build the tooling on this is even worse.