Making a distributor cap

[Peter Twissell]

Excellent and useful document - thanks Ken.

My plan is to simply pour de-aereated resin into the female part of the mould, then introduce the male part, displacing excess resin at the join and finally clamping the two parts together.
Both wax and silicone release agents are advertised as suitable for epoxy and polyurethane.

 

[bluejets]

If anyone is concerned about insulating properties of resin or epoxy, there is an alternative.
Scotch make a variety of insulating resins in a variety of sizes for underground cable repairs.

https://www.3m.com/3M/en_US/company...sulating-Resin-4/?N=5002385+3294176988&rt=rud

 

[TSutrina]

PC board need to have good dielectric properties. So starting there [ https://www.engineersedge.com/materials/phenolic_plastics_engineering_data_table_13470.htm ]

G9 glass melamine 7.12 dielectric constant and 0.017 dissipation factor. Epoxy FR glass is 5.2 dielectric constant and 0.025 dissipation factor, phenolic 5.3-6 dielectric constant and 0.1-.045 dissipation factor, the charge measured by dielectric. Can operate at continuously operate at 285 F. Water absorption of melamine is four times that of epoxy and phenolic is double melamine. Go with epoxy.

 

[Ken I]

Peter T - Yeah, that works with shapes that naturally displace without entraining bubbles. One suggestion I would impliment on your excellent mould is a pair of tapped holes in the flange so you can jack it apart later. Assemble it with the jackscrews + release agent applied to screws and threads so they are in place if needed.
I would also suggest a weep hole at the apex in both sides of the mould with a threaded plug and filled with wax to the component level so that you can remove the plug, the wax and then introduce compressed air (or even hydraulic pressure / plugscrew against soft wax) to push the part out of the female mould or off the male counterpart. I see you have holes there but I can't see if they go through.
In most cases once it has "cracked" loose, the rest is easy. Easier to provide it up front rather than after you have a stuck mould on your hands.

petertha - all resins have shelf life issues. Polyurethanes need to be kept warm - like 30°C - I keep mine in a hot box - even so shelf life is seldom more than a year. Below 10°C urethane resins start to separate into constituent parts and have to be "reconstituted" by heating and blending (at 40-50°C) simple blending creates the illusion of reconstitution but results are poor.

Epoxies on the other hand like to be kept cool.

With soft urethanes, cleaning out is simple - so long as you let it cure properly before stripping. Hard urethanes you need to keep an eye on your residue and when it reaches the consistency of toffee, strip the feeder chamber and pipes - not the mould itself - leave that until it is well cured.
It even pulls out of the tubes (as long as you ran release agent through them) otherwise the tubes are "consumable".

Thanks for "bubble gods" - will mentally file that alongside "shop monster" which gobbles up any small part dropped on the floor.

Above my truck oil filter vaccuum pot - I made a lid with a window and illuminating LED's so I can keep an eye on degassing resisns prior to gravity pours etc. The black "seal" in the lid is gravity poured soft polyurethane and pigment - no release agent used in the groove.

Regards, Ken

 

[awake]

If anyone is concerned about insulating properties of resin or epoxy, there is an alternative.
Scotch make a variety of insulating resins in a variety of sizes for underground cable repairs.

https://www.3m.com/3M/en_US/company...sulating-Resin-4/?N=5002385+3294176988&rt=rud

As I understand it, pure alcohol is not particularly conductive, nor is pure water ... but there might be other things in the Scotch that would make it at least weakly conductive, so I wouldn't use it as a dielectric. Much better just to drink it.

Oh, wait - you were talking about the company, not the beverage? Never mind ...

 

[Peter Twissell]

Ken, the holes in my mould go right through and are threaded for jacking screws. The screws bear on the brass conducors and will be used to separate the male part of the mould from the resin. Sliding inserts in the female part of the mould support the brass conductors. These protrude from the bottom face of the mould, so pushing on them will separate the resin from the female mould.
I made a small trial mould with a single brass conductor and a sliding insert, similar to one of the 'legs' of the distributor cap. I coated the small mould and the sliding insert with wax, then mixed some epoxy resin and poured it in. I didn't use any vacuum, just allowed it to settle and closed the mould. I'll post some photos when my wife has finished with the PC.

 

[Peter Twissell]

The first picture shows the mould fille with resin and with the cap in place.
The second shows the moulded part.

The third picture shows a large void, where the resin trapped air as it ran into the 2mm annular gap. The epoxy resin was quite viscous, like syrup. I expect a lower viscosity would reduce the likelyhood of such a void occurring, but I will start gathering parts to vacuum fill the mould.
Even the viscous epoxy seeped into the clearance between the brass conductor socket and the mould insert, but this scraped off easily with a scalpel. I will fill the voids with wax for the next test.

 

[Ken I]

Pete, I admire your first building a prototype mould. (Off topic - Ferrari have a museum with a collection of its F1 engines in single cylinder versions for testing design ideas.)
Despite its viscosity, resins manage to creep through very small gaps - entrained and trapped air bubbles are the principal culprits for voids.
For some strange reason the harder grades of polyurethane seem to be much less viscous than the softer versions - finished product hardness that is.
With gravity casting a vibrating table is a great help - as a one-off lashup you can mount the mould on a flexible pole and vibrate the hell out of it with an eccentric link to your lathe / mill / drill / powertool etc. You can also drop the filled mould into the vacuum pot - but you must degas the mixed resins before pouring as it will froth up way too much - even then you will need a little headroom or live with the overflow (remember a bubble is going to expand infinitely in a full vacuum) and then top up again.
As you might guess I've had most of what can go wrong - go wrong - on more than one occasion.
As regards equipment - the aircon guys usually have a good vacuum pump to hand - perhaps you can beg, borrow or hire one for limited use.
A bulk spray painting pot makes a good vacuum chamber / pressure pot - my local resin supplier even sells such things along with vacuum pumps etc. for this very reason.
My equipment is salvaged scrap filter bits and a second hand Edwards 5lpm vacuum pump.
I once used my wife's pressure cooker with the anti-vacuum valve replaced with the fitting for my vaccum line etc. Not all pressure cooker seals will take a vacuum - a flat plate and a rubber gasket work well though. However this is also poor for relations with SWMBO (try it when she's out and try not to get caught).
(OT again - when they fill a car's brake system with hydraulic fluid on the assembly line, they draw down a vacuum and then inject the fluid - no bleeding or mucking about whatsoever).

 

[Peter Twissell]

Thanks for the tips Ken,
I have an automotive vacuum brake bleeder, but it's hand operated and slow. I work in automotive design and a while ago I specified a vacuum pump as part of a hydraulic scavenge system. It's a 12v unit with 4.5l/min capacity. I'll see if I can 'scavenge' one. I have spray pots, just need to fabri-cobble a lid.

 

[petertha]

Kind of an interesting take on a a pressure pot. I'm more leery of things going bang failing under pressure vs. no real chance of shrapnel when applying vacuum. But I guess its all about finding the right vessel/lid combination.

I wonder, does increased viscosity of the resin adversely affect the ability to reduce bubbles under pressure? The 30 psi is something I've heard before but maybe this assumes a viscosity of typical casting urethanes which are typically quite low. I think in the case of vacuum, reduced resin viscosity assists because the bubbles are migrating out of the liquid whereas under pressure they are just being squished insitu so as to be less discernible?

 

[Peter Twissell]

It is my understanding that pressure is used only to push the de-aereated resin from the pot into the mould.
Using pressure alone world simply compress bubbles, not remove them.
The viscosity of the resin will slow the process of de-aereation, as the smaller bubbles will take longer to rise to the surface, but a decent vacuum will expand the bubbles so they will soon break the surface of the resin and be released.
The second attempt with the test mould was more successful, with only some small bubbles at the cap end. That was achieved without vacuum, just careful filling of the mould.

 

[petertha]

It is my understanding that pressure is used only to push the de-aereated resin from the pot into the mould.

In the composites world that is just one process (typically infusion & related). But pressurized autoclaves have been around forever. The vacuum bag role is to help remove air entrained in resin and outgassing as much as possible, plus ensure the layup remains conformed to the mold plus leans the net resin content by bleeding excess resin into a fabric/wick like peel ply. The pressure chamber adds additional pressure to the layup to shrink bubbles. Probably an stress riser occlusion thing - small dispersed voids are better than big voids even if the bubbles are never fully vacated. But maybe with vacuum bag side is an 'outlet' the extra pressure energy is helping to drive any gas out. Also vacuum is relatively cheap. Adding 14.7psi pressure capability to a walk in size vessel means lots more $$ steel. The elevated temperature is to cure & strengthen the resin. In pre-preg cloth there is no resin mixing, its pre-impregnated in the cloth so is highly viscous, closer to a solid.

OK now we really have diverted from casting model distributor caps haven't we? LOL

 

[Peter Twissell]

Hi Petertha,
You are of course correct with regard to pressurised autoclaves. Pressure is used to force the lay-up against the mould and force resin into the mat.
However, the vacuum bag should have no gas inside, so the resin is being forced into evacuated voids, rather than gas bubbles.
I'm not sure that a 1 bar pressure vessel is any more difficult to construct than a 1 bar vacuum vessel. The former is loaded in tension, while the latter is in compression and buckling, but the magnitude of the load is the same in each case.

 

[Peter Twissell]

While we're drifting off topic, here's a gratuitous photo of my distributor, waiting for a cap.

 

[Ken I]

A 5mm diameter bubble pressed down by 4 bar becomes a 3.2mm diameter bubble - so pressure is not terribly effective (unless immense).

So if we vacuum it down to 0.95 Atu it expands to 13.6 diameter bubble which is going to float out a lot quicker - assuming the 5mm bubble volume remains (rentrant cavity etc.) and we now compress it from 0.05 Bar to 4 bar it shrinks to a 1.16mm bubble - if we pull 0.99 Atu the bubble will collapse to 0.67mm diameter and so on. Volume proportional to the cube.
Vacuum does more good than pressure - but you can see why both are desirable.

I am skeptical of the claims for the Firefly particularly with more viscous fluids - the centrifugal "sling" idea is commonly used by the artist types (the guys opposite my factory - The Handspring Puppet Company - made the marionette horses in Equus - won a Tony award for them - they moulded the "eyes" by this method but scrap rates were high - they took my advice and went vacuum/pressure and were delighted with the rsults.)

Petertha I like to keep the pressure on - so its more than just a means of delivering the resin - but if your mould leaks (re my tutorial) this may not be possible. I typically inject at 8 Bar - which is as far as my compressor will go.

Pete - I understand you being leery of things going bang - that's why my large mould in the tutorial is held together with 20 M6 bolts - there's about 12 tonnes to kept in check.

P.S. sometimes minor leaks are your friend - particularly if they help to bleed air from "traps" - I have deliberately "scored" some mating surfaces on some of my moulds for this reason.

Regards, Ken

 

[petertha]

I'm not sure that a 1 bar pressure vessel is any more difficult to construct than a 1 bar vacuum vessel. The former is loaded in tension, while the latter is in compression and buckling, but the magnitude of the load is the same in each case.

From that standpoint you are right. But the volume of a bagged part under vacuum is small compared to the volume of a big autoclave. Once vacuum is drawn on the part, it can be maintained by simply pinching off the line or connected to a very small vacuum reservoir for insurance against (minor) leakage. The vacuum buffer reservoir can be attached to the back end of the pump or separated as a dedicated pot if its more practical or if vapors from the curing resin are an issue to the pump internals. So I mean this (negative) 1 atm is cheap compared to adding 1 atm of pressure capability to a significantly larger steel autoclave vessel. And as discussed, we need vacuum for more practical reasons relating to the layup procedure anyways, so it just lessens the size of the pressure vessel. But this is industrial equipment unrelated to a model making solution.

I've seen demos of the resin infusion relating to dry cloth layups. That probably has some practical applications to liquid mold fillup, but its similar principle to what Ken was describing.

 

[Sprocket]

For a long time, my work involved laminating acrylic resins. We used a commercially made vacuum chamber, and pulled 15 in Hg for about ten minutes. (gel time about 25 minutes, full cure 1 hour)
The bubbles would surface to be scraped off or burst, then pour the resin into the PVA (polyvinyl alcohol) bag to be drawn through the laminate with about 7 in Hg vacuum. it made the lamination and elimination of bubbles much easier.
The degasser was plastic, 10-12 inches in diameter and 8 inches high, split top and bottom with big "O" ring and the vacuum sealed it. The acrylic was viscous enough to make the bubbles difficult to get out if not for the degasser.
I'm not sure if that is helpful, but it seemed the vacuum chamber didn't need to be terribly strong, and failure wouldn't be dangerous, but possibly messy .

 

[bluejets]

Couple of videos from Doc who made a vacuum unit for impregnating varnish in his ignition coils.
Good videos.....might apply to the epoxy issue here.

 

[Peter Twissell]

FIrst attempt at moulding the distributor cap in clear polyester (so I can see what's going on, both with the moulding and then with tests on the ignition system).

Lessons learned at first try:

I used wax as a release agent, heating the mould so it would run freely over the surfaces. It worked well as a release agent and the general finish is acceptable. However, I had not noticed build-ups around the bolt lug locations which left the moulded lugs mis-shaped.

After pouring the resin into the female mould with the brass inserts in place, I vacuumed the whole thing to remove any bubbles. Although the resin appeared free from bubbles, a lot came out.
On opening the mould, I discovered that vacuuming had drawn air out from the bores in the inserts (where the HT leads will plug in) and on releasing the vacuum, resin was drawn into them.
It was easy enough to drill the excess resin out, but I'll fill the bores with wax for the next attempt.

 

[awake]

Peter, it looks good! What product did you use for the polyester?