Um…those are really good welds…I am impressed…you have a tig welder in you garage or this in your shop?
I wish I had that skills in welding…but I ain’t…
Kevin Doe is really a robot. Thats why his welds are so good haha!
Thanks guys. I have the welder in my garage. I’m an engineer by day, fabricator by night (hobby). I’m hoping to add composites to my playbook.
What field are you in?
Aviation. I work for a large company that builds jet engines. I design high speed telemetry systems for development test engines.
Not a lot of progress today. I managed to get the thing all plumbed up, and all the pieces mounted to the board. Up next, lids.
IMG_5337.jpg by Kevin Sobkowiak, on Flickr
How will you be capping the top? Plexi, or a hinged setup?
Nice work. Your doing a great job!
BTW, you can use rattle can primer but you need to use wax and PVA. Almost all rattle can paint will chemically react with PE, VE, and epoxy resins, which wax cannot block (no matter how many coats your apply). PVA’s role in the release system is to block the chemical reactivity while the wax provides the slip.
There are some other surfacing agents that will not interact chemically with certain mold surface coats. A test prior to use is always a good idea with an unknown combination.
If you utilize a sandwich construction it’s important to balance and mirror the laminates on both sides of the core. You’d want to go 0/90, +/-45, core, +/-45, 0/90. As the resin shrinks it will create tension within the laminate. If you do a thicker layup on one side of the core, it will cup/bow towards that side. The same thing will happen if one side has a lower tensile modulus than the other (carbon on one side and glass on the other). If the layups on each side mirror each other then the tension on each side is equal. If the part only has 0/90 fiber on each side the sandwich will develop a twist. If only +/-45 fibers are used the sandwich will cup and bow. Since this composite part will be adhered to the rest of the frame the warping of of an unbalanced sandwich may not be such a big deal.
You might consider using a Soric core so you can infuse the whole thing in one shot.
Oh and…I wish I could weld like that!
I finished up the lids for the resin trap and degassing chambers. The resin trap uses a stainless lid, 1/2" thick. The degassing chamber uses a Lexan lid, 1/2" thick. I machined an O-ring groove into each lid, 0.160" wide, 0.250" deep. I am using Viton o-rings in the grooves. The atmospheric pressure seals them tighter as the vacuum increases. On resin trap lid had 350 lbs of force holding it closed at full vacuum. The degassing chamber has 740 lbs of foce holding it down at full vacuum.
I added a tube to the resin trap lid to channel resin into the cup that I’ll place inside it. I did this to make sure the trap works as intended. I doubt anything will make it up into the tube, but better safe than sorry.
I pulled a vacuum on the containers and turned the ball valves to closed. Zero vacuum loss over 10 hours. Perfect. I was a bit concerned about the force of the lids smashing the o-ring so I measured the height of the lid while free, and while full vacuum. The o-ring compresses 0.020". That seems pretty optimal really, so no limiters need to be added.
IMG_5465.jpg by Kevin Sobkowiak, on Flickr
IMG_5463.jpg by Kevin Sobkowiak, on Flickr
IMG_5461.jpg by Kevin Sobkowiak, on Flickr
IMG_5464.jpg by Kevin Sobkowiak, on Flickr
After I did the vacuum test I found out that even with the ball valve on the pump open, vacuum won’t bleed down. So the lids were stuck on with the force of god. Luckily I just have the top port of the resin trap plugged with an NPT plug. I was able to just thread that out to relive vacuum. Given that, I need a better way to do that. In this picture you see a half coupling next to a tee fitting. I’ll be welding the coupling to the fitting and drilling a through hole. On that I’ll thread in another ball valve with a breather fitting on the end to keep debris out when I relieve vacuum.
IMG_5466.jpg by Kevin Sobkowiak, on Flickr
I also started polishing the mold. I am using Meguiars polish that I usually use on car paint. I tested an area and it works pretty well. I’m only done with step one, but its coming along nicely. This is with M105, Ultra Cut Compound. The gel coat is pretty hard so I had to up the speed and pressure accordingly.
IMG_5472.jpg by Kevin Sobkowiak, on Flickr
IMG_5471.jpg by Kevin Sobkowiak, on Flickr
How did you machine such a large disk? Also what did you use to machine the plastic? Your cuts are really clean In the past when I’ve tried to machine plastics it’s fractured or the cuts have been very rough, although I’m sure my speeds and feeds were off.
The Lexan disc OD was cut out with a water jet machine. The groove was cut out on a lathe. I made a hand ground tool from high speed steel which looked somewhat like what a cutoff tool would look like. To cut plastic, the surface speeds are fairly low. For this groove I used 40 rpms (it was ~8" diameter groove). The tool must have zero rake angle ground into it. I tried to feed at a rate of 0.005"/rev manually. It turned out pretty well.
I cut the stainless disc on a lathe as well. Similar technique, except stainless steel requires some rake angle, about 4 degrees.
Very nice man! Great job polishing it up :). Now infuse!
Here is my modified t-fitting with quarter turn ball valve and breather vent all put together.
IMG_5479.jpg by Kevin Sobkowiak, on Flickr
And re-installed.
IMG_5482.jpg by Kevin Sobkowiak, on Flickr
I was unhappy with how much the whole thing vibrated even with the pump mounted on factory rubber bushings. It would vibrate so much that the whole thing would slowly slide off the workbench. So I ordered up some rubber mounts that would put more rubber between the pump and the plywood. 4x the rubber thickness to be exact. I also bought some rubber feet for under the plywood that are really soft and made for vibration damping. The whole thing works great. No vibration at all transferred to the bench its on.
IMG_5494.jpg by Kevin Sobkowiak, on Flickr
Great workmanship. Please keep it coming! I am in the process of working out how to do the same but with a car wing.
Machining, welding and now composites?!?! It looks like you’re quite skilled in all these areas of fabrication. How long did it take you to learn all of this?
I’ve just picked stuff up over the years. It helps having a very mechanically inclined mind, I love learning new things like this.
So, its getting time to order some resin. I’d like to not paint the part, so I think epoxy resin is out due to the lack of UV resistance. I am thinking that vinylester would be a good place choice. Does anyone have a recommendation for me, to use as infusion resin.
Also, I’m looking for thoughts on the feed/vacuum locations on the part. I was thinking about putting a vac spiral (or MTI hose) around the perimeter of the part and feeding in the middle. Keep in mind I’ll be using a core. Any thoughts on that?
Another question I have is layup. Would you try and do this out of one sheet of carbon, or do the top/sides, and then the flange with some overlap?
I was planning on doing an infusion all at once (including a 2 layers, core, 1 layer), but I think I’ve decided against that. I think I’ll just infuse the 2 layers first, then glue the core down with some resin and let that cure, then infuse the next two layers (or even lay them up wet). The reason I’m thinking that method may be better is that I cannot find a structural core that is infusion friendly. I found this Soric XF core, but it says it does not offer any structural properties to part. On the rear long edge of the sunroof (drivers to passengers side rear edge) I’m going to have to cut off most of the vertical flange so it will fit in the car properly. To regain some stiffness in that direction I was thinking of running a fairly large stiffening rib along inside of that area. I was thinking something along the lines of a 1" foam rib. Given the thickness of the rib I don’t think I’ll have a successful infusion if I attempt it all at once. In addition to that, I think I’ll have less of a chance of print-through of the foam ribs if I let the first two layers cure first. Thoughts?
I use Soric in load bearing wings. It adds a significant amount of stiffness to the part due to the increased thickness. The core is functioning as a sear web between the skins.