So, NinjaTek have released a new very hard filament which is petroleum resistant – could be useful for making my composite tank. I still need to understand if the regs require foam filling, or if flap gates are adequate. I could then completely print the pump mounts, flap gates swirl pot, etc.
This comes on the back of this post about making tank bladders.
I’ve been thinking about an internal bladder when I make my fuel tank, and one thing struck me – infusing aramid with viton rubber (very low viscosity) – wouldn’t that get the best of both worlds – really strong bladder and fuel resistance to boot. then sling the entire thing in a CF/Aramid tank.
I can’t see anyone selling VITON compound on its own though – seems to be as a finalised product.
Any thoughts?
So, even though I’ve got my Velleman 8400 printer and not assembled it, I’ve been thinking of what to do with it. Needless to say, with 3D printing that started in the community there’s not only a great set of community based CAD packages, but people like Autodesk are in on it as well with TinkerCAD and a quite interesting suite of free apps as well (including using your i-device as a 3D scanner).
I think my first project (I have some 3D experience with CAD anyway) will be to rework my composite fuel tank. I had to junk the original plan – it wasn’t possible to really make a part from the moulds – I had returns greater than 90 degrees on some flanges and that just doesn’t work.
What I’ll do this time is print a framework that bolts together, and bond on to that framework a set of plastic panels. The end result will be a completed tank that can then be moulded from. I’m also going to simplify the design with far fewer faces, but some of the curves (radius curves so the carbon can go in) will be much easier to make more accurately. I think I’ll make two halves that slot into each other much like the top and bottom of a Matryoshka doll.
I will make the panels out of (maybe) 3mm polypropelene sheet. It has great release properties, but then of course, is difficult to bond to the framework. Nothing that a good keying and epoxy won’t fix. I can use other plastics of course, and if I’m a little rough with the cut, then I will fill the gaps with wax. It’s only a tank and not a display item.
and I’m a happy bunny. The finish is a little rough, and I can do a little mould polishing to get that, but once the part is in, it will never be seen again, so I’m not going to get all artistic about it.
Prepare to swoon at the fitment …
Top Down – here it is, fitting nicely, stopping about 100mm from the gear lever. There’s a good 10mm clearance all round between the tank and the rest of the transmission tunnel.
Open Wide for the Doctor
This is the view right down the tank into the sump.
Bolt Recess
Here you can see the cut-away area of the part I made to allow me to get a socket in to first insert and then secondly tighten the 1/2″ bolt for the top of the diff cradle.
Tank and Chassis fail to meet
The tank nicely runs under the chassis rail. This is just shy of 10mm. The tank isn’t going anywhere once it’s in, so I’m not stressing this.
Diff Clearance
Again, loads of room here. The diff cradle is held in place with 1/2 inch bushes, so there’s no way can it ever touch. I’ll also be making prop-catchers, so I won’t worry about the prop hitting it. And the tank will be in aramid, so i’m not fussing that.
Topping Out
Here’s the tank the whole area behind it is where the old MG Midget tank used to sit. The black brackets are temporary. I’ll made something rigid out of CF that spans the top of the tank between the cross-members. The tank itself will have m6 male fasteners bonded on to it, so it can be fitted from below.
The last picture demonstrates just how far I’ll be able to move the mass of the tank. It’s forward by 300-400 mm, to the point where it’s now inside the axle line. The aramid tank will have better impact resistance properties than the steel one it replaced, and being in the new place it is in the chassis, will benefit from greater impact resistance. The area behind (at the top of the picture) where the old tank used to hang from will be replaced by a crumple-zone.
And … Attach
These are Big Head fasteners. You basically put a big dollop of the special 2-part epoxy glue no the part (after keying it), push the fastener through, let it set and trust it.
I don’t have a solution for the front of the tank yet – I think I may take a bracket off the prop-catcher I haven’t made yet, or make a top-hat shape, cut a hole in the tank and stick the top-hat to it, but inverted so the top hat makes a recess. Then I can stick a big-head to it (I have some female ones) and bolt through from the other side of the transmission tunnel. It’s going to be … specific.
So, before I make the part out of lovely aramid and carbon I thought I’d do the sensible thing and crack a part out of chopped strand mat first. I wanted to do this for three reasons:
Here we have the entire set of parts pulled and trimmed from the mould. As I went along, my trimming got a lot better a lot quicker. I realised it was far easier to scribe the part in the mould for where the trim-line should be, and then to pop it out and trim it. As such, I have a few gaps which have been sorted with how I’m bonding it together (30mm strips of glass, wet laid on).
Now the part is pulled and bonded together, hopefully you get a sense of the size and shape of it.
I managed to bond the whole of the top tank (the top bar of the T shape) internally, but the lower part needs to be bonded externally. I did a reasonable job, but it’s not watertight. It was good practice for when I do the aramid/carbon part.
The entire part weighs just over 2kg, which gives me a saving of 3.7kg over the standard midget tank. It’s a great weight saving, before I even take into account moving the mass about.
So, I’ve laid up two of the four quarters of my composite fuel tank mould. It’s a simple layup of gel-coat, and then three layers of 450gsm chopped strand mat, set in polyester resin. Nowt flash.
Since I’ve done the first half and the second half mated to it, I’ve learned a couple of lessons:
Here are the ‘bad one, then good one’ pictures:
Finally, here is the part with both flanges in place.
Here is the join between the two bits of the tank – it’s taken quite a bit of time and bog to get this right, as well as using a flap disk on a drill to get the radius right. lots of polishing and filling as well. I managed to get some of the profiles right by taking a plastic bog-spreader and cutting a radiuses edge into it. I also found that a wooden tongue depressor also had a very useful profile for putting curves into things.
So, we’re moving further along with the fuel tank. The last post I did on this was more or less finished, but then I’ve done a little more.
Firstly I decided that the recess for the fuel-senser was a little tight (potentially tight) so i modified it, thus:
It’s a nice shaped curve, feels organic and now there’s loads, loads of room (I tells ya) to get the sensor in there.
The black disk is cut from a piece of carbon sheet I had made and then changed direction. It was lying around.
If the bits of bog on there leave me with a surface that’s not flat enough to work with a rubber gasket (which I doubt) then I can also cut another disk and bond it on – the epoxy glue can go on as thick as I need and it’ll leave me with a good sealing surface. Similarly if the angle on the black disk isn’t quite right and the sensor doesn’t go where I need it I can adjust it a little with adhesive. I’ve spent a lot of time trying to get that disk in the right place, so here’s hoping.
Here’s the final tank.
So, the last things that have been done have been to paint a little resin over the wood to lift the grain. This is necessary so I don’t end up with the grain lifting when I spray it. Once the resin has set, the grain is lifted and a quick once-over with 120 paper on a sanding block did it.
Lessons Learned
So, a lot has moved on tank wise.
As ever, pictures speak louder than words. This is the tank now extracted from the car, and you can see it’s more or less like a model from StarWars. In the second picture you can see how it runs down the tunnel. The cut-outs in the top-left and top-right corners are to provide a recess for the sockets (as in socket set) so I can do up the diff mounting brackets. In order to fit the tank, the diff and brackets need to be dropped down to floor level (excluding the height of the the trolley jack it sits on) to get the tank in. What I needed to avoid was boxing myself in with the tank.
It’s important to get decent radius edges around the tank – if the carbon is just shoved into what’s more or less a 90 degree bend then it will bridge between the laminates, or between the laminates and the mould (filling the gap with resin). Not only does it weaken things, but you’re putting unnecessary amounts of resin in. Don’t forget this stuff is £20 per kg, you don’t waste it.
In my first attempt, I had started sanding around the curves and bogging in the gaps. I knew this wouldn’t be perfect because I’m not a sculptor. However, the tank wouldn’t be on display, so I thought I’d not be terribly worried. In retrospect, this clashes with the very pleasing OCD I’ve been brewing around neatness.
For another unrelated reason, I was in B&Q, and I had a revelation. They have 16mm quadrant beading. It’s cheap, and easy to put into the part. When I say easy, you cut a hole, stick it in, and have to start blending with bog when you go around the corners.
So, I went and cut out all the edges I had radiused (sp?) by hand, and put the beading in. Basically, you use your combination square to measure in 16mm from the corner in to each side, cut with an angle grinder and off you go.
There’s a bonus to this, which is the quadrant beading squares up the edges to the eye. The gaps between the edges and the poly sheet are made up with bog. So, it’s win-win.
I’ve tried a couple of ways to stick the beading in after cutting the raw part, and the main way is to use bog again. Hot glue is great for an instant stick, but you can’t sand it and it isn’t as strong a set as bog. The only downside of bog is that it takes about 15 minutes to go off so you find yourself doing all the work in batches, or alternating between one part of the tank and another. It’s fine and efficient until you have to join both bits together.
Here is the near finished tank, and you can see most of the radius edges. I reckon there’s 60 hours of work in this. A skilled part-maker would do this in half the time, to more accuracy.
Myself, I’m chuffed. I think the tank will save me 5kg from the overall sprung weight.