2 Jan 2017

I figured a year-end update is in order given how quiet it’s been at this end – and no, I am not selling the car, hah.

Summed up, I’m burnt out from working on the same (though rewarding) project since December 2007 – 9 years! The years-long design and build, the book, having the engine blow up, the expensive and prolonged rebuild, the recent composite work, the sanding, all combined with our yard being ignored; it all kind of came to a natural decision point after the koi pond literally fell apart.

You know me and projects; I got sucked into the technical aspects; the chance to bring the filtration system up to date, combined with the puzzle of how to do it well at a reasonable cost. The new filter system (drum-type particulate filter and moving-bed bio section) is now operational. To help bring the new system up, the existing bio media was transferred to the new setup from the old filter -the one previously mentioned that’s slowly settling into the ground at a worrisome angle. Until the new pond is ready, the existing pond will utilize the new system. While the new pond will have a gravity-fed filter, the temporary setup uses to the existing pump to push water to the new filter. It’s not optimum but it serves its purpose. With it running for five days now, the good news is that the drum filter works great – and the bad news is that it works so well!

There were two surprises: because the pond has never had a real particulate filter other than a settling tank, there’s all sorts of crap (both figuratively and literally) in the existing system. As a result, the drum filter quickly loads up and cycles the sprayers, as designed, which isn’t a problem. The issue is that the sprayers are loud – like someone next door using a pressure sprayer (I put up a short video on Instagram, see it and other stuff at @midlana1). Additionally, each time it cycles it produces roughly a gallon of yucky water and with it cycling about every 10 minutes, that’s 6 gallons an hour, or about 72 gallons overnight – that’s a big pail! For now I’m putting up with it and as the filter does its job and clears the pond, the cycling will slow though it’s too loud. I need to figure out how to hush it up a bit because it’s just a bit much.

Oh, and after building the drum filter, the drum itself may be modified. While it works, the plastic 55-gallon drum has a large “waste line” (being larger in the middle than than the top and bottom). As a result, the filter cloth took six hours to install due to all the wrinkles and folds. Also, watching it get blasted by the nozzles I worry about how long it’ll last. If a new one’s built it’ll have straight sides with polypropylene sheet for the ends and two layers of stainless mesh to contain the filter element, be it polyester or fine-mesh stainless. The stainless may last longer but there are reports that the wire has microscopic burrs that snag waste that the sprayers can’t dislodge. I’ll probably make it so either filter type can be used, making it so it’s far easier to replace than the six hours it takes now.

Redoing the pond has been a great education in project management and logistics. We don’t have a big yard so I can’t just pile the dirt up in one convenient place out of the way – it’s everywhere. The trick is doing things in the right order so I don’t box myself in. Oh and then there was discovering that we have roots everywhere, looking something like brown carpet under the existing pond liner. The good news is that they’re well-behaved, with them only following the liner but not growing through it.

Pond filtration has changed over the last 20 years and it was a big reason for the redesign. In the old days people used all sorts of media for the biological section. Once popular was lava rock due to its low price and really high surface area. Yeah, it worked great for several years until the bacteria plugged up all the pores making it no better than gravel. People also used pea gravel, again cheap, and an absolute bitch to clean due to the weight (and have you ever tried shoveling rock?!). Over the years, plastic media became popular and “bead filters” were the hot thing for a while. Picture an huge hourglass full of plastic pellets, periodically churned to clean. That worked fine until the pellets got so stuck together that cleaning no longer worked. The most recent iteration is a “moving bed” filter, basically 1000s of barely-buoyant plastic objects in a tank constantly being churned by air injection. It works really well because the air injection supplies the oxygen the bacteria need anyway, the constant motion ensures that the water flowing through the bed is fully exposed to the media instead of taking the shortest path from inlet to outlet, and the constant agitation knocks off the dead bacteria resulting in no channeling or clogging. Shown is the biological container in operation and what one of the bits of media looks like. It’s just a surface upon which beneficial bacteria can grow; an additional 12-cubic feet of the stuff is on order.

Anyway, to keep this remotely car-related, walking into the garage tonight Midlana looked rather menacing with the ducting so a picture was in order. It’ll look much more finished once painted and I know I keeps saying I’ll get around it but with the holidays past (businesses back to work), expect that the ducting will be dropped off at the painters soon.

23 Oct 2016

Bah, it took all day to sand the plug so there was only time for two coats of wax before the day was over. This type of composite construction doesn’t technically require a smooth surface on the plug – which becomes the inside of the final product – but because I know it was rough it was smoothed down any  way just because. There are still small low spots sprinkled about but they’re gentle and unlikely to affect airflow.

After applying the second coat of wax, the car was taken out – it was nice to drive her again! For the first time in over a year, gasoline was added, diluting the ethanol roughly 50/50 just to see how it ran. It should have run fine but even with an ethanol detector it was good to see that it still does no problem. There are plenty of back country destinations around here that don’t have ethanol* so it’s nice to know the car handles either. Also, 85% ethanol (“E85” around here) cuts fuel mileage by about a third, a significant dent in range, so knowing gas works is reassuring when low in the middle of nowhere.

*I don’t understand that lack of availability out there, the desert attracts all kinds of people with crazy off-road vehicles and the higher power ones run on E85 – which owners have to transport from home. Why the small towns out there don’t take advantage of that I don’t know. Instead, E85 pumps show up here in larger towns where it’s great for people like me, but makes zero sense for anyone else, short of those wanting to “buy American” at a premium,  since it’s not cheap enough to make up for the decreased range. I suspect there’s some sort of kickback thing going on with stations to install them because the only cars I ever see at the pump are one strongly suspected of running tweaked engines. It’s always fun to ask the STi and Evo owners about that and listen to them swear they’re stock. Of course they are 🙂

22 Oct 2016

Coated the foam plug of the largest section. Tomorrow it’ll get sanded then covered with mold release wax, though how long that’ll take depends how smooth the surface turns out, and how long each coat of wax takes to harden. If time is short then the first two sections – also coated today – will be sanded to see what’s what, or said another way, to see how many go-arounds of epoxy/micro and sanding might be expected before it’s deemed smooth enough to paint.

With both sections set aside to cure today, the car was warmed up (for the first time in a long time!), then an oil change done with full synthetic (mineral-based oil was used for break-in). The car was vacuumed out to remove as much foam dust as possible because I don’t want it blowing into my eyes while driving (never mind inhaling the stuff).

The last picture is a used “magnehelic” meter I picked up off ebay. It’s a very sensitive (1″ H2O) meter for measuring pressure differences, useful for testing airflow in ducts, intercoolers, radiators, but especially for determining proper placement for vents, measuring diffuser effectivity, airflow around wings, all sorts of interesting goodies 🙂

20 Oct 2016

Ordered enough fiberglass cloth to ensure stock for the last and largest subassembly. There’s enough carbon already on hand that should suffice for stiffening the larger surfaces (by the way, when fabricating, curved surfaces in composite or metal are much stiffer than flat). Expect a lot of progress over the next couple weeks, and after endless sanding, it’s on to paint!

In other news, Lulu is running a 30%-off sale through 24 Oct, code “OCTHIRTY”, book links: coil-bound and regular-bound. Lulu doesn’t notify authors when they have these promotions but I try to pass them on when I see it.

17 Oct 2016

Spent the day touching up the two front pieces to get them ready for a thin layer of epoxy/micro – a layer easily sanded in preparation for paint. Part way through that I figured the middle section should be riveted on as part of becoming a permanent assembly, so that was done. Then I figured there’s no point in fully prepping the front sections with the intercooler section still sitting there in foam.

First up was making a proper aluminum frame that’ll become a permanent part of the assembly, a frame so I don’t have to worry about the attachment points tearing out for whatever reason. It also allows a close tolerance on the fit-up between the composite and intercooler to minimize leaks. With that done, the foam was pushed into place and (not shown) a few areas had epoxy/micro added to bond the aluminum to the foam for now.

In the last pictures is this little guy I found at work  – “little” being relative because a praying mantis is really big for a bug at about 3″ long. It was really cool how how his head could turn just like a person’s and it was intimidating how he’d turn his head and track my movements. In this picture he’s looking at the camera upside down.

25 Sep 2016

It went pretty well. The first shot is the waxed plug ready to go. The second shot should have a caption of “Who says hot chicks aren’t useful in the garage?” Sorry girls, but you’re only useful as a pattern for cutting the fiberglass and carbon plies. When it was all said and done, I give the first part a “B+”, pretty good though I forgot to angle the plies so it more easily bent around the sharp back edge corner of the roof. As a result I had to clamp the plies which of course glued my clamp to the material. It was fun trying to remove it without cracking anything but eventually succeeded. The good news is the goofs are functionally harmless though I know they’re there. The foam plug was removed very easily thanks to the wax and in fact about a third of it came out intact. The end product still needs some work to prep it for paint but it’s very usable. A big thanks to Dave for instilling his years of composite experience and trying to keep me from going too far wrong, and the next two parts should go much smoother.

After the first piece was done it was on to the second, which required some thought due to how I made the “sort-of” plug. That is, the plug isn’t exactly that; parts of it are permanent and other parts are expendable, due to my inexperience. After staring at it a while, it can be used as-is but doing it over it would have been done differently, albeit ending up with the same product. The “icing” (epoxy and micro) is on the parts which either stay or get hollowed out, and the parts of it without will be dug out more to allow the inside surface composite layer to flow out to the outer layers. Pictures as it progresses.

In other news, one of our koi fish died so I bagged it, tied it up tight and put it in the trash can. That would have been fine had it not been for our hot weather and the smell was something that has to be experienced. I needed to get rid of it so went to put it in the truck and just reaching into the trash can was enough to imprint the smell into my clothes, nearly as strong as a skunk – ewww. Anyhow, I’m sure the shoppers at our local shopping center appreciated me putting it into the trash can in front of the grocery store…

18 Sep 2016

The third and smallest subassembly was completed, connecting the inlet duct to the intercooler assembly proper. Several pictures show the overall flow – I’m pretty happy with how the three turned out, somewhat subtle yet all-business (granted, a lot of finish-up work remains on them).

The second-to-last picture shows how well imperfections show up with the good lighting. All of this was corrected by simply using my figures, more on that below.

I’m trying something that might work great – or might not. Instead of following advice from composite experts, I’m going to: smooth out the foam plug, coat with epoxy/micro, let it cure, sand smooth, coat with mold release, cover with fiberglass. The theory is that the epoxy/micro will fill the voids in the foam, resulting in a much smoother inner surface when the foam’s removed, which the mold release will theoretically allow to just magically happen. Guess we’ll all see how well this works.

I’m debating whether to: treat the foam/roof panel as a surface upon which to lay-up the fiberglass, then pop the finished parts off the roof for finish-up work, or lay it up permanently. I started with the smallest (and least visible!) subassembly and plan to pop it off for cleaning after it cures. The inlet section though is more challenging because of how I made the plug – perhaps incorrectly in hindsight. Some of the foam is (or was) intended to be become a permanent part of the final assembly, specifically the foam ahead of the inlet proper, along with the stiffening ribs. This type of foam isn’t structural – it’s very delicate, just running your finger across it visibly removes material. It’s great for forming and finishing, not great for bearing any weight; in fact it’s why the assembly can’t be vacuum-bagged because the 15 psi of air pressure will completely collapse the foam (pretty proud of myself for figuring that out before finding out the hard way)! Meh, a solution will present itself one way or another.

Oh, and I’m using slow-cure epoxy so I don’t rush. I’ve used fast-cure and they aren’t kidding, in a matter of minutes even small batches get warm and it must be quickly spread and the composite applied. The slow-cure allows plenty of time to make sure each layer is in place and any bubbles dealt with. The only downside is that once it’s applied, work on that part has to stop for the rest of the day.