You Already Know This

April 19th, 2015 by Q

wingbracesolid188 Strength versus weight…. the common belief is that you can drill holes in practically anything to make it lighter, without compromising strength. We drill holes in everything, and seldom worry about the strength of what is left behind. A lot of the time it doesn’t matter.

Sometimes strength and stiffness is critical however. In that case, is it true that drilling holes is the best way to add lightness? I recently made some simple bolt together bracing for a car with a very flimsy body being asked to hold up a wing with the potential to make a lot of downforce. The braces consisted of some frame and body side mounting brackets and with aluminum bar connecting them. Simple.

And keeping it simple, what is the best way to keep things light while maintaining enough stiffness to prevent the area surrounding the trunk from caving in on itself at speed?

The easiest thing to do is to cut the bars out of solid 3/16″ aluminum. Pulling out my nameless auto cad program and doing some testing, we approximate about 1mm of displacement under full load, and a weight of 195grams, which came in a little bit lighter than the actual piece I constructed for test fitment on the car.

wingbrace6x1.25The obvious thing to do is drill some holes. Note that my testing showed that hole size has a pretty big impact on the result. A single column of larger holes does a better job of shedding weight while maintaining strength, than does multiple columns of smaller holes. Of course, making the holes too large seriously compromises structure. The best balance I found was 1.25″ diameter holes through the 2″ bar.  The weight savings were 33%.  Stiffness and strength drop 27% and 8% relative to our solid bar.

wingbracesolid125That seems like a reasonable step forward, but if weight savings are important, what about starting with thinner bar? Shaving 1/16″ off the solid bar lends very similar weight savings (35%), but a much larger 34% drop in stiffness and 32% drop in strength.  As you might expect, considering “strength versus weight” there are negligible gains.  But if the structural requirements allow, it’s a much simpler way to save weight than starting with something big and cutting holes in it.  Obviously.

wingbracetrianglesIf this was a race car that was professionally built and engineered 20 years ago, you might see something like this. Weight savings are 46%. Strength falls only 20%, but stiffness is a disappointing 48% that of the solid 3/16″ bar. Stiffness is what is critical to this application, making this design an obvious dump. However, it is also not nearly as easy for a hack in his garage to create. Possible… but, not as simple as just drilling a few holes.

graphI played with a number of different designs, and was somewhat disappointed (although not surprised) that none of them provided better results than a single line of holes. That’s not to say its the ultimate: 3D structures can easily provide superior stiffness versus weight. When we have limited machining tools/skills however… and hole saws are cheap… making swiss cheese of everything gets really tempting.

Back in grey

April 8th, 2015 by Q

debut2015My new shoes are boring and grey, but somehow I like them. If no one ever noticed my car that would make me happy. Any attention is bad attention.

If you think my car looks hotter than ever though, despite the boring wheels, you’d be right. It is. As a result of a recent revision to SCCA regulations, the FRS community was alerted to the legality of TRD optional springs and sway bars for street class. Always looking for a reason to modify my car, I immediately ordered said parts. They come with a mild rise in spring rate, and a 1″ drop. The stock suspension on these cars is already really quite good and I can’t say I notice any difference in performance on the street. Having not yet had my car to the track, I’ll never be able to compare to the stock parts, but at the very least, it looks better.

The disappointing part, is that the car woke up with an oil leak. The smoke rises from the rear passenger side of the engine where there is reportedly a cam seal that fails often. Looking quickly, I saw no residue of oil anywhere and was too annoyed to look further. The car has 3300km on it at this point, and none of those km have been hard by any even my grandmothers definition.

Subaru. Fail.

Shatter Proof

April 6th, 2015 by A


Flaps aren’t just for mud, but they aren’t just for winter either. And they certainly aren’t just for the look. Even an STI that remains gridlocked on Calgary pavement for it’s entire life will still throw slush/gravel/salt up onto the body for about 9 months a year…

The first generation Legacy came with mudflaps from the factory. But the seam where flap ended and fender began was a notorious spot for rust to brew.

I trust that Subaru has better paint/metal/finishing methods now than they did then, but I wonder if they would still have the same rust issues if factory-flaps held grime against the fender for 20 years.

So what then.. We can’t live with them, and we can’t live without them?

Countless chips in the rocker paint say I can’t live without them.

And an already flawed wheel well makes me fear that the fender can’t live with them.

I suppose the answer is obvious -> Maintenance.

A “set it and forget it” approach might be OK with flaps that are installed very carefully when the paint/undercarriage is flawless.

But in my situation, the approach will likely need to be:
Install for winter -> Clean thouroughly often -> remove for spring cleaning -> address and treat with Por15, rocker guard, and curses accordingly -> reinstall for summer -> cross fingers firmly

In any case it is always satisfying to make flaps from scratch.

We learned our lesson about trusting in names a long time ago.

And the Popular Armor that is out there just seems too dainty (and expensive) for my needs.


I started with an online template of the best selling stuff. But before cutting the rubber I played shadow puppets with cardboard and realized bigger is better since I am trying to protect both the inside and outside of that area. I just hope I’m not trapping a lot of junk that high jumps over the flap and gets caught at the mounting points.

I guess I should add ‘full flap flush’ to my bi-weekly service schedule….


Worth Even A Stack of Dimes?

April 5th, 2015 by Q

earlymigLast year, I had a brief love affair with aluminum. I bought a spool gun for my old Lincoln 180, a canister of argon, and went about learning how to weld and work with aluminum. It went so well in the beginning, that I imagined making everything out of aluminum. Nevermind intercooler piping and catch tanks, I was thinking struts, suspension arms, engine mounts… I wanted to make a whole car out of the stuff.

Aluminum was cheap, so easy to cut and shape that giving it the stink eye practically put it into shape, and it welded together with nice clean dimes that made me feel like a hero.

And then one day, something broke.

6061sucksYes aluminum needs to be heat treated. I had been buying 6061 T6 stock from a local supplier (T6 indicated the level of heat treating/artificial aging), and realized that the act of welding essentially voided that designation, but it was no big deal.  Heat treating, treat eating… Most of what I was doing at the time wasn’t really structural, and if it was I just built it extra big and strong. This simple welded bracket that broke though, was extra big and strong.

It was a wake up call, with very small consequences, but I must say it totally annihilated my confidence in aluminum.  Now I’m trying to realize a little better than value of heat treating.

I must admit, I’m no engineer, but as I look into this further one thing is crystal clear.  If you weld aluminium, heat treating is absolutely critical.  I threw 4130 Chromoly on the chart above for another point of reference.  4130 is a whole other thing to understand, so lets compare instead to plain old, commonly available 1020 steel.  6061 T6 is roughly 33% as stiff as 1020, 75% as strong, and 33% the weight.  So again, not being an engineer, if you wanted to built something out of 6061 T6 that was as strong as the same part made from 1020, it would be only 44% as stiff, but also only 44% the weight.  If stiffness is critical, the part would be essentially the same weight, but more than twice the strength.

Of course I’m over simplifying things as there is more to it than that, and aluminum parts can be designed differently than steel ones because of material properties… but I’ll save that for the engineers.

The real point is, 6061 that needs heat treating (i.e. 6061 T6 that has been welded) possesses the same stiffness and density, but only 20% of the strength of 6061 T6 (approximately 15% of the strength of 1020 steel).  In other words, give your sister some popsicle sticks and glue and see who can build the better part.

Now, I have a shop in my contact list that can heat treat the welded parts I make, but that severely impacts both the affordability and simplicity of working with aluminum.  The better alternative is to get things CNC cut from solid chunks of material and avoid welding all together, but that introduces other costs… although it does also allow for the use of some better aluminums that aren’t as easily weldable.

For now though, the summary is: even run of the mill plain ass steel > aluminum.

No Humping. Straight Porking.

March 30th, 2015 by Q

KAAZPressureRingsDrivetrain mods are for max winning IMO. Nothing you can do to your car is more under rated in it’s impact on performance and drivability… particularly differential set up. AE86 was a very simple lesson in this. A stock [Canadian] AE86 with an open diff is basically no fun at all. Tall and soft, it teases you often by playing like it’s going to take a fun set, but when it comes to actually holding a line, the inside tire explodes, the outside takes a nap and you decide you’d rather be a hockey player than a driving hero. Alternatively, you might spend your months wages on a trick new KAAZ LSD that all the hachiroku gods have told you that you need, and then go out there and feel like no matter what you do the car just understeers.

I’ve played with number of limited slip differentials, and from my experience, NOTHING compares to well set up clutch type diff. Yes they wear out, yes they chunk and chatter, and yes they are typically more expensive than helical type diffs, but they are not active/reactive in their torque transfer, and they do the same thing… over and over again, and if you don’t like what they are doing, just adjust them. Viscous humping is cool for pre-pubescent school kids. Helical diffs are fickle women who change their minds every half second. Yes carburetors are deprecated pieces of crap, but like the small block Chev, the salisbury type diff is still king.

What you are looking at, are the pressure rings and 1.5way cam opening for a KAAZ differential. Those hachiroku gods were not wrong to sway me towards buying a KAAZ unit as my first LSD 16 years ago. KAAZ offers great customer support, very good pricing and excellent adjustability in initial torque and total lock up with the purchase of different plates and shims. From my limited experience however, CUSCO units seem to be the bling piece, packing more clutch plates, and center sections with not only adjustable preload, but optional cam profiles as well. I imagine that a CUSCO unit will end up on my ZN6 at some point. This particular unit is for another car…

KAAZcomponents… a dedicated autox car. Although autox is probably the most common place for a an LSD like this to be used, stock settings are typically geared much more towards the higher speeds and larger radius of a road course, with stronger lock up and initial torque settings. While that can make a car feel stable at 150km/h, at 50km/h it tends to make a car feel quite invariable in attitude, unwilling to rotate… especially if there is a lot of weight over the drive wheels. This is the both the greatest strength and weakness of the clutch type diff versus a helical diff. Cars with a helical diffs tend to be a lot easier to rotate, but much harder to manage mid corner. A clutch type diff on the other hand, provides a lot of stability and predictability mid corner, at the cost of some nimbleness.

KAAZxsandos As I said though, clutch type diffs, tend to be easily adjustable. This KAAZ unit comes with 6 clutch plates on either side of the center section. These clutch plates alternate splining to either the casing (X) or center section (O). With six plates there are five friction surfaces that are activated under torque. X-(1)-O-(2)-X-(3)-O-(4)-X-(5)-O.

Without purchasing additional parts from KAAZ, these plates can be re-arranged to “soften” up the diff. Arrangement XXOXOO gives X-X-(1)-O-(2)-X-(3)-O-O three friction surfaces for a 60% (3 out of 5) setting. Other arrangements are possible. XXXOOO would be a 20% setting. Or we can transfer plates from side to side to get other settings. Instead of 3 X’s and 3 O’s on one side, we might have two X’s and four O’s on one side, four X’s and two O’s on the other. That would allow for XOXOXX on one side, and OOXOXO on the other for an 80% setting.

Note that this is a simple large scale adjustment, that will affect initial torque and over all lock up. For autox, and in general this is probably a fair solution, but other scenarios might call for a lower initial torque relative to braking and accelerating torque or visa-versa. In that case, KAAZ offers different cone springs, or additional clutch plates. Parts are quite affordable, and the entire units are also easy to disassemble. Think and play.

Dual Sport

March 27th, 2015 by A


Not to disappoint
Randedge, but I want to make
stuff as clean as this.

Turn Down for What.

March 20th, 2015 by A

There are a lot of neat places to explore by driving around Alberta. And although most people think to go touring in the summer, technically there are more surfaces that will hold vehicles during the winter. All that is required is a vehicle that you don’t mind going off the road with.

This was on a lake not West of Calgary. And a surface not smooth. And on suspension not high enough for my Canadian Tire “racing jack” to fit under the car. Causing me to high center myself before I had finished my morning coffee.

The ‘goofy luggage rack’ that is still waiting to be filled with wild game is the most outwardly noticeable indication that this car has transferred ownership. But in my mind, the most absolute indication that I own this car now instead of Q is the fact that I have turned down the performance by raising the ride height.

The Ohlins were sweet on the cone-courses last summer, but they were not OK for winter driving conditions, regardless of road. And since I go off pavement and off grade altogether, I need something all-purpose.

A used set of Version 7 JDM STi shock&spring combo fits the bill and easily fits the wallet. Days spent on lakes and logging roads has left me satisfied and impressed, respectively. On the ice [where limits are routinely pushed] tire pressure still makes the biggest difference. On the backroads [where limits are intentionally not approached] humps, lumps and severely cambered corners are all handled with authority.

Which reminds me: When people ask what suspension the car has, the simplest answer is “just STi struts” – but that is also misleading. Because the tune is what makes the difference. Control arms, camber bolts, sway bars, polyurethane bushings…. each component only amounts to marginal change on their own – but packaged together they produce a recipe as enjoyable as boxed brownies [very].

It still feels like the WRX is Q’s car, and I am just doing my thing with it. Not sure what will have to happen before it feels like it is my car. Maybe it never will.

Turn Down for REAR copy