Aero Crosstour- Squeezing blood from a stone
 

Howdy, all. I'm setting up a build thread for my aero Honda Crosstour project. This post will be a data dump of everything so far and then I'll use it as my primary build log.

Specs, goals, and completed projects will be updated on this first post. All other journal entries will be new posts.

Currently, MPG is roughly 21 MPG. Regularly updated MPG log is here: https://docs.google.com/spreadsheets...it?usp=sharing

Crosstour curb weight is 3852 lbs. With me and 16.5 gallons of gas weight is calculated to be 4142 lbs.

The Crosstour has Honda's Variable Cylinder Management (VCM) and a Tire Pressure Monitoring System (TPMS).

Goal is an average of 25 MPG (16% increase in MPG)
I expect to have to do more than current prosposed projects to meet this goal. Recommendations welcome.

Wheel covers
Lower radiator blocker plate
Foglight covers
Low profile side mirrors
Lol
Scrum board: Kerika
Image analysis for front surface area cross section:
ImageJ
EcoModders aerodynamics stencil:
https://i.imgur.com/9GeevKH.png
This is the image that inspired a possible rear spoiler.


So, first phase is just tooling up and ordering parts to complete the project.

I installed a Harbor Freight drill guide on my old IKEA desk for drilling the lugnuts. To install the guide, I drilled holes slightly smaller than M5-.08 nuts and epoxied them (SteelStik) into the desk for a semi permanent installation and to improve stability. I used a thread tap to clean the still-wet epoxy out while leaving the max amount of epoxy still in place.
https://i.imgur.com/m0kt5sJ.jpg
https://i.imgur.com/SIHK168.jpg
https://i.imgur.com/klOglGM.jpg
Ya'll like Patrick Nagel?

I also designed and 3D printed a simple centering jig for the lugnuts. It's the black thing holding the nut beneath the drill bit. If I let gravity center the bit, it gets pretty darn close to perfect. I'll verify with measurements before drilling.
https://i.imgur.com/rkImgDu.jpg
Side note: I'm considering using steel zip ties vice drilled lugnuts. I'll have to see how centered I can make everything and go from there.

Other than that, I printed some organization tools for the shop and built a scrum sheet for the build using Kerika: https://kerika.com/app/BQBC4/board/BQBDh
Also printed a drill bit sharpener for a Dremel so everything goes smooth when the time comes.
https://i.imgur.com/8F5GNBT.jpg
I'm also planning on fabricating a vacuum molder but I don't have photos for that. This will be how the foglight covers are made from clear Lexan.


First project tackled was wheel covers, since the pizza trays came in first. I wanted to passivate them, as they're pretty cheap and I don't trust the aluminum any more than I can throw it (which is probably pretty far, actually).

Technique was, well....bathtub. I used what was left of the Barkeeper's Friend under the sink and the minimum amount of water needed to cover them.
https://i.imgur.com/OTnS2mb.jpg
They soaked for a little over an hour with hot water.
It will be hard to tell how well it worked but I'll definitely be monitoring for corrosion and taking notes.

I'm leaning heavily towards steel ties vs. threaded lugnuts. Utility, ease of install/modification both seem better. Aesthetics won't be any better or worse.

Next big project is the side mirrors. I wanted to do a full elimination with a printed vanity cover but the Crosstour's enormous blind spot makes it a non-starter.

I'll be using ImageJ to try and get an approximation for how much the new mirrors take from the front cross section. I jumped into this project head first and the passenger mirror has already been removed, so I'll cut and paste the driver's side mirror into the image to get a closer estimate.

Here's a side by side to show how much reduction there will be. Not amazing but it's something.
https://i.imgur.com/v4dO5id.jpg
ImageJ will hopefully give some hard numbers on how good it was but, of course, final MPG is king.

Speaking of which, I'm going to intentionally pace project implementation such that I can get a couple of tanks of gas after each mod. maaaaaaybe do A-B-A.

I'll also have to build a plate to cover the old recess that the side mirror was integral to. New side mirrors bolt onto an adapter plate (which is for much older Hondas) and the adapter plate bolts to the car. I'll have to fab an adapter plate from this:
https://i.imgur.com/rGftTXz.jpg
Which really shouldn't be too hard. It's a triangle with no major curvature. Measurements are easy to take with a caliper. The three bolts are what penetrate the unibody to mount the mirror. The plate will be 3D printed and those bolts will be replaced with heat set threaded inserts. I'll threaded a bolt into them from inside the vehicle.
Image of inserts:
https://i.ebayimg.com/images/g/orUAA...Ll/s-l1600.jpg


Image of the scheme for the mirror mounting hardware:
https://i.imgur.com/TDYynbw.jpg

Next minor project is replacing the front emblem. This is more of an aesthetic thing but it'll hopefully help with some aerodynamic smoothing, since the Honda emblem is pretty big and cavernous.

I work as an intern at an aerospace firm and they throw away A LOT of carbon fiber. This benefits me immensely.
https://i.imgur.com/wLfwmgr.jpg
This is the CAD mock up of the emblem. Pretty easy. I also own a crock pot.
https://i.imgur.com/55NhmNz.jpg
It turns out, the rivets holding in the factory emblem were either missing or sitting in cracked, ineffective housings. So, it popped right out. Anything will be an upgrade after that, I suppose.

And that's it for now. Hopefully, I'll have an image for ImageJ before the end of the week and can have an adapter plate for the mirrors fabbed up in CAD (real CAD, not the cardboard kind. Okay, it'll be TinkerCAD but still). I also want coast down times but the friend who was trying to help is having bad allergies and I may forgo it or do it myself by filming the speedometer.

Let me know what you think. All suggestions are welcome and appreciated!

Impressive!
Having a supply of free carbon fiber plates and maybe oddly shaped parts that could just fit somewhere is a real treasure.

Have you considered raising the tire pressure a bit? The first few PSI above OEM tire pressure have most effect on economy. The higher you go the lower rolling resistance will be, but the extra gains will quickly get very minimal. Still, an easy 5% gain.

It is standing fairly high, so lowering it or a front air dam would help. Also, extend the flaps (wheel air dams?) ahead of the front wheels a bit and fabricate them for the rears.

Finally, consider the 65+ mods and 100+ hypermiling tips. Tightening the nut that holds the steering wheel is my favorite ecomod. ;)

YEAH, some weird stuff comes out of the composites shops, not that I'm complaining. I hope a nice, wide sheet gets thrown away for the rear spoiler but we'll see.

I'm going to be very, very careful with tire pressures. I'd rather have the traction and lose a few %. I don't know where those two things start to rub against each other but it would be a last or, more likely, never resort. That thing is heavy and needs everything it can get.

I'm dying to lower it. I know a few places make well designed springs that keep your steering geometry right while giving you a couple of inches. That's more restricted by a student's budget than anything else, but it's definitely on the radar.

It has a front under tray that goes to about the rear of the front wheels. Is that what you mean by front air dam? I'd like to get into put more trays under the car but that would come after this first round of mods.

And can you point me to an example of extending the front "wheel air dams"? I think I know what you mean but not sure.

And boy, let me tell you about that steering wheel nut. Right now, I'm running supplies around town for people who sew masks for medical workers. My habits are utter garbage but I definitely intend on returning to the "leave earlier, drive slower" routine as soon as I can.

By extending the "wheel air dams" I mean extending them down.
It probably is better aerodynamically to shape a cowl to gradually lead the air away instead of just a vertical slab, but much harder to fabricate - and you don't want to damage other cars should it come loose on the highway.
I once lost a tiny piece of bagel box plastic (a 5" wing shaped sleeve around my antenna stalk) translucent and all, and the semi behind me went full on the brakes... An enormous overreaction, but I felt bad about it anyway.

The thing with tire pressure is that each car is different and you need to find out for yourself what works best. Find a place to test cornering and braking, bring a pump.

FWIW, my observations:
I run about 42 PSI in mine (OEM is about 34) and economy went up about 5-8%.
There's much better cornering with less tire squeal. It was rubbing the sidewalls in hard corners at OEM pressure, which I definitely do not want to happen. Higher pressure means less sidewall deformation and less thread warping in corners.

I have one corner with a wide runoff that I use to test cornering speed. That's 52 km/h max at 34 PSI for my OEM Bridgestone summer tires when new, and 57 km under similar conditions at 42 PSI. That's about 9% more lateral force that the tires can handle!

My car's straight line stability is easily affected by side wind, and higher tire pressure all but eliminates that - to the point that I know I need to check the pressure if I notice the effect; especially on the rears.

Braking in a straight line seems unaffected.
Road noise went up, sadly, and going over 42 psi makes the ride less comfortable. That's the same for all 3 sets of tires I have used on my car so far.

.

Do you have a instantaneous MPG gauge in the vehicle ?

If not, you need one... It’s the most important mod... Period... ;)

(Scanguage, UltraGauge, Torque Pro or similar app.)

It allows the steering nut to be adjusted and also allows one to evaluate vehicle modifications quickly and easily.

Increasing tire pressure to the max psi as indicated on the sidewall can easily net gains of up to 10% without losing drive ability and or safety. Use the search function to research, as this topic has been covered many times. A member here (CapriRacer) has written or replied extensively about tire related topics. If I remember correctly he works in the tire industry.

I would take care of the low hanging fruit first before doing suspension mods.

:)

>

I'll definitely consider playing with tire pressure, now that you framed it that way. 🤔 I'll definitely look up Capri's post. I'm at 33 out of 44 psi, I'd be flabbergasted if I didn't lose some traction between here and there but, like you said, there could very well be some room for no-cost improvement. If I only I had a skidpad......

*looks at empty Walmart parking lot*

And yeah, springs would come after everything else. I did find some from Megan that aren't expensive and may be worth it to experiment. The tricky part is no one makes just a spring for the Crosstour, explicitly. Everyone I've seen lowering Crosstours gets the Accord spring with the highest spring rate, which I think is for the V6 hatch from the same year. So, I'd need to be able to get spring rate specs on the spring, which shouldn't be too hard.

And I'll look up those instantaneous MPG apps because what the heck. I've been using the VCM eco light as a bellwether for now. 😄

How about the obvious: re arch your existing springs. Places exist that do this for less $ than new unless this is a weekend only thing

I thought you could only re-arch leaf springs? Never heard of re-arching coil springs.

Yeah, the Crosstour might look like a van but it is indeed a unibody with coilovers all around. I couldn't find anything refering to re-arching springs. There's always cutting them but I wouldn't do that to the car that drives me to my funeral.

I did, however, find out that Megan Racing makes a set for same-gen Accords that clock in below $150. Just need to check fitment and spring rate and that will probably be the solution.

Journal Entry 04/09/2020:

So, not much time for progress during the week but got some small stuff done.
Cuts for the vacuum molder were done by a good Samaritan on NextDoor.

https://i.imgur.com/0ZSiDto.jpg

It's funny how simple of a plan a vacuum molder is. If you've ever wanted to use one, here are the plans I'm using:Amalgamized: DIY Vacuum Former for Thermoformed Plastic in Your Home Shop

I'm scaling mine up to be a little bigger for the fogs but you can make them just about any size, assuming your vacuum source can do the thing.

Got a tracing of the mounting plate for the side mirrors. Should get me in the ballpark for a prototype. I'll print the prototype at, say 3% infill and just see how it fits. If it's the right size, I'll print the flight model from HTPLA, bake it, and starting mounting hardware.

If I'm lucky (I'm not), I'll be able to use that factory rubber piece to seal the interior up nice and neat. Otherwise, I can get some thin rubber sheet to do it with.

https://i.imgur.com/QGM5dG7.jpg

Quick plug for carbon HTPLA. Proto-pasta makes this stuff and it's very neat. It's an HTPLA you can bake in an oven to increase hardness and heat resistance. They claim temperature operating limits of 310 F after baking. I made a shift knob with this stuff for my old Toyota MR-S. I was able to cut a thread in it and leave it in the sun permanently with no deformation.

My limited memory tells me they formulated it to crystallize while baking to achieve the higher strengths and heat tolerances.

I would use some spray-on filler, black spray paint, and polyurethane to finish out the mounting plate for the side views.

This is also where the limitations of TinkerCAD are going to stop me short. TinkerCAD is a VERY rudimentary free CAD program. You can do a lot with it but it's more like putting together pre-assembled blocks that it is true CAD. You can't really make a triangle with specified face lengths. You can make a pyramid and a cone and a triangular prism. But not a triangle with specific length faces.

Anyways, there's plenty of more sophisticated, free CAD software out there I can use.

I've used Wing 3D previously. It has a work flow that was better for me than others, but eventually I found things not working as advertised.

Blender has a steep learning curve, but is improving at a rapid pace. They abandoned right-click to select for the industry standard left-click a while back. There're lots of tutorials on Youtube.

These little fellers?

https://i.imgur.com/P7m9nP0.jpg

Search term is 'wheel spats".

Yup, yes. I bet they work by forcing the air around the turbulence ahead of the tire instead of mixing with it.

Also worth considering...
The wheel air dams / spats are deflectors then ? ;)
I'd think spats go behind the wheels anyway. Where they generally cause extra drag.

Be careful with the aero template. I recall there being two versions floating around here:

https://ecomodder.com/forum/showthre...ate-33673.html

https://i.imgur.com/OagWvFZ.jpg

Here's your vehicle overlayed with this version:

https://i.imgur.com/sOwVBT5.png

I'd get a second opinion but this is really low hanging fruit, especially for a heavy vehicle. Most gains from raising pressure come with the first few pounds.

My uneducated opinion: the tires on the Crosstour have always looked stylistically oversized to me, though I had no idea they were that heavy.


You're in luck - your car has double wishbone suspension all around, which (typically) means your camber won't be impacted by changes to ride height. In a pinch, you could just cut a quarter of a coil off and see where it lands, rinse and repeat. To properly lower you would ideally go stiffer as well (to prevent bottoming out). I have no idea how much travel these have.

Never cut coils. That causes more problems than it solves. There are spring clamps available cheap. Or you can use u bolts

https://www.eatondetroitspring.com/c...-coil-springs/


Just as an example, on the rear of my car, I replaced the stock springs with stiffer ones that maintained approximately stock ride height. I then pulled close to 200lbs out of the back of the car, and it was sitting about an inch and a half higher.

The springs were tangential end springs. I took a quarter of a coil off, and it was still a bit high, so I took another sixth or so and brought it back down to stock ride height, albeit much stiffer. They seat the same way when cut. When lifted, the springs are still considerably compressed so no risk of them falling out. I see no problems with taking a small amount off so long as you pay attention.

-Don't cut springs if it will change how they seat.
-Be mindful of suspension travel - going stiffer is a good idea, you don't want to be hitting your bump stops.
-Don't cut so much that they are no longer compressed when the wheels aren't on the ground.
-Be aware that with some types of suspension, you'll mess up your camber.

Thanks for that new overlay, Ecky. Yeah, I've gotten a few different comments on the overlay. In any event, I'm not expecting much, if anything, from that back end, short of going full kammback (which I won't). At this point, I more focused on smoothing out the front clip and reducing front cross section.

Definitely gives even more pause to the idea of any kind of spoiler, not that I'm complaining.

And I'm open to cutting springs, just because I know you can, but you can get a nice, high spring rate set from Megan for about $150. That lets me keep my spares in case the rate drops the car to the ground. Cutting springs is a bit of a "get it right the first time or you're screwed" affair. If it was a dedicated track car and I was on a budget, it'd be game on.

I was blown away to hear about the suspension. I like this car about 25% more now.

Minor Update: 4/13/2020
Going through the beginners' tutorial of FreeCAD. I think I can get a good mounting plate that requires minimal drilling or heat setting to fit and mount the new side mirrors. This will be the only project I work on this week. I need my side mirrors back, badly.

P.S.-
I bumped tire pressure up to about 37 psi in the middle of a tank. Got 23.12 MPG, which is likely lower than actual since I did it in the middle.

Interesting. I would kill for a skidpad and an accelerometer but this ain't MIT, baby.

I've dropped two Superbeetles. I sold one on, and when I did the 2nd one I requested the shop use the same (stock) spring rate but increase the anti-roll bar. Instead they used super-stiff springs and the stock anti-roll bar.

When the opportunity presented itself, I had them replace the stiff springs. With the stock spring rate and 3½" of clearance, the only real problem is ground strikes on freeway bridge aprons.

I should clarify; the springs from Megan are for the same year V6 Accord.

If I have the right curb weights, my Crosstour has a curb weight of 4070 lbs. The heaviest Accord from the same year weighs 3605 lbs.

I'm basically try to get max spring rate with the Accord springs because they're made for a considerably lighter car. As far as I can tell, no one makes just springs for the Crosstour. But everything I've read says the the fitment is the same.

I plan on digging waaaaaay more into and maybe trying to get under an Accord to really look before buying anything but that's the tentative plan.

I'm getting bored out of my mind at work, so you're all going to be subjected to my micro-updates.

I managed to get an okay-ish front cross section image of the new and old side view mirrors. They're here:
https://i.imgur.com/AcJ1RRG.jpg
https://i.imgur.com/QHRJPF4.jpg
Keep in mind these two shots are not scaled the same.
I'll probably be re-shooting the low profile mirror, since it has a lot of non-cross sectional area in the shot, specifically with the mounting bracket. Isolating it isn't hard to do with Gimp or whatever photo manipulation software you have.

I'll be using software called MIPAR that's gave me a free license until July 8 (because student) to analyze the photo and give me an approximation of he mirrors' surface area.

Their website is here:
https://www.mipar.us/

There's plenty of guides on how to use MIPAR and I read one that mentioned you'll need a scale to tell the software what area a pixel is worth. I didn't know this when taking the shots and cursed myself at first. Then I realized an 8.5"x11" sheet of printer paper features prominently in both. xD So, no worries there.

.

So what instant MPG instrumentation did you acquire ?



>

Nothing yet, on this student budget. I'm pretty much relying on the "Eco" light cutting on and off to give me some idea of what's good and what's not, for now. Beyond that, just doing MPG math after each tank and recording coast down time.

Cheapest method is an MPGuino. Parts probably come out to under $30. Easiest is a Scangauge(e). Both I find invaluable.

I've got a raspberry pi that's been sitting on my desk for a year that needs a job....

I'm sure it can be done, but I'm unaware of any existing code that works for it.

.

Achieving results.

Just as in real life, there are certain steps that must be taken in order.

They are.


Crawl

Walk

Run


Crawl - purchase mpg instrumentation, read 100+ hyper milling tips, Practice driving tips. Observe mpg instrumentation to see how it affects mpg by adjusting the nut behind the wheel.

Walk - read 65+ efficiency mods, implement basic and simple mods like increased tire pressure, partial grill block, engine blanket, warm air intake, block heater, etc. etc. Observe mpg instrumentation to see how modifications affect MPG.

Run - add a Kamm back or boat tail , Full under tray, lower the suspension, Replace mirrors with cameras, add wheel skirts, Remote engine start & stop switch, etc. etc. Observe mpg instrumentation to see how modifications affect MPG.


The one thing that they all share in common is instantaneous mpg instrumentation. It is first and foremost.

Just like going to school or beginning a career one does not start at the top. To understand something fully one must work their way up.

The money spent on MPG instrumentation will be paid back in saved fuel quickly. (See my earlier post for suggestions)


I,ll get off my soapbox now... :p



:)

>

It's probably not going to happen. If I thought it was absolutely necessary to improve MPG, I'd do it but this project has a shoestring project. And other maintenance concerns, like my awful, warped rotors.

The link between driving behavior and MPG isn't exactly complicated. I've watched plenty of factory instantaneous MPG gauges and I can't think of a single instance I was surprised by the reading. I've read the 65+ and 100+ hypermiling tips and nothing stuck out as particularly novel. There were a few things that were interesting but most of those fell into a "useful but not practical for me" category.

Several others fall away because they drastically alter the driving experience past the point I'm willing to accept. I'm looking for 25 MPG, some insight into aerodynamics, and some experience fabricating here. Raising tire pressure may have gotten me to 23 but I won't know for a little while yet. If I see improvements toward that after each mod by measuring the MPG after, let's say, three tanks (A-B-A style), I'll be happy. If there's no observable improvement, I'll consider it one of the small improvements that add up. If there's an observable loss, off it goes.

Should I presume that's OEM? That would be the low-ball baby step.

I'm making payments on an 1990 XFi that has an Eco light. Even my 1979 Dasher doesn't have that feature. They both can get 45mpg.

It is. I don't know what you know about Honda's VCM but it allows the V6 to operate on 6, 4, or 3 cylinders. It does so automatically based on several inputs, as far as I can tell.

It seems to follow an instantaneous MPG tracker's behavior fairly closely. Heavy throttle, heavy acceleration, going up hills, etc. all switch the VCM system off. Highway coasting, light throttle, coasting down a hill, etc. all switch it on.

Again, I've never looked down expecting it to be on or off and been surprised.

Well, small update on lowering. K2 and Megan both emailed me back today and said they really don't recommend the Accord V6 springs for the Crosstour, citing a spring rate that was way too low.

No surprised but they also tried the coilover upsell, so who knows? Jokes on them, I'm broke.

Looks the long term project timeline just got 'cut springs' added to it.

Question about drag coefficient for anyone who knows: I got a drag coefficient of "56/44" from a random auto blog. Comes out to 1.27

This is almost as high as a flat plate placed perpendicular to flow, the Empire State Building, and Usain Bolt.

Does that number sound right? I knew the Crosstour was tall and wide but holy smokes.

Definitely not right. Expect it to be somewhere between 0.25 and 0.40. multiple the coefficient (Cd) by the frontal area (A) to get the total drag (CdA). I'd expect that to be in the 2.0-3.0 range.

Edit: that sounds more like the weight distribution.

How strange. Of course, like most of the cars I've owned, there's no data out there. A website called Auto-Catalog claims to have a full performance review with the numbers but their site isn't loading.

Weight distribution looks much more likely.

Looks like coastdown time and MPG will be king after all.

"56/44"
 

That ratio is meaningless.Cds are never expressed like that.
I exhausted GOOGLE,attempting to find an official Cd from Honda Motor Co.
The second-gen Insight was rated at Cd 0.28.The Crosstour ought be in that territory.

Now that you mention it, do you think comparing the front cross sectional area of the Insight and Crosstour could confirm that?

I know that's a relatively small part of the equation and I'm not necessarily sure that's the area used when measuring Cd or D for autos.

It's a multiplier.

Find a black on white silhouette of the two models, apply a Gaussian Blur filter to it's max and read the RGB value for each. The difference should be proportional without a lot of fussy outlining.