Coastdown test for an aeromodded 1998 Ford Escort ZX2: Determining Cd & Crr
 

*Test performed on 05/27/2015*

I was able to get the ZX2 out to do some coastdown testing yesterday morning! This test was to determine the drag coefficient (Cd) and the coefficient of rolling resistance (Crr) of the Ecoscort, a 1998 Ford Escort ZX2. I am using the "measure your drag coefficient" at Instructables.com.

The conditions are as follows: Start/End

Time: 10:44 a.m. - 11:17 a.m.
Temperature: 77*/79*
Dewpoint: 72*
Relative Humidity: 83%/79%
Winds: WSW @ 3/WSW @ 9
Elevation: 197 feet
Air Density (rho): 1.156 kg/m^3
Engine Coolant Temperatures: 205* - 220*
Tire psi: 70 front, 60 rear
Weight: ~2500 pounds

The car has the following aerodynamic modifications:

85% grill block
Covered foglight holes
Windshield wiper cover
front & rear belly pans
Smooth 14x4 space saver wheels
Rear wheel skirts
Driver side and passenger side mirror delete

Stock Cd for an Escort ZX2 is 0.33, stock frontal area is 20.8 square feet, and the stock CdA is 6.9 square feet.

The test was a series of 6 runs in ABABAB fashion. A 7th run was done to test a common effect on aerodynamics(will say more at the end). The road I was testing on was surrounded by trees, thereby negating much of the effects of the light winds. All test were started at 55 m.p.h. and coasted in Neutral for 60 seconds, recording the speed every 10 seconds.

t=0 is when time was started. It is always at 55 m.p.h. t=1 is 10 seconds, t=2 is 20 seconds in, and so forth.

Run #1
t=0 - 55
t=1 - 49
t=2 - 44
t=3 - 40
t=4 - 36
t=5 - 33
t=6 - 28

Run #2
t=0 - 55
t=1 - 50
t=2 - 45
t=3 - 40
t=4 - 36
t=5 - 32
t=6 - 30

Run #3
t=0 - 55
t=1 - 49
t=2 - 44
t=3 - 39
t=4 - 36
t=5 - 33
t=6 - 30

Run #4
t=0 - 55
t=1 - 50
t=2 - 45
t=3 - 41
t=4 - 36
t=5 - 33
t=6 - 29

Run #5
t=0 - 55
t=1 - 49
t=2 - 44
t=3 - 40
t=4 - 36
t=5 - 32
t=6 - 29

Run #6
t=0 - 55
t=1 - 50
t=2 - 45
t=3 - 40
t=4 - 36
t=5 - 33
t=6 - 29

Very consistent!

On the Instructables website, an Excel document is given. The top part is simply plug in your values.

http://s26.postimg.org/3zyaze2s9/Drag_Coefficient_1.png

The bottom part is where things get fun! You are in charge of figuring out your Cd and Crr. The objective is to get the line as close to the dots on the graph as possible. The Sum of Error^2 will show how close you are to matching. The lower, the better.

http://s26.postimg.org/dyj9lvc7t/Drag_coefficient_2.png

According to my calculations, my new Cd is an amazing 0.269! while my Crr comes in at exactly 0.01. The Crr number is a typical value for a non-LRR car tire's rolling resistance. My coefficient of drag has dropped over six hundredths with only the few mods listed, and I have not yet anything to the rear of the car except for the rear belly pan.

To put it in perspective, a 2nd-gen Toyota Prius has a Cd of 0.26 and the 1st-gen Honda Insight has a Cd of 0.25 :thumbup:

Here are pictures of the modifications done to achieve the lower drag:

Wheels & Tires

http://s26.postimg.org/70is3dt3t/0522151048b.jpg

Wiper Cover

http://s26.postimg.org/ajxbjo0bd/0501151335.jpg

Rear Wheel Skirts

http://s26.postimg.org/h90lkiyl5/0421151616a.jpg

Rear Belly Pan

http://s26.postimg.org/ygw3sqc2h/0415151356.jpg

Front Belly Pan

http://s26.postimg.org/3rjfd6yzd/0415151306a.jpg

Grill Block

http://s26.postimg.org/hrlh5c0pl/0415151229.jpg

Foglight Covers

http://s26.postimg.org/gkd3b6yah/0125151800b.jpg

Even after all these years, I still haven't figured out how to use that Instructibles Excel sheet. I'll come back to this thread for questions :thumbup:

I'll be more than glad to help you out jedi-sol!

For a moment there, I thought you left the front wheels un-covered ( that first pic you posted )

Congratulations !

So what has been done about the rear of the car - mainly the edges of the boot/trunk ?
Stock, that car has a very rounded rear*
Did you get around to installing a trip strip or a small spoiler ?

Also, was the front air dam installed at the time of the test ?

( *Where else but EM can you talk about 'rounded rears' with a straight face ! )

Bahahaha!!! Only at Ecomodder could we talk about these things that would make old ladies faint if they walked in at the wrong time :D

Uhhhhhh..... That is because they were uncovered :eek: I did not like the small gap in between the cover and the wheel as the covers cannot lay flat against these wheels. So I left them off. In order to avoid any further ridicule, they will be going back on today :o

I have done nothing to the rear, sadly. I have not had the time since I have been trying to figure out front wheel skirts. I am wanting to install a Kammback and sharpen the edges of the bumper to promote cleaner separation. I am going to try a trip strip on the A-pillars soon and see what that is all about. Eddie25 posted something about one recently and said something happened, I can't remember.

And the air dam was not installed either! I tested the car with it looking like this:

http://s26.postimg.org/qy81llfvd/0526151412.jpg

The rear hatch was down BTW :D

did you have a friend ride along?

do you think a boat tail will do much,for you even with the low speeds you drive and the Aero already in place?

What is the difference between A and B? The direction of travel? When will we find out about the 'one weird trick' on run #7?

I'm with dirtydave, having different persons driving and timing would eliminate a source of bias.

I did a coastdown test on my Scion, and my method of data collection worked well, I think: I video recorded the SG II set to KPH. I paused the video at the instant the car reached the 70KPH, noted the time, moved 10s forward, and noted the speed.

There is no bias.

The challenge with this is to "know" the Crr in order to get the estimate for the Cd.

mods
 

Don't forget
*Passenger side mirror delete
*Coroplast wheel covers
:thumbup::D

My brother rode along and took time for me. He started the timer @ 55 mph and would say 8...9...10. At the 10 I would call out my speed and he would record it.

I definitely think a full boat tail will help. 0.269 Cd is still not considered low drag to me, even with my numbers approaching Prius territory. I want <.20 Cd and I am discussing with Aerohead about getting the ZX2 to 0.15.

A lot of my work has been quick and sloppy, honestly. I want to go back and refine all I can with better materials and with greater precision. My rear belly pan does not have the correct angle for the lowest drag and I need a middle section pan.

My biggest gains are going to come from the engine and transmission. The ZX2 is too deeply geared IMO to maximize the potential of a low Cd. The load on the motor will have to increase in order to reach a maximum BSFC, and I have three ideas regarding how to get there: Higher gears, lean burn w/ turbocharging and having an late intake opening (think Atkinson cycle).

You are correct freebeard, the difference in A and B is the direction of travel. I will comment on the 7th run in a new post. My brother did the data keeping and timing :thumbup:

No bias with my brother. He kept an eye on my speed and did the timing. I have no way of knowing the coefficient of drag of my tires. I have searched Google and nothing turned up :confused: The are Chinese made, regular car tires, so I knew the Crr would be between 0.01 & 0.015 from researching typical Crr's :thumbup: Mine came out on the low side.

Thanks for catching that aerohead! I did not have the coroplast wheel covers on during the testing though. I may have had a crack at breaking 0.26 if they were on :o

How long did the tests take ? I would think several hours.

I'm also curious about what people were thinking. Not that it matters, but can you imagine the comments ? :-)
" Oh look - he forgot something again. "
" What did he forget this time ? "
" He must have dropped something valuable out the window there - he keeps going back and forth ! "

Here is the time it took Cd - Time: 10:44 a.m. - 11:17 a.m. It was hidden in the 1st post. Shockingly short huh? I wasted no time!

Now if I add the time to compile the data into the spreadsheet and zero out the graph, then it took over 3 hours to get the finished results :thumbup:

Haha! Truthfully, I was wondering if anybody was watching and their thoughts. The chances were high for the police to get called on me! I drove past the same houses 7 times, and people start getting spooked when someone does that.

Suspicious scientist is suspicious :D

I have talked of the 7th run and not said anything about it. This run was to test both driver and passenger windows down halfway and their effect on the overall Cd. I run without A/C, so it is common for me to crack the window a little bit so I don't melt :)

However, what if I took it to an extreme for me? It irks me to have one window down halfway when I am up to speed. But am I just psyching myself out?

Let's see the data!

Run #7 Averages of the previous 6 runs
t=0 - 55 \ 55
t=1 - 48 \ 49.5
t=2 - 44 \ 44.5
t=3 - 39 \ 40
t=4 - 35 \ 36
t=5 - 31 \ 32.5
t=6 - 28 \ 29

All bias aside. Having both the windows down had an overall negative impact on coasting. However, the impact was extremely small! The speed of coasting with both windows up was equal to the coasting speed when both windows were down halfway, on occasion. The #7 run averages 1.25 mph at all data collection points during the test in comparison to the average of the previous 6 test.

What you can take away from this - as long as you are not a hardcore hypermiling stickler such as me, pushing limits and trying to squeeze every mist of fuel as efficiently as possible, then I doubt you will notice a difference in the real world having your windows down partially :thumbup:

Edit - There is no graph for the #7 test. I would need more runs to get an honest representation of the effect, and I only did one. Sorry y'all!

I'm not good at understanding math and such. I'm much more visual.
Even something as simple as your charts makes no sense to me.... but i wish i understood.
One thing i would like to know is if you started the test by adding in the factory quoted Cd figure.
If so, how do people know what variable to use when they don't know their cars original Cd ?
It would be interesting to see if a test would show the car as having the factory quoted Cd if you removed all your mods and retested.

Since you are thinking about a boattail, take a look at http://ecomodder.com/forum/showthrea...tml#post481536. He's shedding parts on the highway. BamZipPow has also run aground lately. I'm not a big fan of coroplast and duct tape.

Look for a build quality that would pass tech inspection at a race track. I just got a good part for an airdam on my car. It's the 'towel rack' bumper overrider from the American spec 67-down Beetle front bumper (cut down), and it will hang upside-down under the later bumper with tabs on the bumper brackets. I've got steel sheet for the diverter.

My goal is to get Prius-like mpg, starting with a 0.40 Cd. But first it's off to the Woodburn Drag Strip for the NW Bugrun on Saturday. Oops, that's today.

I'm sorry that you are having a difficult time understanding the data, Cd. Is there anything I can do to help? I know you are not the only one who is having difficulty, you're just the only vocal one. I don't have an Education degree for nothing :thumbup:

I input my original Cd figure of 0.33 into the input location and it was waaayyy off. So I started going down: 0.3, 0.29, 0.28 and so on. I would watch the Sum of Errors^2. This showed the total deviation of the pink line in comparison to my data points, squared. When this was minimized as far as I could get it (0.005 something!), then I knew I was pretty spot on :thumbup:

I can used my points that were plotted from my coast-down test and input the old Cd number back in to show the difference. The pink line would show the 0.33 Cd deceleration compared to the 0.269 Cd plot points of now.

I have seen the carnage of the boat tails recently. Pretty saddening to see their hard work get destroyed.

I don't exactly like coroplast and duct tape either. But for my first boat tail build, I plan to use cardboard and duct tape to keep cost down and its ease of working with. If the results are worth pursuing, then I will go back with aluminum, fiberglass and paint and make it perty ;)

Funny you bring up race tracks and drag strips. I have been wanting to get the ZX2 to the drag strip for quite some time! I think it will perform very well considering the Cd is much lower than stock. Good luck today if you are racing freebeard!

I know graphs do not help you much Cd, but it is all I can do for the time being :(

Here is a comparison of coast-down speeds with my current 0.269/0.01 combo versus 0.33/0.012. The numbers are a stock Zx2's Cd and a hypothetical increase in RR due to placecard air pressure.

The pink line is the hypothetical coast-down during a duration of testing a stock ZX2 in the same conditions as the original test. The dots are the data points for my coast-down test where I came up with Cd= 0.269, Crr= 0.01

http://s26.postimg.org/4pvc0coex/Ori...Present_Cd.png

Note the incredibly high Sum of Error^2. This shows the amount of deviation from my data points and it is quite a lot!

Worth noting that this test only works if your speed signal is accurate. Dashboard speedo's are typically 5% off, but the OBD data will usually match GPS, however, unlike GPS the OBD doesn't lag as much and is probably the best way to go :thumbup:

Good to see people gathering and posting their data.

I think you need to adjust your CRR, so that the dots at the end of the graph better match the model. You'll probably then have to adjust CD slightly.

Good point, OldT! The speedometer is very well calibrated. I compared the Speedo to the GPS and it is spot on:thumbup:

That last graph was a comparison between stock Cd and Crr vs. My coast down test. The line is stock, the dots are modified:thumbup:

While driving a few minutes ago, I thought of something that needed to be added to this thread:

When I deduced the numbers down and got the RRC for my Blacklion tires, they seemed pretty good since they had a value of 0.010. This is right on the line between LRR and normal tires.

Then I remembered - The tires were pumped up to 70 psi for the test!

So! That will throw the numbers off if tire pressure is lower than I tested. I typically run them around 60 psi, so that would mean the RRC is higher than my test. Maybe not now, since my tires have worn down some and "loosened up".

Hmm... makes me wonder what the RRC is if they are at placecard psi (35 psi).

Actually, I think I will test the RRC between placecard psi and hypermile psi eventually. It will be good to have numbers behind what we already know is a benefit.

All in all, the test concludes that I need LRRs... asap!