terminal velocity coast testing
 

I tried some of that when doing my first cardboard stuff on the beetle.
I had two hills, both of which i rolled on at a specific speed so as to avoid conflicts with traffic.
One yielded 59 MPH or so before it flattened, the other 61 MPH.
I think it needs to be slower to really see a CD change, more like high 40's or 50 tops.
What do you think?
My reasoning is, the drag is getting so high up in those speeds that a subtle change will be hard to quantify.

think
 

The only official tests done in the past are top speed testing under full power on a level track.
At top speed,we'd have a chance to discern a minor drag change,as there are formulas to do that.So my gut feeling is that 'faster',not 'slower' testing would be the order of the day.
Rolling resistance and powertrain losses would also have to be a constant during the test,so the test car would have be fully pre-warmed before starting.

Cd is pretty consistent between a lower bound, depending on temperature and barometric pressure, probably somewhere around 20-25mph and transonic speeds—above 250mph. Shockwaves start to form, leading to those sharp pointy noses that are totally unnecessary on a road-going car. Fineness ratio too, land speed record cars have the proportions of a chop-stick.

That simplifies things, I can keep using my two hills. I picked up some 3/16" steel wire to wire form the trailing edge and then add plastic to that.
I have a 400 mile week of bad commuting so I hope to get something together before Sunday evening.

Rather than trying to connect the ends of 3/16" wire, make hoops and tape the edges together. Sort of like making a geodesic dome out of hula hoops.

Unless it's welding rod and you have a welder standing by, of course.

Skyking, I really looking forward to seeing your results.
I suspect quite a bit of fluctuation in the data though.
You have so many variables.
I would guess that you are engaging your cruise control at a certain target well ahead of the test area ?
How accurately will the cruise control keep the car at a precise speed ?
Even a slight dip in the road, or a small twig that you miss can skew the data.
Variables in wind direction and velocity will have a large effect too, of course.
And as the others have mentioned, tire friction / temperature / driveline losses as well.

How many runs do you plan to do ? As you of course know, the more data the better.

I use the digital readout from my scangauge and use my foot to set up the coast. I have a couple of markers, one for me to get my speed on and the other for the kick of the stick.
Wind is not too much of a problem down in the trees here. If it is out of the south or north it can get down to the road level, otherwise it is pretty still.
I got 61 several times on the one hill and only 1 MPH variation on the other so I am happy with the method.
My big concern is catching a dry day. Today if I get everything put on I won't be able to test due to expected rains.
I have a coast down run that was consistent, although not level. I enter that one at 50 and use 40 and 30 for the data points. It happens to be on the way to work so I will get that one done this week.

variation
 

With the hill I coast on I've seen over a 38-mph variation in velocity,with zero modifications to the truck.It's been extremely problematic attempting to learn anything from doing it.I'm not saying don't do it,just bear in ind that it can be a real can of worms,data :pwise.