Practical ways to measure lift, downforce and drag (?) - Rennlist - Porsche Discussion Forums


Practical ways to measure lift, downforce and drag (?)


Old 01-20-2017, 09:09 PM
Thread Starter
Join Date: Jul 2012
Posts: 768
Default Practical ways to measure lift, downforce and drag (?)

There are a lot of things you can do to your car that will affect the aerodynamics, but there is rarely any information about how. And it matters. A small change in front end lift can have a big, noticeable impact on stability (for instance, the difference between a 996.2 C2 and GT3 is only about 12kg at 200 kmph). And reducing front lift without a similar change at the rear Ė which Porsche has always been very careful not to do Ė is often said to make the car quite unstable at high speeds.

Iíve read many many threads on bumpers, lip spoilers, ducts, wheel well vents, ducktails, wings, and diffusers, and with very few exceptions, there is zero data. Iíve called top German tuners and other highly regarded manufacturers of aero add ons, and again, zero data. (Manthey wings are a notable exception.) If you want to know the impact of a change, you are going to have to figure it out yourself. Or perhaps, other members and I will use one of the below methods and find out for you (or at least share our findings online).

The good news is that you can measure changes in lift and drag without a wind tunnel. The bad news is that I still canít tell you exactly how. Indeed, I started this thread to share what I know, and ask members who know some of the missing bits to help out.

Iíll save drag for another post, because thatís the easy part. Iíll also save ďwhy not just go out on track and check the lap times?Ē. As for lift and downforce, if you Google, youíll find a number of ways, but Iíll focus on three that strike me as most promising for our cars. And if anyone knows of any others, please let us know!

You get sensors called linear potentiometers/shock pots, install them to your suspension, and attach them to a very high quality data logger (if thatís the right word), e.g. from Pi or Stack. The gear will cost something like $6k. And to give credit where itís due, I learned this from another Rennlister: stvsxm.

I wonít be making that sort of investment into this. One might however see if one can find a racing team or race prep shop who can lend you the equipment and/or do this for you. Alternately, the sensors are the least expensive part, so although I understand they typically canít be coupled to an amature data logger, Iím wondering if they might nonetheless work with the sort of high end data amature data loggers I can probably borrow from friends, e.g. one of Race Logicís devices that can acccept numerous external sensors.

The second idea is to somehow tap into the data coming from the ride height sensors. (996s with Xenon headlights have ride height sensors. Presumably 986s and all later Porsches with Xenons do to. Donít know about earlier cars.) And at least on some cars, it is possible to tap into such data. Hereís a link where someone did it with a Lexus:

There is a simple way to translate changes in ride height into kg of lift or downforce. Put some weight between the wheels on one axle (e.g. in the frunk or behind the rear seats on a 911) and measure how much the height changes. This tells you how many kg are needed to move the height by X mm, so if you know the car rises Y mm at speed, you can calculate how much lift you are generating. Note though that for the car to sink enough to measure a meaningul change, youíll need a lot a weight, especially if you have stiffer than stock springs. 10 kg surely wonít do it unless you have a very precise measuring device and method.

(Or if you already know your effective spring rates front and rear, then you could calculate the lift or downforce without even measuring the impact of adding weight over one axle.)

Unfortunately, this is where my knowledge of how to do this starts to run out. How to get the data? I may call around to people who do chip tuning. How to save or display the data? On a PC? Or perhaps one could again use a data logger, good ones that accept external sensors. I may call data logger manufacturers like Race Logic.

Iím hoping someone else on the forum will have more of a clue how to do this, and Iím quite certain many (most/all?) will know more about our cars electronics and how to plug into them than I do. So any thoughts/advice/insight about this would be greatly appreciated!

This is probably simplest and cheapest, but also the least useful. Indeed, it probably wonít work, because itís just my idea, not tried and tested by someone else so far as I know, but anyway, here it is: The idea is to use a spririt level or bubble level Ė but a precise digital one, not one where you look at a bubble, so that it tells you precise angles in numerical terms.

First, you mount this in the car. Maybe you use a GoPro mount or a radar detector mount. Or maybe you tape it onto the center console or the handbrake with lots of masking tape to immobilize it. Though Iíd be surprised if wiser readers donít have better mounting ideas than mine. Then you find the angle when the car is standing, then at low speed, and then at several different, high speeds. That tells you how much more the nose is rising than the tail at each speed (and if your car is a 911, the nose will almost certainly rise more than the tail). Then you do your aero mod to one end, and test again, and see how the angles change.

If you changed the aero at the front, then an change in angle is probably (almost) all due to a chance in height at the front. Do a bit of trigonometry (or add weight at one axle as above, but measuring both the change in height and the change in angle) and you can calculate the change in height at the front, and the change in lift.

But if you changed the aero at the rear, then if the angle changes in a way that suggests that the rear is lower, itís fairly likely that it indeed is lower, but also that the front is higher. How much? Apparently increasing the rear downforce by 25 kg on a 996 GT3RS versus the GT3 caused front lift to increase by 9 kg at 200 kmph. And thatís a car with pretty stiff springs. On a normal Carrera, the difference would be much greater.

Still, this method can help with both front and rear. If the car has roughly the same angle as it did stock, then you are not far from having a balanced setup. And if someone posts the angle of their GT3RS (or Turbo) at speed X, and your Carrera has the same angle, then you probably have an even better setup for track (or road).

With all these methods, you probably want to:
  1. Have roughly the same amount of gas when you measure. Iíd aim for half a tank, since thatís probably the average amount.
  2. Make sure you start out with stock ride height if you want to see how the car behaves stock, or at least the same relative height front to rear.
  3. Make sure your test road is level, which you can do by driving too slow for aerodynamics to make any meaningful difference (Iíd think 60 mph/100 kmph is quite safe unless you have a monster rear wing) and verifying that you have the same heights and/or angles as when standing.
  4. Drive fast enough. At least 90 mph/144 kmph, I would think. Ideally 124/200+.
  5. Donít accelerate, brake, or turn while measuring.
  6. Use a GPS device to know your precise speed.

That's what I know so far. Additions and corrections are welcome!!! (Questions too - I hope someone can answer them.)
rs10 is offline  
Reply With Quote
Old 01-20-2017, 11:24 PM
Rennlist Member
Yargk's Avatar
Join Date: Sep 2004
Location: SF Bay Area, CA
Posts: 1,597

I really want to do this on my GT4. I can accurately ballpark the downforce from added a gurney flap with a theoretical guess based on the known stock downforce, but dive planes are not simple so I'd want to get empirical data. Here's what I've found on the subject so far.

cheap options:

"Some better data on the horizon. I just got these ride height sensors from a mid-90's Lincoln Continental's air ride system. $4 each on Ebay."

expensive option:
Yargk is offline  
Reply With Quote
Old 01-20-2017, 11:33 PM
Rennlist Member
Join Date: Sep 2006
Location: Livermore, CA
Posts: 17,147

Couple of things.

At the office and don't have time to read through and digest it all but for Germany the word I get is if a manufacturer offers some kind of aerodynamic aid/add on if there is any real claim of an aerodynamic benefit it has to be backed up with real test data. The TuV requires this.

That you only found one German company, Manthey, with any data suggests only its products offer any real aerodynamic improvement.

A crude but effective test for any benefit is a coast down test to determine the amount of drag and a track test to measure overall lap speed, top speed at various places on the track and corner speeds to determine if the aerodynamic aids/add ons are actually doing anything more than adding weight and increasing drag.

(Years ago a USA car mag tested a car -- I won't name the brand/model -- and found the impressive rear spoiler resulted in a shorter coast down when present and a longer coast down when absent and track lap times and straightaway speed and corner speeds were close enough to the same that the spoiler was deemed ineffective at best and a drag (no pun) at worst.)
Macster is offline  
Reply With Quote
Old 01-21-2017, 03:03 AM
ion_berkley's Avatar
Join Date: Aug 2015
Location: SF Bay Area
Posts: 69

If I was going to play with this then I'd go straight to thinking about using load cells so I could measure the forces directly.
Then the way I see it you have 2 choices:
You fabricate some way to measure the force (weight) applied at the interface of the sprung mass of the car with the unsprung mass (in otherwords where the springs connect to the body).
Or you fabricate aerodynamic features that you are interested in experimenting with and attach them to the vehicle with load cells in such a way that you can measure the forces they then contribute. It would all be very Heath Robinson but probably quite entertaining provided it was an old beater of a car and not your shiny toy you were messing around with.

If I was actually serious about doing this and had real money then probably my first port of call would be again loadcells in the form of a decent 4 corner balance setup placed in a wind tunnel where the environment was highly quantified and stable and I wouldn't have to worry about filtering the aero data from all the vibration and noise that would be polluting the data if the car were actually rolling.

My 2 cents

p.s. Budget load cells:
ion_berkley is offline  
Reply With Quote
Old 01-21-2017, 01:44 PM
Rennlist Member
ace37's Avatar
Join Date: Jun 2015
Location: SLC, Utah
Posts: 1,202

Besides what's already been mentioned race engineers will use pressure sensitive paint (pressure*area=force pressing perpendicular to that surface) and strain gauges (cheapish but hard to install well and require some work to interpret). Neither are really great for amateur racers or total downforce/drag but they work well for answering specific questions.

Also for a free qualitative estimate you want to look at what the rain does, where the rain water lines go, and see how the airflow changes by watching the droplets flow around the car before and after an aerodynamic change. This is similar to using smoke in a wind tunnel.

Sadly I'm not sure there is a better overall answer than paying for wind tunnel test time. It's not cheap directly but it's the gold standard and you get answers without any other heavy technical work interpreting voltages into forces or compromises like installing PITA hardware everywhere.
ace37 is offline  
Reply With Quote
Old 01-21-2017, 09:42 PM
Thread Starter
Join Date: Jul 2012
Posts: 768

johnb55 posted some helpful info in the 997 forum:

we use an AIM data acquisition system with them with shock potentiometers ... An AIM system is phenomenal and not very expensive

For more info, here's the link:

So this is a version of my method one. Hopefully "not very expensive" means less expensive than I feared. But I'm still hoping to find a way to use my method 2 (using the data from the cars own ride height sensors).
rs10 is offline  
Reply With Quote
Old 01-21-2017, 09:50 PM
Thread Starter
Join Date: Jul 2012
Posts: 768

Originally Posted by ace37 View Post
Besides what's already been mentioned race engineers will use pressure sensitive paint ...

Also for a free qualitative estimate you want to look at what the rain does ...

Sadly I'm not sure there is a better overall answer than paying for wind tunnel test time. It's not cheap directly but it's the gold standard and you get answers without any other heavy technical work interpreting voltages into forces or compromises like installing PITA hardware everywhere.
Thanks for several interesting ideas! Rain makes sense, and I didn't know about the paint. If I add a new spoiler/bumper/wing, etc., that could be interesting to try before they get their final paint job. But I won't be painting over any original body pannels.

Regarding wind tunnels, obviously if you have access to one you want to use it. But I'm not sure everything else is worthless. Indeed, my understanding it that while with wind tunnels you get the most precision, but on the road you get the most accuracy, because even the best wind tunnels are only an approximation of the more complex real world. Anyway, if I had access to a wind tunnel, I wouldn't have started this thread ... .
rs10 is offline  
Reply With Quote
Old 01-21-2017, 10:10 PM
Thread Starter
Join Date: Jul 2012
Posts: 768

There's an interesting post in the 991GT3 forum by [email protected] (yes, that's his user name):

"We're developing a data logger that when finished would allow you to measure the ride height sensors (shock potentiometers) to get a way to quantify aero changes on the car (lift and downforce)."

Here's the link:

Sounds like what I'm looking for (assuming it uses the stock sensors, and not just for the 991, and that the data logger is a LOT less expensive than the costs I expected for method one.
rs10 is offline  
Reply With Quote
Old 01-23-2017, 08:20 PM
Thread Starter
Join Date: Jul 2012
Posts: 768

Despite my best efforts to keep the discussion in one place, instead of having 10 different discussions for 10 different Pcars ...

Well, the good news is that there is another interesting discussion about this, in the 996TT forum.

Here's the link:
rs10 is offline  
Reply With Quote
Old 03-11-2017, 12:57 PM
pyropete125's Avatar
Join Date: Sep 2012
Location: southern CT
Posts: 171

I remember reading an article about when they tested the original ac cobra and they glued on pieces of 3" string all over the car to show where the angle of wind was going and they thinned out white out and dripped it on the car to show where the direction was. I think it was an interview with peter brock.

Not really going to show the data you want to measure but it was neat to read about.
pyropete125 is offline  
Reply With Quote
Old 03-14-2017, 10:22 PM
mark_schnell's Avatar
Join Date: Feb 2017
Location: Rabbit Valley, CA
Posts: 224

I just contacted these folks to see what they would charge to run the car through a rolling-road wind tunnel so no math is necessary...

I'll keep you posted.
mark_schnell is offline  
Reply With Quote

Related Topics
Thread Starter
Last Post
996 Forum
03-19-2017 02:26 AM
997 Forum
01-17-2017 10:08 PM
Data Acquisition and Analysis for Racing and DE
10-11-2016 05:30 PM
991 GT3 GT3RS and 911R
08-29-2016 04:09 AM
02-07-2016 01:57 AM

Thread Tools
Search this Thread
Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Forum Jump

All times are GMT -3. The time now is 08:49 PM.

Copyright © 1998 - 2017 We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to and affiliated sites.
  • Ask a Question
    Get answers from community experts
Question Title:
Your question will be posted in: