How Much Power Does the Volkswagen TDI Lose in “Cheater” Mode?

2011 VW Jetta TDI power loss dyno emissions test

2011 VW Jetta TDI

One of the unknowns in the Volkswagen diesel emissions cheating scandal “Dieselgate” is the potential power loss when an affected VW TDI car go into the “cheater” mode. Does it lose any power when being tested on the dyno? When Volkswagen releases a retrofit or fix for the emissions problem, will the corrected cars lose power in daily driving? If so, what power loss are we talking about? TFLcar wanted to find out, and this was the result.

We talked to the local Boulder Diesel Emissions testing station to verify the testing procedures. The Denver area counties that test for diesel emissions, only test for particulates. The Nitrous Oxides (NOx) emissions are not measured at all. This is strange, because it is the main point of contention in this scandal. EPA states that affected TDI cars emit 10 to 40 times more NOx in daily driving than the “cheater” mode during dyno testing.

The acceptable particulate level in Colorado emissions testing is 35% opacity. That means the smoke and particulates that are coming out of the exhaust pipe can block up to 35% of light and pass the test. This is a lot of smoke. Most TDIs we observed emit far less smoke. The red 2010 Jetta TDI emitted a maximum of 0.9% opacity, or 35 times better than the passing level.

Next is the actual dynamometer testing to simulate regular on road driving on an all-wheel-drive dyno, versus front wheel only emissions testing.

vw tdi power loss graph dyno emissions dieselgate

Solid Red: Test #1 WHP, Solid Blue: Test #1 Torque

The silver 2011 Jetta TDI put down 138.5 horsepower and 260.0 lb-ft of torque on the dyno with all four wheels spinning. This is the simulation of regular on road driving. We saw no warning lights or other indications from the test car with all four wheels spinning. (Two consecutive runs were performed with four wheels spinning. They were nearly identical.)

The same car registered 136.5 horsepower and 228.4 lb-ft of torque when the rear wheel were left stationary. The same dyno was used with the same dyno technician. And the tests were performed within several minutes of each other. (Two consecutive runs were performed with front wheels spinning only. They were nearly identical.)

The math shows a total loss of two horsepower at the peak, but if you look lower on the RPM curve the difference is significant. The car lost 15 horsepower around 2,800 RPM, and 32 lb-ft of torque near 2,700 rpm. In fact, there are power disparities between the tests going up to 3,800 rpm.

The bottom line is, we observed a power loss of approximately 10.5% when simulating the emissions testing procedure.

Get all the details in this video.

2011, diesel, dieselgate, jetta, Volkswagen, vw tdi

Original URL:

Original article

Using htop to generate a live website background

We were looking for a nice background image for our new blog. So we stumbled over several candidates until Fabian Trampusch came up with the idea of showing a live htop summary.

Everyone was happy with the idea, so we began to take screenshots and tried to automate the process. I mean “what could be so difficult at taking automated screenshots?”

…until we realized, we are in an ssh session on the remote server. A screenshot without the X window System is not as trivial as it sounds.

The Imagemagick attempt

After some bit of googling we found our first attempt:

Use ImageMagick to print the output of a process to an image.

ls -lah | convert -resize 500 label:@- image.jpg


The output was not very appealing so we went on to a more sophisticated solution with postscript:

  • Using a2ps to convert… anything (?!) to postscript,
  • … gs to convert it into a png-file
  • … and trim it with ImageMagick.
ls -la | a2ps -=book -B -q --medium=A4dj --borders=no -o 
gs -sDEVICE=png256 -dTextAlphaBits=4 -q -r300x300 -sOutputFile=out2.png <
convert -trim out2.png result.png

The result is better than the first solution, but also unsatisfying. Also a problem is that a2ps needs piped data from your process. Htop is a process which doesn’t terminate by itself and produces no direct pipable output.


The Snapscreenshot attempt

After some more research we found out about the tool snapscreenshot, which is able to capture the output of a TTY.

After download, compilation and installation, we were able to use it to capture the output of /dev/tty1… which is somehow unspectacular.


The next step is to start htop on another TTY:

read _ _ sid < &1 strace -esetsid setsid  sh -c "exec $PROGRAM  /dev/tty2 >&0 2>&1 ")

Linux… are you drunk?

Due to the fact that we need the process id of the new htop instance, this command is somehow a bit complex. For writing a proper daemon, you have to store the PID to terminate the process later. So This commad starts htop on tty2, and stores the pid into $sid.

The difficult part of the project is done. We’ve got a solution to start htop on tty2, can capture it and it is also daemonizable (what a word).

The next steps are trivial. Write an init script, modify the webpage to load a certain Image as background and go for it.

Again thanks to Fabi for building a quick javasript function to periodically reload the beackground.

Now to the initial question…

Is it intelligent to show everybody what your server does?

Definitely not!
… But we thought everyone with a bit of knowledge about webservers and WordPress knows the disclosed information anyways. So the only new parts are that we have one gig ram and two cores.

Original URL:

Original article

Proudly powered by WordPress | Theme: Baskerville 2 by Anders Noren.

Up ↑

%d bloggers like this: