OSHW Turbo Diesel Tuning Box

[luky315]

I am developing an Arduino-controlled Turbo Diesel Tuning Box (MAP and Common Rail sensor signal manipulation).
It works on my car, but it needs some testing and software work before releasing the design to the public.
Are there other car enthusiasts willing and able to help?

[dr.diesel]

I'd be happy to help, but all of my diesel's are old school rotary!

If you don't get any hits here, try the guys on www.dieselplace.com, numerous EE level guys there.

[Wilksey]

If you want people to test it you are going to have to post the project or link to it, as I know I wouldn't be happy to test anything like that until I saw how it specifically worked.

[dr.diesel]

I think that's a given, he just wants a peer review/test or two before posting publicly.

[Wilksey]

Ok,

Well, at a minimum we need to know:
1. What he has currently done and what interfaces he is using.
2. What he has currently tested with and what the results were, in particular what the reading of the "adjusted" sensors are.
3. What he wants help with.

I.e. what software work needs to be done.

[luky315]

I have working hardware on a stripboard and am currently designing the PCB. The circuit is able to manipulate 2 "standard" 0 to 5V (or more realistically 0.5 to 4.5V) sensor signals. Before publishing it is necessary to "clean up" the Arduino programm (with librarys) and writing a PC program for generating and downloading the new engine operating map to the tuning box.
And I need a good name for the Project :-)

[Wilksey]

Are all engine sensors between 0 - 5V? Or do some run from 0 - 12V?

A new map to the tuning box? Are you intercepting ALL of the signals going to the ECU? The MAP / LUT is stored in the ECU, which I assume you are not attempting to change?

I guess you are doing something similar to the old resistor trick on the MAP sensor on petrol engines?

[luky315]

I've only encountered 5V Sensors. If you have seen something else please let me know. By the way: There is no "12V" in a car.
I don't want to manipulate the original ECU so I manipulate the sensor signals (dynamically, the old trick with the simple resistor / resistor divider is not working on newer engines...).
The idea is to persuade the ECU to inject more fuel under certain conditions.

[rob77]

unless you publish something.... it doesnt' exist !
how the hell are you going to manipulate the "sensors" on a common rail injection ? apart from the sheer fact that there is literally nothing "to manipulate" on a modern common rail injection, the modern common rail injection systems are kind of "fragile" - a failure of the high pressure pump will destroy everything (been there - Bosch CP4.1 injetor pump failed and flooded the whole system with aluminum fillings destroying all the injectors - a FREAKING EXPENSIVE repair, and i was just using the car without any kind of mods) so it's not a good idea to manipulate anything from outside of the ECU.... THE ONE AND ONLY WAY how to tune a common rail engine is to replace the maps in the ECU, definitely not by manipulating the sensors.
but anyways... there is a rail pressure sensor, and airflow sensor available for manipulation - the others are not easily accessible for a "man in the middle" manipulation, so i'm really really curious what your "solution" is

[luky315]

The principle is simple. The ECU uses the rail pressure sensor in a closed-loop control loop to regulate the pressure in the rail according to the value stored in the engine map. If the sensor signal is attenuated (not by much and not always!!) the closed loop control thinks that the desired pressure in the fuel rail is not reached yet and increases the pressure.

There are a lot of tuning boxes out there which work this way, but no OSHW solution...

[rob77]

the closed loop control thinks that the desired pressure in the fuel rail is not reached yet and increases the pressure.

and exactly that excess pressure will kill the high pressure pump (higher temperatures long term - more material fatique and finally kaboom).
futhermore - if you increase the pressure and amount of injected fuell - you need to adjust the airflow as well - otherwise a lot of soot will be created thanks to incomplete combustion and that will kill your DPF and catalytic converter....

adjusting the control loops and processes inside a modern common rail engine is kind of a "black magic" ... any kind of "tuning box" is a quick way to destroy the engine... either tune the maps in the ECU (end even that way you can easily damage the engine) or simply accept the stock setup.

if you want a more powerful car , then buy a more powerful one instead of tuning a weak one - it will be MUCH CHEAPER at the end of the day

[luky315]

There is basically no difference between manipulating the engine map inside the ECU or "outside". The outcome is the same: more fuel in the cylinder. And as long as we stay in a reasonable range (10-25%), there is no problem with the air mixture. A diesel engine works with excess air (not like a petrol engine with Lambda = 1). And we have another mechanism working for us: More burned fuel means more exhaust which is driving the the turbocharger. And we can also manipulate the MAP-Sensor (Manifold-Absolute-Pressure).
A lot of commercial tuning Boxes work this way.
As I wrote earlier: You DONT raise the pressure all the time. Just when a "power boost" is needed.
Of course you increase the mechanical stress on all components when using the extra power. If you don't want one: fine. Your choice.
I am still looking for car enthusiasts willing and able to help with the software and test subjects.

[wraper]

A lot of commercial tuning Boxes work this way.

You mean, commercial scam boxes?
You will unlikely to get any positive effect and moreover ECU will get a fault code completely crapping performance. Resetting fault code will be only possible with resetting it over CAN bus.

[dr.diesel]

Guys, luky is absolutely correct.  Fooling of sensors is done ALL the time.  Done correctly and in the right portions, while fooling others to compensate is very effective and works.

[rob77]

There is basically no difference between manipulating the engine map inside the ECU or "outside". The outcome is the same: more fuel in the cylinder. And as long as we stay in a reasonable range (10-25%), there is no problem with the air mixture. A diesel engine works with excess air (not like a petrol engine with Lambda = 1). And we have another mechanism working for us: More burned fuel means more exhaust which is driving the the turbocharger. And we can also manipulate the MAP-Sensor (Manifold-Absolute-Pressure).
A lot of commercial tuning Boxes work this way.
As I wrote earlier: You DONT raise the pressure all the time. Just when a "power boost" is needed.
Of course you increase the mechanical stress on all components when using the extra power. If you don't want one: fine. Your choice.
I am still looking for car enthusiasts willing and able to help with the software and test subjects.

fully agree , your statements are  100% valid for diesels with electronically controlled rotary injection pumps and pressure driven injectors (single pilot injection and the main injection) ... those engines got few variables - air flow, boost , amount of fuel to be injected... very simple, elegant and easy to tune engines...... but please stop convincing anyone that it's the same for modern common rail engines ... just have a look at the 3rd generation of VW diesels, apart from the shitload of sensors and way too complicated engine.... those even got an intake manifold flap - something seen on petrol engines only before.

[luky315]

"Modern" engines are not Black Magic (nor is high speed digital ;-) ) if you understand how they work.
If you are afraight - don't touch them and look for something else to do.
My car has a Kia / Hyundai CRDi engine and its working fine...

[rob77]

"Modern" engines are not Black Magic (nor is high speed digital ;-) ) if you understand how they work.
If you are afraight - don't touch them and look for something else to do.
My car has a Kia / Hyundai CRDi engine and its working fine...

and what sensors/signals are you manipulating ? since when is arduino high speed digital ? how did you solve the automotive requirements for the electronics ? if you're manipulating just a subset of the signals... isn't the engine going to default/safety mode ? how did you measure the actual results ?

so back to my very first comment.... show something or it doesn't exist   proove me that i'm wrong .. becaue i think you got nothing usable yet and you just talking bullshit here

i'll apologize for my previous statement once we'll see something

[dr.diesel]

i'll apologize for my previous statement once we'll see something

 

Better just move on luky..  I suggest the project name R.O.B.

[rob77]

i'll apologize for my previous statement once we'll see something

 

Better just move on luky..  I suggest the project name R.O.B.

i'll rather see people "facepalming" at my comments than people "facepalming" after the "arduino power box" destroyed their engine. and that's all just because of someone got the brilliant  idea of "arduino manipulating" the sensors of a common rail diesiel engine

so common luky... we're all corious

[luky315]

Nothing I can show will convince rob77 and people like him that the thing is real. Or would a schematic really convince you? Don't feed the Troll...

Is someone here who is able to understand the function of a common rail diesel engine and likes writing software? At the moment the necessary parameters in the arduino firmware are "hard-coded". It works, but only on my engine and certanly not optimal. A PC interface to show the signals and change the parameters while driving would be the next step.

[wraper]

Nothing I can show will convince rob77 and people like him that the thing is real. Or would a schematic really convince you? Don't feed the Troll...

Is someone here who is able to understand the function of a common rail diesel engine and likes writing software? At the moment the necessary parameters in the arduino firmware are "hard-coded". It works, but only on my engine and certanly not optimal. A PC interface to show the signals and change the parameters while driving would be the next step.

And how do you check that it is really working and not a placebo effect. Like audiofools are hearing how power cable changes the sound 

[TheEnd]

It's a simple idea, the rail pressure sensor will output something like 0.5-4.5v with pressure from 0 to about 1800bar, and the MAP sensor will do similar for the Boost pressure between 0 -3 bar (absolute)

If you decrease these sensor voltages, the ECU will adapt and increase the boost and rail pressure.
Almost all diesel tuning boxes just do this, and skew the voltage.
It's not going to be hard to take some more sensor readings and make the amount of "skew" change at different pedal positions or RPMs of the engine.

You might be able to tap into the pedal potentiometers to gain a pedal angle signal, RPM might be harder, but there may be an analogue output, otherwise you'll be able to read these from the CAN bus, or with an OBD2 reader.

[luky315]

You don't need an RPM signal or the gas pedal position. It is possible to get all the necessary information from the slew rate of the "original" sensor signal. If power is needed / demanded by the driver, the sensor voltage goes up as the ECU increases the Rail pressure. If the box detects this, it decreases the sensor voltage seen by the ECU, which leads to more pressure in the rail.
The schreenshot shows an old version of the circuit. CH1 is the original signal and CH2 is the manipulated signal. The new firmware is able to smoothly switch between "Original" and "Boost" state.

[m100]

I've no diesel modifying experience but having modified ECU maps on petrol turbo vehicles before, and IMHO in a manner both more effective and safer than that of some of the kludges of so called 'professionals'  I tend towards the view you need to get access to a rolling road or in car data logging and a means of measuring the composition of the exhaust gases to ensure you are getting real measurable effects,  that they are controllable, and you are not causing issues with other ECU functionality such as DPF regen for instance.

(P.S. Modifying the output of MAP sensors with zener clamps was a crude technique I came across in the distant past)

[Scrts]

Oh yes, we do hundreds of tests for months in automotive, calculate everything precisely, do a bunch of reviews, prepare tons of documents and then someone in the garage outsmarts all the engineers and does a better job on ECU stuff. Nice, I am so happy that we have so many talents!