X-EMS now has it's own dedicated website! Head over to xems.eu to keep up to date!

It's always a good day when the smoke comes out the correct end of the car, especially when testing new and untried parts! However, sadly hardware testing with the X-EMS proto1 board is coming to a close.

Why? Well because it has completed all the tests that we had planned for it. Yeah! And completed them well!

It reads all the needed engine sensors, including crank and cam position sensors, engine intake air temp and coolant temp sensors, throttle position sensor, battery voltage, and wide-band lambda. It also outputs data via the native USB plug to a laptop, calculates fuel maps, does timing, and all the other things needed to make a engine run.


The X-EMS proto1 board was made just to test out the hardware, and to do some basic code and timing testing on a actual running engine. It successfully started up our test race engine. And also successfully manages to idle the engine, and even provides a quite stable idle!

This is a full engine control box, it runs the ignition coils and injectors directly from the board. It's even capable of running the cooling fans directly off the board!

But enough with the text, here are our "first start", and "first idle" videos to show off our stuff!

First start

First idle

The X-EMS project took another jump today!

We have completed the code for testing the hardware, and just got the benchmarking setup for the proto1 X-EMS board back from 5 Watts.


In the coming days we will be doing some pure electronics benchmarking, to test out the code performance and get some confidence in the code and hardware.

The plan is to first debug the hardware testing firmware. Once we are satisfied with the firmware and hardware performance we'll hook it up to the er7, and start testing on a real engine.

The code is in no way made to work as a "real" EMS, but more to test out the hardware on a real engine. It should be capable of running the engine with a minimum amount of inputs and outputs though. Just the stuff that is needed for bare-bones testing.


Once we get it running a real engine, we will sit down and take a deeper look at the economics of the project. We will then hopefully have some idea of the time needed to develop the firmware and hardware to a consumer product. We'll see how we proceed from there!

The target is, and has always been, to make the most versatile and expandable EMS on the market. Because that's what we want for our own projects also!

Our X-EMS project is moving along swiftly!

We just received a photo of the completed proto1 X-EMS board from 5 Watts, looks super-sweet! Also discussed the parameters for our test setup.


The real product will of course be much more versatile, this bare bones proto1 is just being made to test out everything on a tight time table.
Some parts have been deliberately over dimensioned, and a lot of protection circuits have been left out to speed up testing. The 30A drivers for instance don't need to be capable of that much current, but they will run cooler, so less attention will be needed on cooling for the proto1.

Something might go up in smoke, but hopefully not. The circuitry is relatively simple, so hopefully the bug count will be low as well. Even on this first proto.

Hardware specs (X-EMS proto1):

  • 2 hall sensor inputs (crank and cam position sensors)
  • 2 temp sensor inputs (coolant and intake temp)
  • 1 tps sensor input
  • 1 wideband lambda input
  • 8 power outputs (30A)
    • 4 injectors
    • 2 ignition
    • 1 cooling fan power
    • 1 spare


Firmware mods are coming along nicely also. Timing stuff and some code conversion left to go/check. But the testing will really get underway once we get it hooked up to our test car.

And we will have the er7 ready shortly also, to test the X-EMS proto1 on. Projects that move along quickly are really inspiring!


We had some discussions with Five Watts Ltd about how we're going to modify the timers from one of the building blocks that we're basing X-EMS on.

We're going to go with having the ECU time and date aware. That will simplify datalogging, and should also provide a good basis for future expansion, and new uses. Most likely going with UTC. Although there are some other good time bases also. There are ready and tested C++ code libraries for all of this in one of the building blocks we're using!


Five Watts should be getting the first proto boards shortly. In the mean time we've been busy preparing the er7 for X-EMS testing, and getting other testing done and out of the way.

We're also going to be showing the X-EMS at our stand D 335 at Alihankintamessut 2013 on 24.-26.9.2013

Page 1 of 2