DIYEFI.org Forum

After years of watching FreeEMS progress I think the time has come to jump in.

The car I want to FreeEMS is my Honda Civic with 1.6 SOHC VTEC engine.
It uses the Honda OBD-I ecu which is already completely hacked and remappable. See pgmfi.org for more info
So FreeEMS is not a real market for these cars so it will most likely just a testbed for the FreeEMS project.

The hardware I am planning is a PnP ECU to replace the stock Honda ECU. I have a spare ECU which I am going to disassemble to get its connector. I am planning a dual layer 160mm * 100mm board with all the interfaces for the stock ECU.
The stuff I am going to use (actuators and sensors):
4 injectors
distributor output
Idle air control
VTEC solenoid
Fuel pump
Radiator fan
Check engine light

2 VR inputs (stock dizzy has 3 inputs for batch fire, I think full sequential without batch fire will do so only 2 VR inputs will be used)
MAP
TPS
CHT / engine coolant temperature
Battery voltage
Lambda
Vehicle speed sensor

Stuff that I have to check for the engine / car:
Alternator control
Lambda heater control
Starter switch

I am also thinking of adding a knock sensor and logic spark outputs for COPs.

As a starting point I will use the schematics of Jaguar A3 and RavAGE.
I have quite some experience in designing hardware so I do not need very much help on the hardware part.
As soon as I will start committing designs I obviously like to hear your comments!

I will make a GitHub repo with the KiCAD files soon!

DaWaN wrote:Battery voltage
Battery reference voltage

What's the difference? :-)

DaWaN wrote:Vehicle speed sensor

No support planned for this right now, will likely require some sort of external IC counter to support until later fw revisions. And even then, something like 4 wheels + driveshaft will require something external.

DaWaN wrote:Alternator control

What's required? Pretty simple or not?

DaWaN wrote:Lambda heater control

If Lambda is narrow band, it's a waste of time, more or less, though you could log the voltage. I thought you said you had a wideband? Or do you have both?

DaWaN wrote:Starter switch

Hmm, what needs doing here?

DaWaN wrote:I am also thinking of adding a knock sensor and logic spark outputs for COPs.

If you do, keep in mind that you'll need to drop back to semi sequential fueling, or run pseudo wasted spark with the COP units firing together. Later firmware will do what you want, though.

DaWaN wrote:I have quite some experience in designing hardware so I do not need very much help on the hardware part.

Keep the comms channels with us open anyway, as it's more about the right thing to do, than doing something that's theoretically good :-)

Fred.

Fred wrote: What's the difference?

Oops, my mistake

Fred wrote:No support planned for this right now, will likely require some sort of external IC counter to support until later fw revisions. And even then, something like 4 wheels + driveshaft will require something external.

The VSS signal is coming from the differential. In the OEM ECU it is used for VTEC engagement and idle control. I also find it very useful during datalogging as I can see in which gear I was driving

Fred wrote: What's required? Pretty simple or not?

Yes I do not think alternator control is rocket science. The ECU can control the charging voltage as far as I know. Just needs investigation

Fred wrote: If Lambda is narrow band, it's a waste of time, more or less, though you could log the voltage. I thought you said you had a wideband? Or do you have both?

I would like to see closed loop fuel control (needed for MOT here in NL), I do use my wideband for tuning and narrow band for normal driving (take the wideband out)

Fred wrote: Hmm, what needs doing here?

Another thing to check, I do not know why the OEM ecu has this, might not be essential

Fred wrote: If you do, keep in mind that you'll need to drop back to semi sequential fueling, or run pseudo wasted spark with the COP units firing together. Later firmware will do what you want, though.

Thanks for the info, I will decide later if I add COP support or not (not very essential)

Fred wrote: Keep the comms channels with us open anyway, as it's more about the right thing to do, than doing something that's theoretically good

Yes I will do that, I just wanted to let know that I do not want and going to consume much support time. In the mean time I do not want this to fail so I will ask for help if needed!

Info about your VSS sensor here please: viewtopic.php?f=41&t=1876

Closed loop fuel control is a "later" thing, don't expect any closed loop anything in the short term. All closed loop control requires optimised open loop control first anyway. It'll be unplug for MOT for a while :-)

Might be helpful for alternator control planning: http://i1091.photobucket.com/albums/i39 ... rnator.png

Fred.

I checked the alternator control, the starter switch and the lambda heater control

The alternator control can be skipped, if the pin is left floating the alternator will work fine
Grounding it will make it turn off, probably for economy / idle control but not essential

Same for starter switch, that is used for the ECU to know when it is started, most likely to open idle air valve or enrich engine or engage batch fire mode
Starter switch then does not seem to be vital

Lambda heater is there to let the lambda work fast on cold engine, not essential too

I also decided not to have hardware diagnostics (to detect open connections / shortcuts etc. on inputs/outputs), but only make the hardware resistant to any common wiring fault.

That said I know pretty much what is needed to make the schematics, I will start on that this week Already made a repository on GitHub.
CAD package will be KiCAD

You can already control the lambda heater with coolant temp, on under X :-)

Adding logic to turn off the alternator for some weird reason should be very easy too.

You may as well log the starter input as a bit too, no problem there.

Link to github account/repo? :-)

Fred.