Members Page/Gerbenvd Lei - VEMS wiki www.vems.hu

MembersPage/GerbenvdLei (2006-03-16 04:43:08)

Hi,

My projectcar is a 205 GTI with a 1.9 litre 8 valve engine. I turbocharged it some time ago and am now fully rebuilding/upgrading the engine to suit high boost. Have a look at my project webpage http://www.205turbo.nl

specs in a nutshell:

alloy 1905 CC wetliner engine

16 valve cyl head

Mitsu td05 16g turbo

Staging

I added 4 injectors to my engine. So I now have 4 injectors aiming at the valves and 4 upstream. The secondary injectors are about 140% of the primary injectors.

Configuration:

I configured VEMS to do injector staging.

rate to calculated pulsewidth to 60
- is this the inj_stage2_rate= ... ?
  - Yes indeed:

-snip from http://www.vems.hu/wiki/index.php?page=GenBoard%2FUnderDevelopment%2FStagedInjectors

config.inj_stage2_rate - Secondary injectors firing pulsewidth rate. Give it in 0.4 percents relative to primaries. Because a typo now 0x00 means 100%, 0xFF means 200%. (Will be corrected soon to 0-100%)

-snap-

This means if my secondary injectors are about 140% of the size of the normal injectors. I would have to use: 0.4 * 140 right?

END OF ADDED TEXT

TPS position to 0%
MAP value to 120 kpa.

But when te engine reaches 120 kpa is stalls. So I tested it with min MAP @ 1 kpa and the secondary injectors are fired but it seems the primary pulsewith is not reduced i.e. it runs VERY rich when firing the secondary injectors. It dies because of the amount of fuel.

BEGIN OF ADDED TEXT:

You mean the primary pulsewidth is not reduced compared to MAP = 0.5 kPa ?

indeed. It SEEMS the primary injectors are fired at the same pulsewidth even when the secondaries are fired. With the secondaries diconnected the car behaves normal. (does not lean out above the set kpa) I will have a try tomorrow with the primaries disconnected and only the secondaries connected.

1kPa threshold does not seem reasonable.

I use 1kpa to see if the injectors are fired at all. And to do some testing while the car is idling. i.e. this is not a common situation but nice to do some troubleshooting without the need for boost.

Just a thought: could it be it has to do with alternating of the injectors? banks 1-4 are alternated. If the secondaries are fired at each primary event that would mean it would deliver 400% of the required fuel.

END OF ADDED TEXT

I have the secondary injectors on channel 7 (ie pin 17) and the primary injectors on channels 1-4

Anybody suggestions?

firmware version
full config (in mcd dump form)
tables (especially h[0])

Tachometer output

I wired the tachometer to EC-18 pin 12 (SPECFET 0) and set the folowing parameters in megatune:

Tachometer channel: SPECFET

Tachometer channel: 0

Tachometer divider: 0

in the help it says:

Most GenBoard outputs is only pulling the output low (0 volts)

If your tachometer wants 12V or 5V signals, wire up a (say, 1kOhm to +12V or +5V) external pullup resistor. Configure Tachometer Divider if you want a lower frequency to your tachometer.

Check out the latest MegaTune release package (mt-r27), it has 1.0.19 firmware which has the tach output working.

OLD / solved issues for reference by other users perhaps experiencing simmilar problems

Megatune showed wrong voltage:

Only problem is the volt meter in Megatune. It gives 20 volt readings while battery voltage is 13 v.
- You probably missed to calibrate it, adjust battery calibration. Default batt_cal=0xBC (decimal 188) or so
  - thanks for that!

Ignition problems. Fuel worked but no sprak.

Publish full config, saved by mcd command.
Try to compare your ignition related parameters with a working config, such as MembersPage/MarcellGal/EngineSwap .
As the last chance, start again from the working config, only change what is necessary.
check again if you missed something from GenBoard/UnderDevelopment/FirmwareChanges
Eg. make sure secondary_trigger=02 is used ?

Thanks for the feedback! It turned out to be the secondary trigger indeed.

Took it for a drive this day. But the engine ran verry lean (afr 1:16) and it appeared all the tables are ampty somehow in megatune. I had this problem with a more recent firware before. Ans switched to a bit older one that seemed to work fine. But now the tables are empty again. Does anybode have any idea how this could be fixed?

A: There was a ini update on the magatune page.

http://www.vems.hu/wiki/index.php?page=MegaTune

<<This package has bug in ini file causing settings lost when updating to ECU, here is [fixed ini file]>>

The engine didn't start -> no fuel injection.

the pulse width megatune displays is 0 though the crancking pulsewidth is set. And the the req_fuel is set(and all tables are filled with the right values)

A: somehow the decel fuel cut rpm was set to 0 rpm after a firmware update. I set is back to 2500 rpm and it started again.