steed wrote:this level of detail during bench testing pre-fitment is phenominal! most impressive work
Yesss, i want to know for shure all the components i throw under the bonnet are reliable, good working and within specs. Good testing and step by step install saves lots of time and energy afterwards.
The car is a dayly driver, mostly driven by my wife, so i cannot permit any errors or problems
Thanks,
The MAP sensor needs a decent vacuum connection to the intake manifolt. Ofcource the best place to connect the MAP sensor is the plenum area.
In my opinion the best place is the same location as the vacuum hose for the pressure regulator.
The simplest solution is a T piece in the hose to the pressure regulator, but after a good look at the lid where the vacuum connection is situated, i thought this is the best place to make a second vacuum connection for the MAP sensor.
This location in the lid/plenum is best protected for any dirt/oil getting into the vacuum hoses.
So i pulled a vacuum hose connector pipe out of a old carburettor which i had laying around. this is a 4mm brass pipe
Drilled a second hole in the lid, 3,5mm through, and then drilled again with a 4mm dril, almost through
Then the brass pipe fittet perfectly in the lid.
As always, a picture tells so much more:
For the MAT sensor i drilled another hole in the intake manifolt, near the vacuum connections, and Threaded it.
I've cleaned the injectors in a Ultrasonic cleaner bath, and at the same time squirting brake cleaner fluid through them.
In the ultrasonic cleaner bath i've used a water based detergent. the injector is connected to a 2 bar air pressurized cannistor (soda bottle) which holds a few CC brake cleaner. While haning in the ultrasonic bath, once and a while i activated the injector for a short time so a little brake cleaner flows through it.
This is my setup:
After cleanining the injectors i did a new bench test:
Bench testing the injectors
My second bench test setup is a little different from the first, this time i've used clear beerbottles, an hooked up a scope to inspect the injector output voltage and the flyback clamping cirquit.
Here are some pictures and a very explaining video of the bench test:
ECU with stimulator, laptop and scope hooked up:
Fuelpump and injectors with fuelrail:
Scope signal,
Yellow = output voltage over injectors, clamped at 39.5 volts
Blue = voltage over flyback clamping cirquit (base of transistor)
And here the video with explenaition in it:
After the test all bottles are equaly filled:
Results:
injector......weight bruto....weight bottle.....weight netto (squirted fuel in test run)
..1.................424..................205..................219
..2.................426..................205..................221
..3.................425..................205..................220
..4.................426..................204..................222
..5.................428..................205..................223
..6.................424..................204..................220
Not bad, no more than 2% difference in flow rate between the six injectors.
Today i've welded in a bung for the O2 sensor in the exhaust and mounted the O2 sensor.
I've got a new O2 sensor (narrow band to start with) from ebay for only € 14,05 include shipment and a O2 sensor bung for € 2,97 include shipment
I had planned to weld the bung in the front exhaust pipe coming from the front manifold (cil 1-3) so i could place the O2 sensor 50cm downstream from the manifold. Unfortunaly the bolts would not come lose, and i did not wat to break a bolt.
This forced me to take out the second exhaust down pipe (cil 4-6) , luckily these bolts came lose. Now the bung for the O2 sensor is welded in the downpipe only a few centimeters away from the manifold. Hopefuly the hot exhaust gases won't destroy the O2 sensor.
When the whole setup is working well i'll throw in a wideband O2 sensor, but they don't come cheap from china
Here some pics:
Bung welded and O2 sensor test fit:
O2 sensor in place:
I've done a test drive (still running the 4A1 carburetor), and the "AFR" sensor in tunerstudio goes wild from 0,1 volts to 0,9 volts and back. Most of the time 0.9 volts which means "running rich". Sensor seems to do the job.
The exhaust system needs replacement soon, then i'll weld in a bung on the right (best) location.
It's started to rain, and the car is on the driveway. so a little time for a update, and a cup of coffee
You cannot attach a fully assembled intake manifolt to the engine, when the thermostat housing is already mounted. the fuelrail interferes with the bleeding notch on the thermostat housing. not a big issue.
I've encountered 2 minor problems:
The oil dipstick is right in front of the air intake, is there a way to unmount the dipstick assembly from the engine?
And the vacuum assembly on the throttle housing interferes with the brakefluid container:
Fuel line would not hold the pressie and leaks now. The metal fuelline is a little bit rusty at the end and sprays a fine dust of fuel when the pump is running. Approx 10 cm at the end.
Bad news first : the thermostat housing is leaking coolant, so i'd to drain the system again to fix the problem.
Second, the fuel line problem is fixed for now but the lines need to be renewed for the next MOT.
Good news: engine started right away and has a good throttle responce.
Now fixing the leaky thermostat. Want to make a testdrive.
Stay tuned
Last edited by BertjeConti on Sun Jun 15, 2014 10:19 am, edited 1 time in total.
Made a first testtrip, ran good cold but when engine warmed up throttle responce went bad.
Fixed it by turning of the "wall wetting algorithem". After that the engine ran good , had steady idle at 1000 rpm, and proper throttle responce.
Made a 50km trip without any problems, no misfires , good accelleration, and more torque at low rpm.
Did some flatout sprints during the testdrive, the engine runs much better at high revving.
Seems that my initial fueltable is almost spot on.
Only thing that i saw on the computer is that the automatic fuel correction wasnt working properly. It corrected each time to a lean mixture and stayed lean. I'm running a narrow band o2sensor. Wide band is a prefered option.
Still lots to do:
- Fabricating a bracket to hold the airfilter box
- cleaning up the old wiring (dizzy and carb )
- and tuning ofcourse (thats the goal and fun factor)