gruss
Part of things
Posts: 242
|
|
Sept 11, 2012 19:35:40 GMT
|
Give it a sniff of gas (butane) and scope the injectors, if there is a hidden O2 sensor then you will see the Ecu attempt to take fuel away by decreasing the injector Duty. Then create an air leak and watch it try to add fuel.
What wires have you got for the maf, where do they go to and from...
|
|
Daihatsu Mira TR-XX Suzuki Alto Works
|
|
|
|
|
Sept 11, 2012 22:35:14 GMT
|
It definately does not have or need an O2 sensor, the original exhaust had a blank spot where the sensor would be, but there wasn't even a hole drilled. It also had a large bump where there could be an catalytic converter, but it was empty with no weld marks (also completely intact heat shielding). The manual also states that non-catalytic converter equipped cars do not have an O2 sensor but have a CO adjusting screw on the MAF. And last but not least, there is no shielded wiring going from the engine in the front to the ECU in the rear (under the parcel shelf). vitesseefi; You're right, they all get terrible mileage but mine went from 10L/100km to 20L/100km which is even for the biggest car nut a bit enthousiastic for a 1.8 with 150bhp You're also right about scoping the power feeds, that's the next thing on my list. Especially since I realised today (college was boring) that the main power feed to both the fuel pump and ECU goes through a manky old immobiliser/alarm which could somehow create interference between the two or just generate its own. News when it happens! edit: sorry gruss, forgot your question about the MAF. It has 5 wires, +12V, GND to body, GND to ECU, "burn off signal" (this activates after the engine is shut off to clean the MAF's hotwires) and a signal wire going straight to the ECU. 1 Of the grounds is not good, it's almost 160k ohm instead of 10 ohm but like I said before, a jumper wire brought it back to 6 ohm but didn't fix/change it/anything. This is also on my to-do list.
|
|
Last Edit: Sept 11, 2012 22:38:20 GMT by joostvdw
|
|
gruss
Part of things
Posts: 242
|
|
Sept 12, 2012 6:57:01 GMT
|
Ok scrap the 02 sensor then...
With the ignition key on engine off carry out some voltage checks on the ground wires. Probe the maf ground pin and the chassis ground point then the maf and Ecu ground. You want to have 100mv or less. Sounds like a ground issue. When you scope the power and signal to the maf does it change at all if you wiggle the plug or press on top if the maf. Also the signal voltage check should be 50mv or less. Basically full test the wiring. If in doubt just build your own harness between the maf and Ecu as a temp measure.
|
|
Daihatsu Mira TR-XX Suzuki Alto Works
|
|
|
|
Sept 12, 2012 22:32:07 GMT
|
Guess it didn't like be probed continuously, cold idle hunts between 1700rpm and 500rpm, it spits, coughs and backfires and is giving me all sorts of trouble. Just not a trouble code I can work with.
In other news I jumpered the immobiliser and no change. Tomorrow I will be disconnecting the TPS and/or adjusting it. And if the weather allows for it I will measure/fix the MAF grounds.
|
|
|
|
|
|
Sept 16, 2012 13:55:10 GMT
|
Is the short pulse an actual injection event, or does it just occur at random points in the cycle? Things that affect pulse width are air mass, air temp, coolant temp, rpm, and load. I would check all of those sensors first the take a look at coil primary current. What is the dwell doing when that short event occurs? A two or four channel scope would be a big help here.
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 16, 2012 14:00:52 GMT
|
Can you capture a longer timebase? I like to see at least two engine cycles on the screen when evaluating thingslike this.
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 18, 2012 9:12:08 GMT
|
Thanks for your replies Rev. The short pulse is as far as I can see an actual injection event. I can take a picture of some more pulses in one screen if you like? But I can tell you right now there doesn't seem to be a logic behind it... Sadly I only have a small single channel digital scope which has questionable accuracy.
I removed the MAF today and with the engine off, just blew into it, this created a stable signal. So the signal I see on the MAF is a result instead of a cause (!). I also solved the earthing problem by reseating all ground points on the engine and battery. I may need to invest in some new ground leads and battery terminals though, just to be safe.
Unplugging TPS, knock sensor, high idle solinoid (since I don't have AC this has no function but may be malfunctioning) and coolant sensor which made no difference to the injector pulse.
The ignition seems to be working just fine, I probed the ignition amplifier and the signal was very stable and consistent. I'm not sure if and how I can check the ignition dwell or how this is relevant? Not saying you're wrong, but I just don't understand;)
|
|
|
|
|
|
Sept 18, 2012 13:25:43 GMT
|
The reason I mention dwell, or rather dwell command, is that its a good indicator of the ecu reacting to an input. if your dwell time changes at the same point as the injector then there's a good chance the ecu is responding to a signal. therefore we would be looking at a symptom rather than a fault. does this vehicle have a map sensor or boost pressure sensor? also have you measured voltage drop at the ecu with the engine running?
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 18, 2012 14:45:21 GMT
|
Could you explain how I can measure dwell with my simple oscilloscope? The distributor has a slotted wheel with 2 sensors which the ECU uses for injection and ignition. As mentioned before, I did check if the ignition displayed the same behavior but it seemed very constant and consistent.
It has a MAF and a boost sensor, but the latter is only used for a boost light on the dashboard and popular belief is that it also triggers the ECU to go extra rich as a form of charge cooling since it's not equipped with an intercooler. But I think it's just a fancy light.
I have not checked voltage drop at the ECU, the ground is new and secure though. When it stops raining I will check.
|
|
|
|
|
|
Sept 18, 2012 16:17:36 GMT
|
The easiest way to check would be to scope the signal coming out of the ignition module. This is the primary voltage and your scope needs to be able to handle upto 100 volts to measure this. If not just measure the signal going into the module from the ecu. This should be a square wave on and off signal. something is ringing alarm bells about the dizzy. Is this where the engine speed signal comes from? if so I would be scoping the output of both of those sensors and make sure they both look identical.
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
|
Sept 18, 2012 17:58:54 GMT
|
I just replaced the dizzy with another one i had for spares and it didn't change a thing:( I did take some pictures of more injector waveforms for you so I will upload them shortly. I already scoped the amplifier output (this is the coil input) and it seemed fine by me. It just started to rain as I was typing this and the light is fading so it'll have to wait...
|
|
|
|
|
|
Sept 19, 2012 11:22:55 GMT
|
Just done a bit of digging and it looks like your ecu only controls fueling so the whole dwell thing is moot anyway. Do you know where the ecu picks up its speed signal? Is it from the dizzy?
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 19, 2012 12:09:29 GMT
|
As fas as I know the ecu does control ignition, the service manual describes a system of the ECU which advances the ignition till it starts knocking and then backs off a bit, this is to compensate for different octane fuels.
Like I said, inside the dizzy is a slotted disc which I guess is also the RPM signal to the ECU. It might pick it up from the coil though... I'll have a look in the manual.
|
|
|
|
|
|
Sept 19, 2012 12:37:53 GMT
|
That information did come from autodata so it wouldn't surprise me if it was wrong. I would hazard a guess though that the two pick ups in the dizzy would indicate seperate control units for fuel and spark. Do you have a wiring diagram?
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 19, 2012 13:15:56 GMT
|
It has 1 ECU which controls both spark and injection. But the 2 different "tracks" in the dizzy do indicate that the systems have their own singnals (the dizzy also has a 4 pin connector). I have a comprehensive subaru factory service manual, several variants (but sadly not exactly the one I have). But it's a PDF and I don't know how I can post 1 page from a PDF file in sufficiently high resolution so that you can actually read it. I can send you the whole PDF which is 14Mb?
Thanks for the help so far:)
|
|
Last Edit: Sept 19, 2012 13:16:20 GMT by joostvdw
|
|
|
|
Sept 19, 2012 21:49:07 GMT
|
Yeah that would be a great help. My email is mewster600 at gmail dot com. I do love a nice weird fault like this. just a shame I can't get my hands on it in person. I remember driving one of these Subaru coupes back in the nineties. Crazy thing that went into 4wd when the wipers were switched on.
|
|
R.S. Autotech. Servicing/Repairs/Diagnostics.
|
|
|
|
Sept 20, 2012 11:06:39 GMT
|
I've send you an email with the pdf attached, but I forgot to mention that page 39 is probably the page you're most interested in. Also, you're more than welcome to have a lookfor yourself, it's only a 700km drive from essex;)
|
|
|
|
|
|
Sept 23, 2012 14:04:36 GMT
|
Not knowing what to do next I decided to check the obvious, I checked the spark plug leads which were within factory spec, checked and measured both cap and rotor. Replaced the coil with a better looking one which also fell within factory spec. All to no avail I'm starting to wonder if this ECU might be bad as well, slim chance I know but there is not much more that can cause this. Next step will be measuring all sensoroutputs at the ECU in the boot, see if the wiring loom causes trouble.
|
|
|
|
|
|
Sept 28, 2012 19:40:26 GMT
|
It's weekend again (yay!) and with predicted fair weather I'm having another go with this. I'm going to check all outputs and inputs at the ECU and during the week I had another idea; the service manual states that the crank angle sensor uses a sensor signal and a reference signal, how would this work? What is the reference signal for and what would it look like? Maybe if this is corrupted somehow this means incorrect timing and whatever. Going to check that tomorrow:)
|
|
|
|
|
|
Sept 30, 2012 17:42:17 GMT
|
Alright, turns out the crank angle sensor works perfectly, took it all the way apart but nothing wrong with it, the signal to the ECU is exactly as I would expect from such a sensor. All the other in and outputs were also fine at both the engine and ECU side of things. Just to be sure I also did a compression test which was fine as well.
I'm really at an end now, I have no more ideas, perhaps the fuel pump, but that's more wild guessing and throwing parts at it than anything else and that's not the way forward. By looking at the amount of views and replies posted I'm also guessing that you don't have a clue either. But anybody who has any idea I haven't tried yet, please speak up as I'm open to anything right now...
|
|
|
|
|