Need help with starting my VW bus

[Heribert]

On the subject of :

quote from Lance´s post Some serial ports are a bit 'funny' in that they seem to mess with the 5V supply on MegaSquirt.

Has been seen by some people on MSforums over the years, including me.
A sure fire way to avoid this totally is to modify the DB9 cable going from the MS to the
"messing" PC.
Open the contact in one end of he cable and make sure that only pins 5 ( ground ref) and 2/3 ( Rx and Tx) are connected.
Simply cut the other wires.
Pin 1 is the offending pin, since it carries +5 V in some PC RS 232 serial ports.

A cable thus modified will perform all MS duties, including reflashing of code both for MS1 andMSII

But it will certainly not function with the megaprogrammer to flash ablank CPU and probably also the Wilette programmer will dislike it.

Heribert

[Bernard Fife]

Heribert,

Yeah, that's a good solution.

Another way to do this is to break the pins off in the cable end that attaches to MegaSquirt using needle nose pilers. You have to make sure you leave pins 2, 3, and 5, as you say.

The pin numbers are usually imprinted in the plastic surrounding the pins (though they are small and can be hard to read).

Lance.

[panel]

Just to confirm.........cut pins 1 and 4 if I wish?

[Bernard Fife]

Yes, and 6,7,8,9 too.

[panel]

Cut the pins and ran the newer laptop and ran TS again. Got it running a little easier and then also tried that hot start with '0' in the 'Cranking Position'(extended)(steps) bin and it didn't do the high rev when hot problem. This will be nice when I start driving her again.

Didn't start awesome but I think I'll still have to tweak the 'Cranking Pulse Widths' a little more. I tried raising it up in the 0.2ms increments from those lower settings until it started fairly reasonably ........but like you said Lance there might be additional fuel in the manifold from possible multiple 'Prime Pulse' tries used first for starting.

I'm going to try one more time when I get home from work tomorrow and hopefully those small changes in the DB9 cable,PWM Current Limit(%) raising value, smaller starting with the cranking pulse widths,CLT location change etc. Seems like we might be onto something(fingers crossed).

Let me know if there is any resets you see or any problems in the cold log. It's been sitting all day(24hrs).

One question........how would you know what a good 'cranking pulse width' should be from looking at a log when you see a ?#ms while cranking but have to take into affect ?#+open_time+volts_adjust+ etc.

For instance at 864.938s and the CLT is @ 65.9F it looks like it's starting pretty good there. Though I've played with the cranking pulse widths to get to that spot. And the 'cranking pulse widths' are 3.00/2.99ms.

My question is this.......at that TEMP bin what's the actual ms # you put in there from the above calculation?

~Marc.

[Bernard Fife]

Marc,

Cranking pulse width is a bit like accel enrichment, in that it's hard to tell what you should do from a log. In theory you can check the time between the start of cranking and running on a log, but this doesn't take into account the history (like if the engine has just been run, has had multiple prime pulses, etc.). And like accel, an EGO sensor is of no use at all.

So it's one of those thing you just have to work on by 'feel' and common sense. If it starts poorly when cold, it's either too lean or too rich. First assume it's too lean and reduce the fuel - if that makes it worse, increase it above the original value. Once you get it close you can fine tune over a few dozen cold starts (which may take weeks, but only a minute or two at a time). Hot starts are easier, because you can test those repeatedly (though you should leave at least a few minutes between attempts - but nowhere near the 10+ hours required for cold starts). And getting the hot start dialled in can help guide you on the lower temps too (it should be a fairly smooth increasing progression in pulse widths from hot to cold).

However, once you have it all set up, you never have to touch it again (unless you change hardware).

Lance.

[panel]

Hi guys,

Tried it out again this afternoon and it seems as though it's getting easier to start. I did a log for anyone to have a look and it smelt a little rich while cranking again. I did go up in .2ms increments but didn't really help I don't think. So I went the other way and then it seemed to start better. Maybe cause there was fuel in there from the prime pules or from the previously cranking pulse widths.(that's where I need some help I guess).

I think this is where it finally started. Question is though.........it shows 3.2 PW at a 66F CLT .........



...........but in my tuning it shows less?



Later.

[Bernard Fife]

Marc,

At 66*F, you would expect the 'base' pulse width to be between 2.1 (at 60*F) and 1.8 (at 80*F).

Let's call it 2.0. Then you add:

- 1.0 milliseconds for opening time (I think this is what you were using in one of your MSQs, which I think is the default value),
- a small amount for the voltage correction on the opening time (assuming your volts while cranking is less than 12.0V). I think the default is 0.2 ms/V, so if your cranking voltage is 10.5, that adds 0.3 milliseconds,
- a small negative amount for the air temperature being colder than the standard 70*F (this would be a small effect, say 0.1 msec).

So 3.2 looks about right to me.

Lance.

[panel]

Also it backfires a few times while trying to start.

Does anyone know what causes this? It does it every now and then. Always in the mornings never when hot. Should I be retarding it a bit more than my cranking at 10 degrees?

Set ignition at 5-10 BTDC until you get reasonably close to a decent cold start.

<----from Heribert.

[panel]

Haven't driven it in a few days so I thought I'd try to start it again. There wasn't even a hint of it trying to start. I tried lowering the cranking degrees and that didn't do it. I can smell fuel coming out the tailpipe and little puffs of smoke out the tailpipe but still no fire in the hole.

Any more suggestions fellas?

[panel]

Maybe my injector placement is wrong? Could it be hitting the walls and just dribbling in the cumbustion chamber? When the bus runs maybe there is enough suction/vacuum to bring the spray towards the back of the valve more? I've seen CB Performance selling them inboard just like the one I've built but also some factory manifolds have them outboard.

[FixItAgainTony]

No new ideas, but rereading this - was a start attempted with the SAW line disconnected to force the EDIS to limp home mode? Assuming limp home mode is set for 5~10 degrees BTDC, that might help to eliminate ignition wander as part of the problem.

Very nice work on the engine.

- Charles.

"puffs of smoke / backfire"

Set ignition at 5-10 BTDC until you get reasonably close to a decent cold start.
<----from Heribert.

EDIS is great. The slower it cranks, the closer to TDC the cranking spark fires. Try unplugging the SAW ( pin 3) on a cold start. Any difference?
<----from Heribert.

[panel]

Thanks Tony,

Just tried this today and it didn't really make any noticeable difference. I've had the cranking in my MSQ set for 10deg for awhile now just to help the motor start up better but I think it's mostly a fuel problem I'm dealing with

I haven't posted the MSQ because it kept changing the 'cranking pulse widths' so it's not really the right file. I was changing it through out the starting process so I don't think it would help anyone out.

I did get it running again but look how long it takes me. I was probably on and off cranking for close to 5-6 mins.

Have a look.

[Heribert]

Again, nice work. Real sweet!
I can not imagine that it is anything but overrich fuelling, since it actually smells gas out of the tailpipe.
Taking plugs out of your engine can be a bit of a bother, but if it were mine I would certainly do that at the next hard start occasion b4 it fires up.
That should tell you what is going on.

I am a bit stumped how you actualy get it to start after 5-6 minutes , w/o pulling the plugs and ventilate the cylinders.

Take away priming pulses( the injector primping pulses.

And how about , after extended cranking without luck, you unplug the injector wiring plugs and then crank. My guess is that after a
couple of crankshaft turns it will catch for a brief instant ( < 1 s) and the die.

So, base tip is overly rich cold cranking, somehow. And BC is not all that cold even.

Heribert

[panel]

Today I tried no prime pulse and still nothing. Even after awhile of cranking I pulled the inj fuses out to see if it would just run a little but it just kept cranking the same way. I also tried to take a lot of the cranking pulse widths right down to 2ms and even lower but with those compensations like Lance said in the above posts it's in the log as a higher value.

Man does this thing really back fire so LOUD I even checked the timing while cranking to see if I was off and it looked good on the video replay.

~Marc.

PS.........just pulled the plugs and their wet

[Heribert]

Wet plugs? Not a good sign for easy starts .

Is there any risk at all that your injectors are leaking so that the fuel seeps into the intake tract when you shut off.
And then gets drawn into the cylindrs at the first suction stroke.

In my experience , the pressure on the fuel system should hold good for several days ( Bosch FP and Bosch FPR, off of old Volvos) after a switch off and the leak back is backwards thru the pump and to some extent fwd thru the FPR. But nothing shlould seep thru the injectors.

Heribert

[panel]

Hi again guys,

I tried going higher with the cranking pulse widths and then after awhile I went low again and no fire at all.

Just finished trying this again and after about 5 mins of trying again the fuel pump is staying on full and the injectors are clicking away all the time. And now the the motor won't turn over
What did I do?

[FixItAgainTony]

Marc,
“…fuel pump is staying on full and the injectors are clicking away all the time. And now the motor won't turn over”
Did this happen with the engine not cranking? If so, check for hydro-lock.
If the fuel pump stays on even when not cranking it might indicate something wrong on the electrical side. The fuel pump should time out after about 2 seconds of the ignition on and no crank signal. Also, the injectors should only fire when there is a tach signal. The earlier logs all showed a solid tach signal, so this does not make a lot of sense, unless something else is causing the injectors to come on.

Is the fly-back circuit installed on the V2.2 board? Looking at the photos on “click”, it looks like it is there. Are all of the wires / components on it okay? It might be worth checking the U7 injector driver area if the injectors are on when they should not be. Your board was built in 2005 - If memory serves, U7 was an MC34151P at that time which has since been upgraded IXDI404PI-ND. I seem to remember the older injector driver failing in some instances.

- Charles.