Babalouie's 911 - start on p7

[irsa76]

I think what you really need though is a 1 piece moulded carbon fibre liner. That way it would cover all the "messy" bits and offer protection from errant detailing product bottles you carry. And look damm cool.

[Cplus]

Wrap it in the same material. If it still looks ass, just don't bother at all?

[Babalouie]

/me looks at the last 42 pages

Riiiiiiiiight...

:D

Alright, alright, it was bugging me and I had to fix it :D

When we aligned the bonnet the other day, we did it a really (in hindsight) dumb way, with the bonnet propped open. So I had to have two mates holding up the bonnet, to prevent it from sliding down on its slots when we loosened the bolts. But of course, the easier way to do it...is like this:


With it only open part-way, you can loosen the bolts as much as you like, and it'll just sit there. So then you can fudge around with it to your hearts content, nip up the bolts and drop it to check the gaps. And after a little experimentation, I think this is the best compromise:

If you look closely, the panel gap on the left is slightly thinner than on the right, but the bonnet edge lines up quite nice with the fenders now...and it's certainly better than it was.

And the gaps at the cowl end are reasonably uniform too.




From some angles, the right side corner maybe pokes out a smidge more than the left, but I couldn't get it any better without closing up the panel gap on the sides even more. I think I can live with this.


And...to celebrate Rennsport Festival this weekend, I've reinstalled the RS wheel caps :D

[Babalouie]

Lately, I've been having a few attempts at getting to the bottom of a few driveability issues. They seem to be quite common 964 maladies, and the first one is the hesitation after a period off the throttle: say you're rolling down a long hill with the throttle closed, and when you hit the pedal again, it seems to take a few heartbeats for the engine to respond.

The second issue, is the poor idle control in low speed, stop-go traffic. My car was particularly bad in this respect, being easy to stall, and reluctant to lope along in gear at tickover without a bit of kangarooing-along. I dunno how a 3.6litre engine with a 20kg flywheel becomes easy to stall, but in truth the Hako is considerably easier to drive smoothly in slow traffic (notwithstanding its Webers, massive cam and button clutch...)

My first thought was that it might be related to the air flow meter, so I begin by rummaging among the pile of bits that I inherited with the car. At the time, I thanked the car's seller (but was a little nonplussed as to what I'd do with all this junk), but over time I've come to realise that all this stuff is actually in really good, low mileage condition.


So the first step is to remove the airflow meter from the spare manifold, and first you unclip the air horn from the airbox


That gives access to the nuts that you have to remove, to separate the airbox from the AFM


The airbox is also bolted to the manifold, here...


..and inside the airbox here.


Then, the airbox lifts off, and you just undo a hose clamp, unclip the AFM plug, and the AFM can come right off, too


Now that the (spare) AFM is on the bench, I can test it, and with the flapper closed, the resistance reading is .55 ohms


If I crack open the flapper a hair, the resistance rises smoothly to .57 ohms


And with the flapper all the way open, the resistance falls to .3 ohms


Now, I have no idea if this is actually a good way to test an AFM or not And the factory manual gives a voltage range for testing the AFM instead of resistance (0.25V when closed, and 4.6V when fully-open). But when I repeated the same test with the AFM out of the car, the results were broadly the same. And...I figure, if there was a dead spot on the AFM tracks, then surely it would have given a funny resistance reading? Anyway...notwithstanding that, I swapped the AFM in the car for the spare one anyway, just to see if it might make any difference.


And you know what? It did make a difference. Low-rpm running was a bit crisper, and after a week's driving, I haven't encountered that off-throttle dead spot anymore.

But I figured that there was a better, and easier way of testing the car's electrical parts, and so I went and bought this, a T-OBD from BergvillFX.


What it is, is hardware that allows you to plug a laptop into the diagnostic port of the car, and it works with the free Scantool software. These are tools developed by guys on the Rennlist forum, which is a pretty awesome forum for Porsche stuff.

So the first thing you do, is kick your kids off Minecraft, and take their laptop down to the garage


The Scantool software is free, and the instructions at BergvillFX look pretty daunting to an IT-phobic like me, with lots of things like "edit the scantool.ini file with the COM PORT selected"....but there are plenty of screenshots in the step by step instructions, and it's pretty idiot-proof (hey, if I can do it...). Then you just plug in the adapter and fire up the laptop.


The first screen you see once it connects, is this, and thankfully...no error codes on the ecu.


Click on where it says "Motronic", and there's a drop down box, which gives you access to the ABS, Climate Control and TIP menus. You can scroll thru each menu to check for error codes, but the only ones I had were these, in the climate control section. They relate to the engine bay heater blower, which has been removed because I have a Cup bypass pipe back there, so this makes sense.


Then we go back to the "Motronic" menu, and check the ecu inputs. The engine is off at this point, by the way. The AFM (or MAF sensor) is reading 0.25V, just as per factory specs, so this is good. To test if it goes to 4.6V with the flapper open, I'll have to pop the airbox lid off, and move the flapper by hand, while someone reads the screen, but so far, so good. The injection system on the 964 is a little antiquated for its time, and this is reflected in how it doesn't have a proper throttle position sensor...instead it has two crude microswitches that work off the throttle linkage for idle and WOT. You can also see at this screen, that the idle switch is on, as it should be. Floor the accelerator, and the idle switch should go off, and the WOT switch come on.


Start the engine, and the battery charge goes to 13.5V (which is good) and the AFM voltage rises to 1.4v, as the airflow prises open the flapper a hair.


So far, it looks like the idle and WOT switches are working, and the AFM is fine (at low rpm anyway). Next thing to check is the operation of the cylinder head temp sensor. With the engine off, and stone cold, the sensor reads 18C, which sounds about right. Fire up the engine, and the temps rise to 40, then 50 degrees pretty much straight away, and after a long drive the temps max out at 117C (which I think is the upper limit of the sensor, which will read no higher). Phew...cylinder head temp sensor seems ok, too.


Now the last step is the famous Idle Adaptation. Click on the button marked...um..."ADAPTATION", and this box pops up. First, the engine has to be up to operating temp. Turn off the a/c, lights and anything else you got, press the Adaptation button, and then start the engine.


After about 30 seconds or so, it's done!


Now what is an Idle Adaptation? Well, as far as I understand it, the adaptation is like a hard reboot of the ecu's idle function, so that it returns to factory baseline settings. For some funny reason, you can unplug the battery and walk away from the car for a week, but the ecu still will retain the idle settings it's "learned". The only way to reset it, is to do it via the diagnostic port...

And it...WORKS!


Cold starting is definitely easier, and the idle is much smoother and more stable. It's much less easy to stall and the car's happier to mooch along at idle in slow traffic, with your foot off the throttle and no more kangarooing. Replacing the AFM helped remove that dead spot, but the adaptation has made it much more solid at low revs.

I'll report back after a few more days, but so far the adaptation seems to have done the trick. The Scantool and T-OBD is a revelation too, in that you can test a lot of things in the engine, very quickly and without dismantling anything. The next things to play with on the Scantool, is testing the O2 sensor and the max reading on the AFM.

[Boxer]

LOL @ stealing kiddies lappy - c'mon admit it, thats your wallpaper......gaming nerd

Scantools/flash tools are a good fun, the early systems are pretty basic though eh

[31GUN]

Hey this looks like fun Kev!

http://www.youtube.com/watch?feature...hjBNTA-I#at=71

[Babalouie]

Lately I've been noticing a bit of a squealing noise on cold start (and I'm paraphrasing here...) which goes something like "yip-yip-yip-yip-yip" It seems to go away after a minute or two, but then one day the belt sensor warning light went off, but when I popped the decklid to check, everything was fine. Everything pointed to a dud belt sensor, so I decided it was time to investigate.

When I first looked in the engine bay of the 964 years ago and noticed that the fan ran on two belts, I figured it was a nice little double-redundancy from Porsche: should one belt break, the other is still there as a backup.


But actually that isn't the case. The alternator lives inside the fan housing, and is driven by one of the pulleys. The fan rides on a bearing (which then rides on) the alternator shaft, and that's what's being run by the second pulley. So while the fan is supported by the alternator shaft, it actually spins at a slightly different speed. If the alternator belt breaks, then the charge light will illuminate, but if the fan belt breaks...

...then there is this sensor which will trigger a warning light. It consists of a little spring loaded wheel, and if the belt breaks, the wheel pops out and triggers the microswitch which then illuminates the warning light on the dash.


To replace, it's really a 10 minute job. It's only held in by one allen head bolt, which is easy to reach. Oh by the way, I wholeheartedly recommend wedging a rag underneath it all, so that the bolt doesn't fall out and roll into that spot under the crank pulley (from which lost bolts and washers take hours to extricate).


Before I ordered a new one, I just pushed-in the belt sensor wheel of the old one by hand, and the wheel did seem to spin in a very notchy and gravelly way. The new one is noticeably more smooth, so I'd say the bearing on the old sensor wheel has had its day.


With the new one in place...no more squealing!


...another day, another random Porsche repair :D

[itsnotagsr]

If your day job doesn't work out, you'll be able to open your own workshop soon Kev

[Babalouie]

That's not possible...because I will be too busy fixing random shit on this car for the next 100yrs

[Gammaboy]

Haven't contemplated a alpha-N conversion on the motronic to bin the flapper? Will require a real tps though.

[Babalouie]

Babalouie's 911 - start on p7

Nah, I might do a Ruf ecu at some point, but something like a Motec conversion sounds like a PITA

[BigMuz]

I missed a couple of updates- but when you are marking out on your aluminium you need to use a much finer marker- a scriber is best but a biro will do. The thick texta lines you use add some errors in and it adds up to making things look a bit amateurish. You'll get everything bang on centre/ in line if you use a much thinner marking device.

Cheers

[Babalouie]

Fairy nuff.

[Gammaboy]

Nah, I might do a Ruf ecu at some point, but something like a Motec conversion sounds like a PITA

Maxx Alpha-N is a plug in unit that keeps the rest of the motronic mapping, but bins the afm. I was seriously considering one for the 164. The E30 M3 guys use them a bit.

[clutch-monkey]

Maxx Alpha-N is a plug in unit that keeps the rest of the motronic mapping, but bins the afm. I was seriously considering one for the 164. The E30 M3 guys use them a bit.

that sounds about ideal..

someday something like this would be tits