[T3] AAR woes

[Jim Adney]

On 23 May 2017 at 21:24, Daniel Nohejl wrote:

> > There is also some amount of leakage from the outside (not the inlet) 
> > because the sealing of the lid is minimal and there is also leakage up along 
> > the shaft. All of this is normal.
> 
> But is it not abnormal that my motor stalled when I covered the hose
> b/w the AAR and IAD. I figure it´d have to be pulling loads of extra
> air for my thumb over the hose to stall out the motor. 

My understand was that you did this after adjusting the idle back down. 
When you did this with the AAR not completely "closed" there was still 
significant air coming thru the AAR, so that with the idle screw closed way 
down AND the AAR airflow now cut completely off, there was no longer 
enough air getting into the IAD to sustain the engine.

So, if I understand what you did correctly, what happened was completely 
normal.

> > The way to check them is to close off the inlet hose with your thumb and see 
> > how much this effects the idle speed. There should be little or no effect if the 
> > engine is FULLY WARMED UP.
> 
> It was definitely fully warmed up both in the morning (highway and
> traffic conditions) and in the PM I´d been driving for over an hour in
> traffic and 45 mph conditions which is always far more than enough for
> the idle to drop down. 

Okay, good. In this case, it sounds like the AAR was not adjusted correctly 
and I should re-do both (all?) of them. Sounds like I need to rethink how I do 
these.  

> It seems I don´t understand something about the calibration. I´d
> assumed that the +/- adjustment simply determined how open the valve
> is on cold start and perhaps how long it takes to close. I didn´t
> think it´d determine whether the valve closed at all though. I say
> that understanding that generally when our AAR is working right
> thereTMs only a ?0-75 rpm drop when I pull the hose and cover th? end
> and that itTMs never really entirely closed. 

The only adjustment one can make is to try to make it be fully "closed" when 
the bimetal spring reaches ~122 F. There's an internal stop that it should 
reach at that temperature and that's what counts as "closed." What I've done 
to try to achieve this was to make a fixture that I can use at room temp that 
should get the shaft to be turned against the stop when it reaches ~122 F. 
It's possible that this approach just doesn't work, but in the case of your 
AARs, I'm afraid that I did the adjustment in a cold basement when I should 
have done it warmer.

> > I know I've said this many times before, but it's worth repeating: The AAR 
> > has absolutely NO EFFECT on the car when you're driving. It only effects 
> > the idle speed. It does NOT change the richness/leanness of the FI at any 
> > time.
> 
> Not that it matters and not to pit one person against another, but I
> recall Ray Greenwood admonishing me in the past on The Samba about
> making sure my AAR closed all the way. Allegedly if it doesn´t, there
> are a variety of timing, injection timing, and MPS related issues that
> follow from an AAR that won´t close. 

I hardly know where to start here.

Injection timing is completely determined once you get the ignition timing set 
correctly to TDC. If you adjust the ignition timing at an rpm that's higher than 
1000, the mech adv will already be working its way up the advance curve 
and this will not be correctly set. This means that both the ignition and 
injection timing will be somewhat retarded, which is bad for both. The 
injection timing does NOT change with either the mech or vac advance, but it 
does depend on the ignition timing being set correctly.  

I've heard various people claim that correct injection timing is very important, 
but this ignores the fact that Bosch advanced the injection timing by about 
30 deg (60 crankshaft deg) in the '72-3 distributors. You can swap 
distributors between early and late engines and notice almost no difference. 
Bosch/VW claimed that the change was made so that the injected gas sat on 
the back of all the valves a bit for all cylinders. In the early engines, 2 
cylinders got their injection pulses during the intake stroke while the other 2 
got theirs ahead of time, while the intake valve was closed. Personally, I've 
not been able to see any real difference.

The MPS senses the amount of vacuum in the IAD. That vacuum depends 
on engine displacement and RPM and on the amount of air that's getting into 
the IAD. There are 3 sources for intake air: the intake butterfly valve, the 
AAR, and the idle air adjustment. The MPS doesn't know, or care, which 
path the intake air came thru. To the extent that the MPS is able to 
accurately gauge the manifold vacuum and the ECU is able to turn its data 
into correct injection pulses, there's no difference between extra air getting in 
thru an AAR that's still partly open and an idle air screw that's open a bit 
more.

The idle air screw is nothing but an adjustable leak. The AAR is a 
temperature-dependent leak. The intake butterfly is just one more imperfect 
valve. It never closes completely either. The idle air screw is there to allow 
us to add just enough more air at idle to get the engine to keep running at 
~900 rpm; it's manually adjustable because the other "leaks" change as time 
passes and parts wear.

So what happens when you open the throttle a bit? You let in more air via 
the butterfly but some continues to come in via the idle path and the AAR. 
You move the butterfly to give the engine more air. The MAP sensor 
measures the vacuum that results and the ECU sends it's calculation of the 
necessary gas to the injectors. You get more power. If the AAR is open a bit, 
you end up applying less pedal, but you don't notice the difference because 
it's imperceptable and your brain/foot feedback does this automatically 
without your having to think about it.

The short story is that the MAP sensor is always active. It and the ECU are 
always responding to the vacuum in the IAD. Nobody knows where that air is 
coming from and nobody cares. It's just what's there and that's all we care 
about.

One caveat: What I've said about the various air "leaks" applies to any air 
leaks that occur at the IAD. Air leaks that occur out at the ends of the air 
runners DO confuse things, because those don't have the same effect on the 
vacuum in the IAD.



-- 
*******************************
Jim Adney, jadney at vwtype3.org
Madison, Wisconsin, USA
*******************************