Lago Blue

is really only incidental to this post. But that is just a more fun title than looking for and finding better low-speed drive-ability and fuel economy. However we, being geeks, tend to sometimes focus on individual components, rather than how they can work together as a group. This is about how on those terms (drive-ability and economy), my pretty-much OEM induction system is currently functioning better than ever before.

Background (my system previously):

Intake Tour II

Update:
- removed fancy work-shopped I/M from the above post; and replaced with a gutted stock one;
- installed a big-mouth 3-wire MAF; &
- the latest vacuum leak fixes.

<u>Stock Intake Manifold:</u>
I expect that my current I/M performs as the original design intended from idle to just below switch-over. It has no secondary butterflies (only because I needed the parts to complete the other one), no port-work, and has only been cleaned, resurfaced (trued flat) for improved vacuum seal on all of it's 13(?) facets/surfaces, then aviation-sealed and torqued back-together. I also expect that further gains in economy will be had with the fancier manifold, but my point here is that for the results shown below, the one with the port-work; wasn't necessary.

<u>My big fat green DIY MAF housing mod</u>
Why is it green? Because, after cabinet-blasting it's exterior, I treated it (on the outside only) with Easy-Off oven cleaner which has resulted in a flat dark olive-green finish. This may allow more heat transfer out (hot-wires run at 120° C apparently); or in (does it pick-up engine heat from the exhaust manifold &/or rad air?), I have no idea if either or both are occurring at various times, or is a good or bad thing. I'm just listing it as a change. For the purpose of this post it probably matters not.

I wanted to lean out the mixture, in cruise. Would it be possible to use the MAF to move the point at which the system balanced itself when it was looping in cruise? If one was able to move the point where the system ran, would it learn around that change and go back to as before, or would the change; stick? Apparently my '93 has less learning ability than newer versions, but I'm not sure if that gives it any advantage here WRT the MAF signal.

I saw a System Authority chart for a Motronic system which showed the influence of the MAF in "Warm Cruise" on Base Injector "On" Time (or Pulse Width). I can't find that little chart right now, but even better, here is an Audi training manual that talks about that and more, see "System Authority", the second paragraph on page 28 (29 of 61 on the toolbar)...

http://atomic.speculation.org/Graphi...rse_941002.pdf

On the following page someone has filled in the entire chart by hand. It's for a newer car, but I presumed ours may be similar in some respects. When I saw this chart, I started to understand what VAP was doing with the O-rings. There must be a range within which, the system would stay running leaner yet stay happy; no excessive lean-ness and no lean codes.

I wasn't unhappy with my modified medium-mouth MAF, so why do a "big" MAF housing? (I would really like to avoid an old debate, so let me say that I make no claims for this DIY housing, other than given the right conditions, it alone; will generate lean codes). Because, despite having a CTS and T-stat that I'd tested to prove they where working well; and fully functional O2 sensors dithering nicely (fast enough and full-range), I still had both:

a.) rich looking tail-pipes; &
b.) yet still rather wide injector on-times.

These two things seemed to be contradictory to what the O2's where "saying" (I was able to "listen" with a very nice DMM), both signal's averages said slightly "lean".

The reason I wanted to "do" (i.e. de-post) a MAF housing was not just for increased bulk flow (see post #4 in below link):

https://www.audiworld.com/forums/sho...AF+restriction

but increased bypass air; detoured around the sensor. Why increase the by-pass? I had several reasons to think this might work. Thanks to other folks posting their work here, I knew:

a.) de-posting a "medium" housing apparently doesn't work;
b.) that one could <i>get</i> lean codes with a fat MAF. I don't really want a lean code, I want to <i>nearly</i> have one (see post #2 in below link);

https://www.audiworld.com/forums/sho...light=MAF+ring

c.) even though I was concerned that with such a mod, low speed sensor response (and consequently, low-speed drive-ability) would suffer, that at least full range MAF sensor output voltages (1.35 to 4.85V) could be achieved; at least with VAP's crudest fat housing (see below link);

https://www.audiworld.com/forums/sho...ght=MAF+output

and most importantly...

d.) going into this that if I went to far, I could correct back towards "stoich" using O-rings, as VAP also describes here:

https://www.audiworld.com/forums/sho...ight=lean+code

And that is exactly what happened, sort of. As I removed more metal, I would stop work, re-install it on the car and see what happened. On the fancy manifold which was installed at the time, no lean codes ever occurred. Life was good.

After re-installing the stock manifold, I attempted to run the fat MAF. Eventually, while in cruise at 80 kph, lean codes would appear. First one, then the other side would illuminate. They would also disappear at first, but ultimately return; and "stick" on. Re-installation of the medium MAF would prevent the lean codes, and so did an O-ring on the fat MAF's sensor. Mileage suffered a little, but low speed drive-ability was better than with the lean codes.

Which was all fine, until:

<u>The Latest Vacuum Leak Fixes:</u>

-The ICS Adapter Block that fits between ICS and I/M. This was revealed by doing a smoke test. It's rubber coating has deteriorated through age and I suspect oil contamination, similar to what happens to the EGR Solenoid valve. Another reason to do VAP's cam-cover vent stand-pipes mod.

-The Seal for the large diameter hard vacuum line from the I/M to the Brake-booster that installs into front of the Booster's shell. Previously I had used a short section of soft silicone hose to fix this leak, but the OEM seal was now just too worn. The leak may be detected by hearing the trapped vacuum leak away as the big vac line is wiggled slightly after the engine is shut down. The new replacement, set beside the old part really high-lighted why the leak occurred, the old part was shrunken, deformed and noticeably worn on it's I.D. where the barbed line fitting plugs in. The new seal's resiliance was such that installation required that the seal be lubed; and the car be in-gear.

-The Seal for the Master Cylinder where it installs into the Brake-booster. It's a square-sectioned O-ring. This leak I could hear if I shut the engine down while the hood was up. Makes a goose honk/slide-whistle kind of noise while the vacuum bleeds away.

<u>Results:</u>
I've run out of time for again today, but will add to this as soon as I can. Hint: the result of fixing these leaks got me to where I am now and involves both the...

FPR and MAF: see FPR function explanation at the 2nd link above at page 14 (15 of 61 on the toolbar);

...but not quite as I'd expected. Can you guess what happened after I fixed the above latest leaks?

Lago Blue

Here are some exact (with the help of Google Maps & friends) mileage calculations:

<u>Highway:</u>

Kingston to Ajax (near Toronto); & Rtn on the Trans-Canada: fuel pump to fuel pump, exactly: 428 kms.

Total fuel consumed: 16.040 (east to west) + 15.442 (west to east) = 31.482 Ltrs.

31.482 Ltrs. / 428 Kms. = x Ltrs. / 100 Kms.

x = 7.3556 Ltrs/100kms.
= 38.4 Mpg (Imp.)
= 32.0 Mpg (U.S.)

<u>City:</u>

Ajax to Richmond Hill; & Rtn: 91.6 Kms.

Total fuel consumed: 8.042 Ltrs.

8.042 Ltrs. / 91.6 Kms. = x Ltrs. / 100 Kms.

x = 8.779 Ltrs/100kms.
= 32.2 Mpg (Imp.)
= 26.8 Mpg (U.S.)

Ambient temp. was 23.5 - 26.5° C. You can see there was a little head-wind going one-way, but by retracing my route, it's effect is zeroed.

As I've said previously, I think the 20v I5 and 12v V6 are extremely similar:

"If I may attempt to illustrate by way of comparison as follows:

The B3 20v N/A'd engine may have the near exact bore, stroke and compression ratio as our beloved 12v (I even use the identical spark-plug in both).

2309 cc (20v displacement) divided by 5 cyl. times 6 cyl. = 2770.8 cc (which is near exactly our 12v's true displacement)."

Furthermore, as the coupe and sedan are pushing the same frontal shape through the air, I think that despite the body-profile, (gearing? I'll have to check) and engine displacement differences, if the 12v was working as efficiently, the fuel consumption of the two cars should be; very close to mirroring one-another.

My I5 Coupe mileage:

https://www.audiworld.com/forums/sho...php?p=18603387

Even in rough form, these are the best mileage numbers I've ever achieved with the 12v. I am confident these numbers will improve as I sharpen my pencil (done). I am also confident in these numbers as they don't include the distance for four small excursions (3 during highway-time, and 1 during city-time) for rest stops throughout.

That I am now within striking distance of the coupe's numbers, I am very happy.

There was nothing remarkable about how my car was configured:
- needs an alignment;
- old snow tires;
- A/C off, passenger side rear window down and sun-roof open.
Oh, and higher ground clearance than the slightly lowered coupe.

So, of course I think I can do better, but what strikes me is how ordinary the car is, and that this should be easily exceeded let alone achievable; by anyone else here reading this.

And (see VAP's remarks on I/M vacuum) this:

https://www.audiworld.com/forums/sho...hlight=mileage

Luxus Panzer

how did you go about this smoke test? I would like to do this myself

thanks.

z.

Lago Blue

I have come to think of smoke-testing as the other half of vacuum-testing, it's that helpful. Also like vacuum testing, it is fast, simple and cheap.

Similar to doing a vacuum test, you choose what components you want to include in your test; and then isolate them. The isolation helps in two ways, it traps the smoke; and limits the area under study. Instead of applying vacuum, you are going to apply slight pressure.

You probably really need a draft-free location indoors, or it would have to be a very calm day.

For instance, to test: the Air Filter-box Lid to MAF Housing seal, the MAF Sensor seal, the Flex-hose & it's seals and the Noise Damper, and it's to Throttle-Body seal; together:

- Remove the Noise-damper and it's Seal from the T/B; put a piece of thin plastic bag material over the T/B opening; and re-assemble it (note that the installed bag will now be helping this infamous seal, so if you haven't ever replaced it...);

The above installed plastic creates a sealed physical barrier between leaks that are simply un-metered; and those that are also vacuum leaks, including the EVAP system and ICS. Strictly speaking you may put barriers where-ever you like; or not!

- Open the Filter-box and seal the exit (leading to the MAF Housing) closed with some adhesive tape. I have VAP's velocity-stack installed, so some clear packing tape works very well here across the V-stack mouth;

- Detach the plug in your Noise-damper for the vacuum relief air for the EGR Freq. Valve (It's plugged because "smoking" is as bad for your EGR Freq. Valve; as it is for you! See EGR Mod. if required)

- Attach a length of hose to said port and blow in some cigarette smoke. You don't need a lot of pressure, just a few seconds for the smoke to spread out and fill the enclosed cavity you are testing. Clear packing tape allows you to see the smoke inside.

- Watch for where the smoke leaks out from!

Any leaks found are simple un-metered (by the MAF) air-leaks.

- Repeat as necessary.

For the time being, you can leave the T/B bag in place, and repeat the smoking on the I/M side of the T/B bag.

- Locate the vacuum hose connection to the I/M near the FPR. You have already done Vacuum leak-down tests on all the components that get fed vacuum from this point so no need to contaminate them with smoke.

- Attach a length of hose to said port and blow in some cigarette smoke.

Any leaks found are vacuum leaks.

In the ensuing excitement over the smoke-signal sources (or not!) don't forget to remove and install the required plugs and lines afterwards.

I have yet to try it on the crankcase, but fully expect it to be as helpful there.

Lago Blue

stock vehicles. Two cars, fraternal twins, separated at birth, divergent appetites.

The recent acquisition of a near mirror-image car (a '93 90q sedan) allowed me to exactly replicate a high-way drive of sufficient duration, which served to high-light the single significant difference between the two vehicles when doing a simple transit from point "A" to "B".

This previously loved (extensive maintenance history), yet recently unwanted and neglected (I installed used <i>matching</i> plugs before taking possession) "pearl" may prove to be a diamond in the rough. I admit to being seduced by it's great body condition, if not it's colour; and certainly not by it's fuel mileage.

My point here is that what we should consider as normal fuel consumption or perhaps typical for cars of this age is entirely unacceptable because my fairly original (and high accumulated mileage) one does so well. The unattended one does so poorly because of lack of attention, not any fault <i>or default</i> of the car itself.

The thirsty car starts well and suffers no drive-ability issues whatsoever; and I think this is <i>key</i> to why this issue receives such little attention. Without an immediate need to attend to it's defects, the car continues to be driven, and it's reduced mileage performance is deemed to be; <i>normal</i>.

However "The trouble with normal is, it always gets worse." (Bruce Cockburn)

Google maps plus a site such as this:

http://oee.nrcan.gc.ca/transportatio...ts.cfm?attr=16

even does some of the calculations for you, just from your receipts info.

Distance traveled, a 244 kms (151.6 mi.) journey, identical gas pump to gas pump, same direction, speed and conditions, three days apart:

<b>Car #1:</b>
Fuel Amount
17.4 litres
4.6 US gallons
3.83 Imperial gallons

Fuel Economy
7.13 litres per 100 km
32.98 miles per US gallon
39.61 miles per Imperial gallon

<b>Car #2:</b>
Fuel Amount
22.2 litres
5.87 US gallons
4.88 Imperial gallons

Fuel Economy
9.1 litres per 100 km
25.85 miles per US gallon
31.05 miles per Imperial gallon

The difference between these two car's consumption is a startling 27.5% penalty over the better performing vehicle. My simply cared-for car routinely buys me a free lunch.