heinrich

Glenn I don't know, possibly. It's about turbulence not flow. Brett Hi Buddy ..... .... walk up to a chain link fence. Nkay? Now look at what's on the other side of it ... see it? OK now put your cheek against the fence and look at a 5 degree angle almost toward the other end of the fence. Still see what's on the other side?

Mike Simard

Deleted, erronous assumption about MAP sensors, LH Jetronic doesn't use one. sorry.

Brett928S2

Hi Heinrich

The point is was trying to make lol....was if the mesh in the K&N can affect the airflow...what about the much smaller mesh in the MAF screens ???
Surely they would affect it even more....as there are 2 of them and smaller mesh ...


All the best Brett

PORSCHE 928 S2 AUTO V8 4.7 LTR 1986 IRIS METALLIC BLUE WITH X-PIPES ,RMB ,PROMAX CHIPS, KICKDOWN SWITCH.K&N.NITROUS KIT 150 HP.
PCGB UK 928 FORUM MODERATOR - Email managingdirector@pchealthcare.co.uk

heinrich

Well we don't know but my theory is that the air hits the MAF screens straight on. But they hit the KN screens at a very sharp angle, thereby reducing the path to a crooked one, creating turbulence..

PorKen

The factory did.



16V US engines have a vestigal cast hole in the heads (square hole between center intake ports) which was the exhaust crossover for the carb'ed intake manifold!

worf928

Thanks for saying that I was 'simply confused.' The way I see it I was 'complexly confused'

heinrich

Yeah .. there is nothing simple about our good friend Dave

the flyin' scotsman

Dave.............just read your long thread from a couple of years ago re. your 'gutless GT'. Did you derive that the 'hall sensor' was the issue because you had performed all other tasks? I don't see what pointed you in that direction?

Just had the car out...........no chirp from full throttle in 1st or 1st to 2nd at WOT.

marton

the enclosed graph is supposed to show the performance effect of MAF ageing. It is not especially accurate as MAF ageing is a function of switch on/switch offs

Marton

worf928

Malcolm, the post that provided the 'last straw' on my decision to R&R the hall sensor was post #7 from "Scott M" with whom I've been acquainted since the old PorscheFans days. You'll notice from the title of the thread that I already suspected the HS as a likely candidate. I had, prior to that thread, already exchanged thread posts or e-mails (cannot remember now) with John Speake on HS failure modes he had observed in the past.

The original purpose of the thread was to poll for any other cases where known HS failure resulted in a similar or identical failure mode. When I received a second confirmation from a trusted source (Scott) and given what I already knew about what I had personally R&R'd on the car, I decided that the HS was the most likely candidate. However, the thread, after post #7 provides a lot of anecdotes on other failure modes that are relevant to this thread's subject.

From the final few pages you'll see that the HS was the primary culprit for my "Powerless GT" but that a dyno test confirmed that my '89 was still down by 25 to 30 hp as compared to other '89 GTs. (IMO, 295 to 300 rwhp is the benchmark for a perfectly running _stock_ (except for RMB) '89 GT.) I have not yet done anything to 'find' that missing hp because 1) I decided that other ailments of the '89 and the suspension upgrade had higher priority at that time and 2) all signs pointed to that missing 30 hp being deeper in the motor - as in head damage, or low compression. Having been in and out of the 'shallow' parts of that motor three times in the previous year I was ready for something else anyway. My next steps on the '89 are to do a leak-down test and possible a compression test.

My diagnosis of the HS - aided by John Speake and Scott - was as "easy" as described above because I had an near-absolute baseline (except for leak down / compression numbers) for the maintenance of the motor as I had over the previous year fixed or tested-good all the other possible culprits (IIRC. Careful reading of the thread might find one I missed.) But, the thread shows just how elusive a HS sensor failure can be without such a known baseline. Flappy? LH Brain? Fuel pressure? Vacuum? False air? Wiring? Injectors? Plugs? Wires? Coils? MAS? Cats? Exhaust?

Since that thread I have found yet another problem with the '89: The Hammer diagnostic port is as dead as a door nail. The Hammer, PWIS, and PWIS II (The latter two being Porsche's newest replacements for the Hammer IIRC.) cannot find any of the ECUs or test any of the actuators. So, the Hammer - like on an '87 or '88 - would not have helped find this failure.

So, we wind up back at *my* methodology for the HS: Replace it when you have easy access to it. Else, and especially if you do not have a maintenance baseline, or a diag port and easy access to a Hammer, when it fails it becomes a real hail-puller-outer of a problem.

j.kenzie@sbcglobal.net

I know I've only got L-jet, but I noticed a huge improvement when I changed my AFM. I had a flat spot at about 2000RPM that went away instantly. I had read that the sweep arm contact of the AFM can cause wear patterns that interfere with predictable response. After checking my other sensors and a new O2 sensor, I took a stab at it and had a good result. Got a rebuilt unit from 928 Specialists; thanks.

Louie928

Ok. I'm not too eager to get into an air filter discussion that's a lot like religion, but I see it this way. The stock air filter/MAF/throttle/intake tubes, etc., (in decent shape) all have the capability to flow more air than a close to stock 928 engine can normally use. The fact that a K&N will show more air flow in a plastic box on an auto parts store counter doesn't mean anything, or that someone's uncle used one on a lawn mower and it did work better doesn't mean anything as it relates to the 928. The K&N has holes that would make it ineffective as an insect deterrent window screen and only a small percentage of the area of a 928 micro fiber filter. Hardly a viable choice for a filter on a long life engine.

The one design flaw in the 928 intake system is that the air enters the ends of the air filter housing and has to go through the filter, travel to the center, make an abrupt 90 deg turn and immediately enter the MAF. It doesn't really like to do this and much of it travels horizontally across the top of the filter and goes pretty much straight down near the MAF opening. Notice how it's the center of the filter that gets dirty first, not the ends by the inlets. The MAF likes undisturbed moving air to produce an accurate measurement. The MAF doesn't particularly care which way the air flows across the wire. Downward, upward, sideways, will cool the wire and produce an output. It's only the air mass that gets into the combustion chamber that counts in an engine. When turbulant air goes through the MAF, it will produce an unpredicatable, erroneous, and widely variable output. How will the LH process the variable, fluttering, signal? Who knows? I'm sure it has some electronic filtering on the input to average out the variations. The electronic filter would be optimised for a stock filter, not a K&N. I'm pretty sure the LH mostly is confused by the variable air flow and outputs an unstable injector pulse width. Whether it ends up being rich or lean is anyones guess, and is probably a little of both. Quite a few things in the 928 intake contribute to turbulent air entering the MAF. The close proximity of the MAF to the air box and 90 deg turn for some of the air to enter is one. The close proximity to the air filter element is another. The air source at the ends of the box doesn't help either. Anything you can do to get as undisturbed air as possible into the MAF will help. The coarseness of the K&N pleats contributes to turbulent air entering the MAF. Remember that the air is somewhat going through the filter element sideways other than in a spot directly over the MAF. Like H said, look through a very open chain link fence sideways as an example. I've done some experiments with air flow in the airbox. I tested on my GT, Heinrichs S4, and a few others. Mike Schmidt made a nice example for his S4. I made a raised section around the MAF opening with a larger radius. Like a toilet seat. I called it the Einholer. It looks similar to the air box bottom on an '85. I also made a spacer around the bottom section of the air box to raise the air filter element to give more space between the air filter and air box bottom. The Einholer opening netted about 5-6 lb ft of torque below 4500 and the raised air filter helped above 4500 to give 5 -6 hp on top. Note that in the K&N vs stock filter dyno charts I posted yesterday that the K&N was down all across the range. Even more so on the low & mid range than at the top. Note that the dyno lines with the K&N are more wiggly than with the stock filter. With the Einholer and filter spacer, the dyno lines are smoother. I'm not sure the engine actually used more air with the air box changes, but I'm pretty certain that the MAF measured the air more consistently and accurately and the LH outputted a more stable injection pulse.

Now, could you tune with a K&N to get good power and then show less with a stock filter? Probably. You could cram a potato up the exhaust and tune for that too.

There is a place for filters like the K&N. Extreme performance where you don't want the filter to clog with dust. Vehicles with inadequate stock filter capacity systems. Installations where there is some distance between the filter and the MAF should work ok. None of this applies to normal street driven 928s.

Here is another set of dyno runs I did comparing the stock filter with the K&N. I believe this is when I was testing the Einholer air box MAF hole. I was still trying to get decent test results from using a K&N filter since at the time I still believed the hype. The air filter would then be about 5/8" closer to the bottom of the air box around the entrance to the MAF than a stock air box. It exaggerates the effect of the turbulance near the filter and MAF inlet. Both show about the same numbers. Which would you rather drive?
https://rennlist.com/forums/attachme...d=141154&stc=1
K&N_Stock.jpg

worf928

The (theoretical) advantage of the "H" hot-wire sensor over the flap-type AFC systems is that the H is supposed to actually measure the mass of the air. This measurement would be therefore be independent of changing density due to temperature or pressure (elevation.)

So, if that is true then the measured mass of air should not be a direct function of the absolute rotation of the throttle plate. It would however be an indirect function of the mass and density assuming that nothing else is different.

This is - I think - quite possibly the thing that 'is different' with a K&N. The algorithm that is used to derive mass from the hot-wire (IIRC it's current required to maintain a specific temperature? (Probst not handy...)) must make implicit assumptions about the flow characteristics in the vicinity of the wire. One might then assume that a MAS could be calibrated for the flow characteristics of a K&N filter?

(Note the question marks above John )

I'm sure others have seen more stock 87+'s on a dyno than I have (Louie has I'm totally sure.) The 20 or so that I've seen all ran richer than the ideal 12.6 A/F ratio for max power. Lower 12s, mid-11s and sometimes high 10s. So, if the theory of the K&N upsetting MAS calibration is correct (in the 'right' direction) and if the theory that the K&N flows more air is correct, then mid-11s become low 11s and high 10s become mid-10s and there is power loss.

worf928

Louie beat me to the punch and has the data to prove it!

ew928

In a previous discussion about the damn K&N filter with Dave C., it was noted that the K&N is a flat filter whereas the stock paper element is a raised filter. (We were trying to decipher whether the arrow points up or down on the paper filter)
The conclusion was to install the paper element with it being raised from the MAF opening. Much like a mini version of the Einholer.

It's also disconcerting to look through my ex-K&N at a bright light source and be able to see through the filter. Can't do that with the paper element filter.


Ernest (NYC)
K&N free for over 2 months.