After all the trials and tribulations by some of the Ljet crew and their struggles with cold starts, it came to me in an early morning haze this morning that maybe I should try bypassing the AAR valve and rigging some sort manual one. I figgered if inadequate air was creating the low idle at start situation, then maybe this would be a way of seeing that an increase in air would remedy that.

Ok, so I pulled the AAR, and set out to get what I needed. Well, here’s where a little laziness took over, I thought before I through that, may I’d see if a total bypass would work…

Click to view attachment

Here’s one of the AAR hoses going from the plenum directly to the intake boot.

So being diligent on startup as to not allow a potential crazy high idle, I gave it a shot.

Wouldn’t you know, while it was a bit balky to start, when it did the idle shot right up, and I shut it off. Knowing what to expect on the next attempt, it went up to just about 2k rpm, then crept up more. It was enough to convince me that inadequate air is getting into the engine with the stock AAR (which functions perfectly) and that a slightly larger one like VanB has installed will probably be the ticket.

Thanks to @Van B @StarBear and @Wonkipop for their quest for perfection

@Porschef

nice work.

yeah a straight through hose shot is bigger diam the the biggest aav would ever be.
she would easy hit 2K+.

fair bit of gas supplied by the CSV and also ECU enrichment on start up.
i reckon it lasts for about 30 seconds and then comes off.
so she will want to go if she gets a lungful.

ps @Van B is the perfectionist.
he is the one who cracked it.
i put up with mine but its nowhere near as cold in australia (north antarctica) most of the time as it is for folks in his and @StarBear 's neck of the woods. mine will usually idle steady after 30 seconds or 1 minute if i hold it steady at 1200 when i first start it.
lazy way to make up for a lazy aav. as it is i am still on a junkyard quest to find one i can sub in. have discovered late 80s and 90s SAABs had same identical looking aavs as the 914. likely different guts which might give BMW type outcome Van got. the aavs are out there to find if you snoop around. but its a matter of just how good are any of them as well given they are all 25+ years old as well.

Ha! Yup, I did the same bypass when I was messing around with mine and I thought it was gonna be a runaway situation lol. On the second attempt, I got the steady upward creep. 2k, 2.5, 3… and then said enough of that.

I also learned that the CSV still sprays even with a dead TTS, but only while cranking. To get the full 8-9 sec, you need a TTS… and that’s another simple but unobtanium part.

Thanks Gentlemen, yes, I did go back and read through a bunch of posts, and saw my idea was not original...but it did confirm what’s necessary. I’ve been using the foot method for years, and it hasn’t really been an issue; I’m used to it. But I figured if I had to sell the car, God forbid, having it be a true turnkey start would add more to the value than the cost of an AAR

@Wonkipop You’re correct, I could probably find a used unit at a wrecker. ‘Round these here parts BMW’s are mighty common, with lot’s getting wrapped around trees and whatnot, so it probably would be relatively possible. Not sure how I’d streamline the search however...

@Van B I apologize, I’ve tried a bunch of times to spell out your username so you’d get the notification but I’ve been unsuccessful. Great job ciphering this stuff, especially finding a replacement.

I gotta go back and see how @StarBear engine is doing


Now if I could find the test for the TTS, while I’m in there...

@Porschef

Click to view attachment Click to view attachment


as a side note porschef, mine starts fine. and will make itself run.
literally make itself run at about 700 rpm for about 30 seconds to a minute.
i just don't like it doing that. its not good for it. got to get the oil circulating and the engine heat in. which is why i prefer to force it to sit at 1200-1400 for at least a minute.
it will cold start and make itself run but i just don't think given its artheritis of the aav thats a good thing. i think anyone buying a 914 in otherwise good nick gets it.
i wouldn't worry. its like nursing a model T into action. part of the whole thing.
its not a lexus or a toyota camry.

but i do like the american obsession with having it all running right.
like Van B.
thats why you guys got to the moon and not the russians.

i'm just an aussie, we stand around and drink beer and bullshit.

Awesome, thanks Wonk

Mine will do the same, and doesn’t stall since installing the 123. Like you, I don’t wanna let it do that, so I keep my foot on it for 30 secs or so, till the head temp hits around 175°.

There ain’t nothing wrong with standing around bullshittin’, especially around a 914

Good stuff

Sorry for the late response as a bit distracted with getting ready for the New Hope Car Show. Shooting to better last year’s 3rd place result.

As for an update, Frogger is running great but still struggling a bit at startup (cold or warm) for 2-3 minutes. The dwell is a bit off at 50.5 degrees so that might (?) be a factor. The current AAR closes fine at 12V bench testing but maybe isn’t opening the whole way? Going to test that Standard BMW AAR and see how that works. Van is helping.

Otherwise, still confused a bit about the AAR function. At startup the mixture should be a bit richer, but the AAR adds air, which would lean the mixture (???). Also the AAR does not pull in fresh air, but pulls it in from the intake plenum, for which the air has already been taken in and metered. So, the AAR works more like an air recirculating unit, REDUCING the amount of fresh air at warmup, gradually closing and reducing the recirc air amount thus increasing air to injectors and thus leaning the mixture?
I now have several LJet guides; will check them out for detailed AAR functioning description. The general descriptions aren’t very helpful.

no, its not pulling air from the plenum. its allowing it into the plenum.
its pulling it via the AFM and its moving the flap.
its adding air. while its open.

that added air slowly goes away as it closes.

its more or less a clock while its doing that. just electrically heated. in the case of my car it may as well be a cuckoo clock you wind up. back when new timed to be synchronised with roughly the amount of time the engine takes to warm.

while its going tick tock, the ECU is adjusting enrichment to a pretty intact regime hardwired in that adjusts to the input that is by and large from the cylinder head temp sensor (and a bit from the air temp sensor in the AFM).

thats why it goes out of whack. the one thing in ze whole german prezishum komputer that behaves like a bavarian lunatic is the aav with its cuckoo clock technology. it just goes decrepit as it gets older. a bit like me.

all you are going to be doing is finding an aav that somehow behaves in the present time like the original one did 50 years ago. a bit of guesswork and trial and error. with an opening about the right size for the amount of fuel that the ECU is already supplying (which is a fixed amount with the throttle closed and TPS saying to ECU we are in idle circuit) and which gradually closes in roughly about the same amount of time the engine takes to warm up which is monitored by the cylinder head temp sensor telling the ECU this much fuel for this much engine temp. (again all the while the ECU knows its idling, the TPS tells it that).

you got to think of it as the engine will be getting this much fuel period.......during that cold start warm up phase.
and the air is supposed to be there.
cept its not if the aav has fallen over.

its actually sitting there running rich as hell, and barely turning over, but ze german computer keeps it running.....just. you want to get it out of that state of affairs as fast as possible. all zat fuel running down ze cylinder walls. so just rev it if the aav won't do its part. temp comes up. cht communicates. ECU stops enriching. and everything settles into normal daily life and gets on with it.

its not that different to a 914/6 hand throttle on the floor in the end.

van b just managed to fluke it with his BMW unit (or thats what he says).
personally i think he had access to military intelligence we ordinary folk are not privileged to see.

Hehehe. Either military intelligence or alien technology! We’ll see what happens when one is in my car. Yes, hoping for a return to “original normal” would be just fine.
Thanks for the AAR primer @wonkipop .
Now that makes sense that the air is flowing in the other direction. With the vacuum at the air intake horn I presumed the AAR line was in vacuum, too. Still learning…..

yeah. its the engine sucking air in. like it usually does.
and the aav is giving it the route to do it.
the fuel for that assumed amount of air is already down there at the injectors.

don't forget the TPS still has a role in L jet when the throttle plate is closed.
its telling the ECU that i am in idle mode.
combine that with the CHT telling the ECU i am cold (freezing cold) and you have the fuel routine.

now what i am not entirely clear on is that the AFM will definitely be pulled well open by that large air route down into the plenum. and normally that amount of AFM flap movement would be via the throttle body and throttle plate which will have switched the TPS into off mode. so i am never entirely clear about this but i assume that what happens is the TPS over-rides the AFM.

a real expert like @clayperrine would know exactly what is going on there in idle mode when the AFM flap is being activated a bit more than usual by the open AAV. which one overrules the other. ?

You can’t think of vacuum as a singular state. Pressures vary both above and below atmospheric pressure. Its the pressure differential and the tendency of systems to strive for equilibrium that determine which way things flow. The manifold is higher vacuum/lower pressure than the intake boot air. So, air wants to go from the boot into the manifold as long as the AAV allows it.

The more air through the AAV, the more the AFM flap moves, with more fuel through the injectors as a result. This loop will continue until the “system” stabilizes. Thats why @Porschef saw a super high RPM when he had a straight hose connection, and why a worn AAV only provides mediocre performance.

Oh and if aliens were real, and we had their technology, then I would expect these cutting edge technologies I am supposed get into operation to be much better than the over priced same old same ole… our stealth tech is little more than funny angles and paints that manage emissivity.

i need the angle measures and paint spec.

got to get under the radar so i don't cop another 51k-in a 50k zone-$200.00+ fine in the danny-dick-head-ville suburban operations theatre. i think i can figure out how to cloak the heat signature on the aav with unclassified documents.

Great Explanations...

Put the AAR back in the car yesterday so I could drive it briefly, and took a second look at the “adjustability” of it. Yes, there’s a minor increase in opening that can be had by loosening the nut on the back and pushing it to the extent of its travel and tightening.

Made a marginal difference...maybe. Engine obviously wants more air on startup. 2056 with a 9550 cam, slightly oversized throttle body, 912 injectors.

@Van B Out of curiosity, any chance you took a look down the replacement AAR to compare the orifice size to the stock unit? Such a fine balance.

Ain’t this fun?

Kind of like sitting around a campfire spinning tales of fantasy and horror. Marshmallows anyone?

@Porschef

I posted pictures of the two in my original thread that I linked a while back on @StarBear s thread. there should be a boresight photo in there.

There's no all-aspect protection. You need to assess where the cop is and then your approach angle. from there you can assess what features on your car will reflect back at the receiver. Then you have to either redirect the reflection away from the receiver or absorb the energy.

The above is a near impossible task for cars unless you want to get rid of your plates, turn signals, reflectors, windshield, and everything else that makes your car road legal... but then the cop can just pull you over for not having those things and write you several tickets instead of just one

So the real tactic for defeat is early warning capability. you need to see his system, before the system sees you. a good radar detector with GPS memory is a fantastic threat mapping tool.

it is -------sort of.

i did come across a thread years ago when i was recommisioning my car.
some buy with an alfa or maybe a bmw had got into tearing the aavs apart and seeing if he could rebuild them and get them functioning properly again. same kind of later style aav that is on our L jets. he seemed to succeed but he had a few aavs to practice on.
didn't manage to do it on the first go. since i have only my aav i'm not willing at this stage to tear into that.

i did pull one off a saab but it was fairly heavily corroded inside it so i didn't bother even doing a tear down on that one.

ah well. looks like i am f$5ked then.
going divine wind.
i know where to get a red dot head band from in a tourist shop in cairns north qld.

banzai.

if i'll be getting my license suspended and my car crushed for doing 35 mph in your failure to be independant imperial measurements, may as well go down in flames on the carrier deck.

(ps i just read a story that the b52s are going to be stationed in darwin and they are expanding the runways. might have to take a trip up north and watch them take off, never seen a b-52, what a plane. seen a f-117. but never a b52. things are getting interesting down here in the southern hemisphere. too old to join the airforce, but it sure is getting kind of curious and reassuring down in these parts).
keep up the good work on the paint specs etc. at least some of us care about it.

Back in the 80s, Carswell Air Force Base in Fort Worth (Now Carswell Joint Reserve Base) was a SAC base with B-52s on alert there 24x7. When I got my pilot's license, the field was due north of Carswell. We would be up flying and an alert would come on the radio, and we would have to vacate the airspace at maximum speed either east or west. The B-52s were scrambling and headed out at max speed in 2 second intervals. It was an an awesome site watching all the BUFFs headed out like someone kicked an anthill.

FYI... and the BUFF is the nickname for the B-52. Big Ugly Fat Fellow/F**ker. It is amazing that a military airplane that had it's first flight in 1952 is still operational in 2023. I read an article online a while back about how a new B-52 pilot was assigned to the same plane his Grandfather had flown when it was new.

Ok, fellow Ljet brethrens,

@Van B
@StarBear
@Wonkipop
@emerygt350
and anyone else...had some time to read through the extensive thread Van started back in Nov ‘21. Lots to see there, but I’m gonna try to remember the questions I had plus a couple comments.

It’s encouraging to see the effort and research there, I don’t know what happened but I left this thread at some point and never caught up. So, first comments-

Emery, that valve you rigged was very cool and even elegant. Great idea using that garden unit, had I seen it earlier it would have been the way to go instead of a quarter turn unit like I was gonna try till laziness took over and I just went for the straight bypass. It does confirm that there’s not enough air getting through at cold start.

The search for substitutes for both the AAR’s and TTS’s is great. Options are a beautiful thing.

The TTS is the only part of the system I haven’t checked yet. Would like to find one that sends the signal at 95° instead of 55°.

The metric system is great, especially if you can do close conversions in your head. But temps are better in Fahrenheit than Celsius IMHO, allow for more precision if you ask me.

Installing a 123 is the best thing I’ve done to make the car run better.

I’ve adjusted both the dynamic spring and static setting in the AFM to meet the needs of the 2056 as best as I can. Readings from the Innovate wide band shows 14.7 at idle and low 13’s at cruise. It’s very happy there. Hard acceleration drops down to low 12’s.

Ok, enough of my blabbering, here’s a couple questions...


Van, I know you replaced your AAR with the BMW unit. Was your original one fully functional? If not, do you feel a good unit would have made your cold starts the same as what they are now? I saw your pics with them open, it didn’t seem to me that they were drastically different. I also saw that you replaced some of the hoses that were improperly sized, do you think that would have made a difference with the original AAR?

Emery, I watched your cold start and warm running videos this morning. Engine sounds great. That’s how my Djet started and ran when I first got the car until the engine dropped a seat the following year. Been running the Ljet since. Are you still using that garden valve as part of your warm up procedure?

Wonk, have you changed your’s? Loved the couple SAABs I had, the 900 turbo was pretty bulletproof, and the 9000Aero was the best road car I’ve ever had. A six hour trip felt like 90 minutes...


Ok, I think that’s it for now. I know these are silly little problems in the grand scheme of life and I’ll probably have a couple more, but they do provide some welcome distractions from the 600 pound gorilla in the room or however that saying goes...

Thanks all

Joe

@Porschef my original tested good. It heated up, I cleaned it and everything moved from full cold in the freezer to heated while connected to 12v. Looking back, I never confirmed the actual amps sent through the harness. So, when connecting to a battery for testing, it’s possible I was giving it more current than it would normally receive.

That said, I think that 50yrs of heat cycles had fatigued the spring steel and probably the heating element… it was tired and needed to be retired.

There was a considerable amount of time where I was convinced that my AAV was fine. But changing the part was night and day. But the journey was good, because I found a lot of DAPO items that needed to be addressed… like spark plugs that didn’t seat lol.

BTW, your AFRs are perfect!

Thanks, it does seem happy there. And no pinging since the 123 install.


I only ask to verify the functionality of your original AAR because I’m somewhat reluctant to pull the trigger on the BMW unit unless it’s virtually certain to remedy this last little bugaboo. But seeing that you also did the same bypass (and here I was thinking it was a original idea of mine ,not so much) and got the same results does make me wonder.

Anyway I’ll look into it Monday. Thanks

Got it

Thanks for the summary @porschef .
Based on your numbers, looks like I might still be running a tad rich at rpms.
Will do another little experiment and temporarily install a borrowed larger BMW AAV and report back.

Great! Looking forward to the results

i wouldn't classify that as a hijack or off topic. but not an administrator or mod.

strikes me the bee five two ( auslish) is just like a 914 (nine one four).
rebuild rebuild rebuild. keep the air frame going.
because its.............just so good?
not that i would really know, but it must be or the USAF would not bother?

classic. also the 914 was once considered ugly. (still is in some circles?).
so the car and the plane have a lot in common.

huge fan base as well?.

Rock Auto 1988 BMW M5 idle air control valve $215.79

Thanks, yes, I saw that, I’m gonna wait to see how Steve’s swap goes.



Oh, the suspense...

I second that.

down here in north antarctica we have this bloke here.
he is up north in queensland.


my mechanic mike found him.
but he doesn't really sell new/replacement. he is an exchange reconditioner.
he wants mne on exchange basis.
which i am slightly reluctant to do.
but what am happy for him to do is sort out one i can find on something else.
ie a SAAB or BMW.
apparently he has the original airflow and time duration specs for all the different models.
so theoretically he might be able to adjust one from another model to suit the 914.

https://k-jet.biz/auxiliary-air-valves/

Very interesting Wonk, that’s the site I used to pick off both the Bosch and BMW part numbers…

Just waiting for the results from StarBear before the next step

Thanks

apparently he is a genius.
working away in his little shop.
i have no doubts as to his credentials.
but i just don't want to be left AAV-less if some kind of accident happens.
like lost in the mail.

so i will only be sending him a sacrifice he can cut the head off and eat the flesh of ancestors (which happens the closer you get to papua new guinea! )

Here are the results of the AAR temp tests - OEM Bosch/VW 101 vs NOS Bosch/BMW 125 units:
Click to view attachment
The BMW unit starts with a much bigger opening (expected). When 12V source applied, while the 101 unit is closed completely at about 2 minutes the BMW unit is maybe 80% closed. In 20 minutes the BMW unit was still just a tad open, but that can be adjusted by the little screw on the backside (I didn't, as it's a borrowed unit).

Videos of warm up with Bosch 101 unit coming soon in another (or a few) post). Videos and results of Bosch BMW unit after I swap it in and fire it up.

Results of current AAR (Bosch/VW #101) unit testing during warmup provided below.
Air temp 80F, 65% humidity, in the shade.
Dwell 48 degrees; Timing a bit advanced with red line about 1/2" to the right of the V notch.

@15 seconds - ~600 rpm; idle not stable
@1 minutes - ~600-650 rpm; idle not stable
@2 minutes - 700-750 rpm; idle not stable
@3 minutes - 800-850 rpm; idle not stable
@4 minutes - 900 rpm; idle stable
@6 minutes - 950 rpm; idle stable
@7 minutes - 950-1000 rpm; idle not stable
@9 minues - 1150 rpm; idle stable.

Otherwise, I just let it sit for a minute in the driveway, then with stumbling idle I back into the street, and take off happy as a clam. Idles fine and stable by the time I get to the end of the block.

A great 30 minute run this A.M. with cassette of Doobie Brothers and Jefferson Airplane complementing the air-cooled engine sound perfectly.

Have tried to post videos of the engine analyzer but dont have software to edit and reduce file size (or site doesn't load .mov files)

Will install the borrowed BMW unit in later today and post results. Stay tuned (pun intended)!

A little Sunday afternoon suspense...that’s pretty much what I got going on from a cold start. Read through a couple old BMW forum threads that showed a small adjustment to the AAR making a difference. The pics look encouraging.

I’ll be here waiting

Ok; second test, with bigger BMW AAR. What a difference!!!! Here’s the data:
Temp 84F; 45% humidity. In shade.
Started right up with authority. Way up to 1500 rpm. Turned TB idle screw down to 1000 rpm.
15 secs - 1200 rpm idle stable
1 min - 1000 rpm. Idle mostly stable; a bit of swinging.
2 min - 925 rpm. Idle stable.
3 min - 920 rpm. Idle stable. Opened TB screw 1/4 turn
4 min - 920 rpm. Idle stable. Opened TB screw another 1/4 turn
5 min - 940 rpm. Idle stable
6 min - 920 rpm. Idle stable. Opened TB screw another 1/4 turn
7 min - 880 rpm. Idle stable. Opened TB screw 1/2 turn
8 min - 900 rpm. Idle stable.
Opened TB screw to set at 1000 rpm.

Will let it cool a while then do a third test with the initial TB screw set at desired final idle rpm.

One thought - maybe this BoschBMW AAR is better aligned with my new Bosch AFM, while original AAR is better aligned with my original (failed temp sensor) AFM?

Stay tuned for the thrilling concluding episode!

@StarBear , you gotta set the TB air screw when the engine is at full operation temp ( after driving) and then leave it. Otherwise it just remains a variable. That's how I discovered that my TB is worn with the infamous groove. Once I left the adjustment screw alone, I knew the slight variation in idle had to be coming from a change in metered air flow.

Very promising Steve!!

Couple things, and I could always be wrong... Your ultimate idle speed shouldn’t be affected, so I’d think that closing down the AAR opening with the nut would be the ticket to bringing down the initial start RPM’s . I understand why you wouldn’t want to as it’s a unit you’ve borrowed. If it’s remaining open just a hair that’d bring the idle up, but also let in unmetered air that you don’t want.

Second, I don’t think your AFM would know the difference, it’s kinda like a few components working as one. But yeah, looks really good!

Thanks for all the info


*Edited* What Van said!

Yep; will do the third test then a short drive to confirm all and do final TB screw setting.

Great ideas. Will order one if third test and drive check out, then do bench tests once it arrives and adjust the nut accordingly.
In the home stretch!!

Ok, Test 3. New Bosch /BMW larger AAR with initial TB screw adjustment.
Same temp and humidity.
15 sec - 1400 rpm. Stable idle
1 min - 1200-1500 rpm. Idle wavered a bit.
2 min - 980 rpm. Stable idle
3 min - 940 rpm. Smooth
4 min - 920 rpm.
4.5 min - 880 rpm. Opened TB screw 1/4 turn
5 min - 920 rpm. Stable idle.
6 min - same
7 min - same.
Adjusted TB screw to 1000 rpm.

Test drive: AWESOME! Flawless and great acceleration. Very stable idle at 1000 rpm. Don’t think it’s run this well in several decades!
A bit over 2 years to get through AFM temp sensor fail, replace and dial in new alternate AFM, install Air/Fuel sample port and adjust AFM, and polish it off with resolving startup idle issue with alternate AAR.

Thanks to all for your support, wisdom, and patience. Happy Motoring!!

In the winter it will obviously stay at the 1200-1500 range longer. But overall, that definitely looks like what you want to see!

Awesome Steve! Thanks to you and Van for the info and testing

Thanks for sharing all your research and testing!

nice work @StarBear .

inspired me.
soon as its springtime here i am off for a wander at pick-a-part junkyard.
see what kind of aav i can find in the relics.

Ordered the BMW AAR from Rock Auto last Tuesday and it arrived Thursday. Pretty quick. And yes, it’s a Bosch unit of French manufacture. Installed it this morning, I had to dial down the orifice some to cut back on the 1800 rpm cold start. A bit of trial and error but I think it’s good now, about 1200, with no touching the throttle. I’ll let it sit overnight and check it in the morning. Seems pretty good so far

Mine arrived Saturday, also from Roch Auto. Some things to do around the house and test new (also NOS Bosch) unit with pics as before to compare with borrowed unit.
OT: Long and hot day at the New Hope Car Show, but worth it:
Click to view attachment
Click to view attachment

Congratulations Steve

Car looks fantastic!