As getting ready for the 2022 Northeast Gathering, I'm encountering a semi-recurring idle issue on my 1974 1.8L LJet FI. When outside air is cool and dry, idles just fine at 950-1000 rpm. When outside air is warm and humid, after warmup the idle falls off to barely running. I've just been adjusting the air flow screw (maybe 1.5-2 full turns) to open it. Of course, the reverse happens - if I preadjust the screw it idles near 1800 then settles down to 1100 after the engine gets totally warm (15 mins or so).
I presume (?) there SHOULD be some regulation, perhaps with the ECU or AFM, to monitor air temp and humidity and make adjustments automatically? I didn't have to manually do this adjustment in the past.
The AAR works and tests out fine. Just annoying to have to pull over after running a while and adjust the screw.
Your thoughts @Van @Van_B and @Wonkipop ?

CHT Sensor? Rpm seems to stay steady if air stays cool and dry, though (not often here in NJ this time of year).

only thing i can think of is the other temp sensor.
temp sensor 2 i think it is called in factory manuals.
this is inbuilt into the AFM unit.
i don't think you can do anything about that one if it starts to play up.

that sensor measures incoming air temp.

its possible with high humidity perhaps the electrical connections are affected by moisture? i seem to remember that living in chicago it got monstrously humid unlike anything i experience in australia. perhaps try cleaning the electrical plug connection to the AFM with electrical cleaning fluid. also the plug to the ECU. nothing to lose. check all the engine grounds for being clean and corrosion free. thinking you had it standing stored for your horrible winter and you have only set the car free recently?

@Van B might have better ideas.

Will try cleaning those plugs. Grounds are A+ and it doesn’t sit for more than a few weeks at a time during the winter. Will explore that temp2 sensor. Thanks, @wonkipop !

Steve, sorry for the delay, I’ve been traveling for work and forced to do things like socializing after hours lol…
Anyway, next time it’s all hot and steamy, you should grab hold of the throttle body and try to force the butterfly closed. If the idle drops back to where it’s supposed to be then I think you may have a worn TB that’s not seating and reveals itself when the metal heats and expands just enough to not seat properly.

On these cars idle is far more likely to climb from metered air than any fueling change. When you mess with the AFM flapper while the car is running you can better understand how small the fuel window is for idle. A bit too little fuel, idle drops, a bit too much, idle drops.

oh thats a very good suggestion.

I’m actually having the reverse problem - when hot and humid the idle drops. Sort of as if the less dense air and added moisture yields less O2. I checked the LJet guide and 74s don’t have the air temp sensor in the AFM.

Check it all the same.

But, it seems to me that you need to set idle for hot and humid.

With my new AAR, the heated wire alone is not enough to completely close the AAR in cold conditions. The cold engine bay sucks the heat right out of the aluminum housing. But after I drive a while, everything is warm enough that the AAR closes up completely.

I’m thinking you had your idle set for a season and engine bay temp when the AAR was open just a fraction.

thats interesting about the sensor I in the AFC.
what L jet guide is that steve?

i have the service guide to the 74 model i downloaded from mr b's website - guide says its there. and it tells you how to check it - adding that it cannot be replaced.

factory workshop manual also says its there in the EFI section.

i mistakenly called it sensor 2 in posts above. sensor 2 is the cht.

hmmmmm?


not a bad theory of Van's that the AAV is playing up.
bloody things. a bit like a half century old pop up toaster when you think about it?

@Van B and @wonkipop Will check them out. I have the LJet 912E guide from mr B, which says it’s there but I have a note that it for later models. Will check again! Wonder What one does if it can’t be replaced - a work around or just reset the idle screw and live with it?

The Porsche manual does a terrible job of breaking out year to year changes. I know the 75 has a 7-pin and the 74 a 6-pin. And the manual only says to test resistance between two sets of pins for a pass fail.
Either way, I still think there's no issue with your car other than seasonal allergies!

Hehehehe. To my best (?) reading, the D-Jet and 912E LJet have an air intake temp sensor but the 1.8 doesn't as it only has 6 pins - until maybe later with the 7-pin.
@wonkipop I checked Mr. B's website and can't find your reference; there are a few LJet manuals but none seem specific to the 1.8. My guess is that the 1.8 has Temp Sensor II (CHT sensor) but not the intake one?
Meanwhile, will try to figure out testing the butterfly as Van suggested.
Have cleaned the AFM and ECU contacts with electrical cleaner spray followed by air blast.

Ok, looking at Haynes page 177 on fig 9.87 current flow 1974 USA part 3: Shows G19 as air flow meter but no temp sensor. Temp 2 CHT shows as G18.
Going to 1975 on page 185 it shows G19 as afm, G17 as temp sensor 1 and G18 as temp sensor II.
No temp sensor I in our 1.8s that I can determine. Will head to butterfly test.
Also found the oil fill cap a bit loose, so that does funny things, too. Tight now.

you could well be right as the plugs are different and i have always wondered about that.
and thats a good explanation for that plug difference.

the stuff from mr. b' site is here.

https://bowlsby.net/914/Classic/zMan_1974_1...iningManual.pdf

its in factory tech manuals section.
it is a service training manual for the introduction of the 1974 models and covers the 1.8.
however it does not mean it is correct.

i will now look into this as you are on to something.
possibly a good thing its not there. one less thing to screw up.
entirely possible they held off until 75 on the temp sensor even if they intended it to be there in 74 just to keep it simple. would just mean its slightly less refined in terms of mixture fine tuning and wasn't needed for emissions until 75 when the standards tightened?

good stuff pointing this out. the mysteries of the first L jets hey.

oh yeah steve. a loose oil cap will make your idle stumble and die, even a tiny bit loose.

Van; could you explain just a bit how to do this? I get the concept but not the action. Thanks!!!

When the car is idling just grad the arm where the cable attaches and try to twist it further closed. Sometimes you won’t really feel movement but your idle will drop. That’s how I learned my TB was worn out.

Got it. Thanks. Will test with oil filler cap now tighter first, then if still happens will do this test.

@StarBear

i looked through the electrical schematics in the factory manual.
additional current flow diagram AFC engine.
you are absolutely right.
only list temp sensor II for 74.
temp sensor I and II are listed for 75.

thats a relief.
74 1.8 = a classic mate. the old keep it simple stupid principle and one less thing to go the way of a toaster.

ok @StarBear .

i had to keep digging. as i do.
i think we do have a temp sensor I on the 74.
but its different to the 75.

i've never been inside my AFM.
so i'm going blind here off documentation and diagrams.

but as far as i can guess or surmise, its somehow inbuilt in our 74 AFM and it appears to have been separated out in the 75 units. perhaps in order that it can be replaced.
i think it is integrated into the potentiometer section of the AFM in 74 but it might have been shifted to another location in the 75 unit, more directly in the airstream.
i have never seen a 75 unit, and being down here, i never will.

but here are the electrical diagrams from factory manual.
this would imply its not there in 74.
but not necessarily. it might be that its just considered an integrated part of the AFM potentiometer unit.

Click to view attachment
Click to view attachment

this is the 75 wiring diagram.
clearly says its there as a separate thing.
it has a dedicated pin too. 27.

Click to view attachment
Click to view attachment


74 AFM. 6 pins. no dedicated pin.

Click to view attachment

75 AFM. 7 pins. 27 is a dedicated pin according to wiring diagram. hence additional pin.

Click to view attachment


diagram page from factory workshop manual.
seems to imply the sensor I is there in 74 and is integrated part of potentiometer.
but is not in airstream.
of course it is a diagram. and i have never popped the lid to go in there and see.

Click to view attachment

so.......maybe that training manual i downloaded from mr. b's site is not incorrect.
here is the page in question.
interestingly it tells you to test it using pins 6 and 9.

Click to view attachment

so i am thinking it is there.
but if it screws up in the 74 there is not much you can do about it.

(i leave this with @Van B - i think he has opened the lid on his to look inside).

Yes, though that extra control would be useful. Makes our 74 1.8s just that one more aspect unique!

I’ve had mine open in the past and don’t recall seeing anything but the pins and tracks.

I've got one AFM that I did a full tuning session on and there's nothing like depicted up top.

Also, still think there's nothing really wrong with StarBear's car other than the fact he lives in NJ!

Hehehe. Yeah, but NJ is looking a lot better than some other states lately! Born in Ohio and lived in WI and gre up in VA, but been here since 1974 and married a Jersey girl, so guess I’m “native” by now.

Yeah, you’re a native for sure. Did you buy the car in NJ?

------

i agree re steve's car.


--------
thanks for the image of the afm. cool. saves me murdering the pterodactyl perched on top of my engine.

i think i can see the two wires that go to it in yours.

here are a couple of good clear pics i dug up of an AFM off a BWM K100 forum.

an early one.

Click to view attachment

a later one.

Click to view attachment

i believe it is the thing in bottom right of AFM in images with a couple of wires to it.
i think it just gets wired in differently in the early AFM if indeed i am right about it being there. it looks like its replaceable in these later AFM. undo two screws. but its probably fused in to the unit in someway in ours and can't be replaced easily?

the drawing in the workshop manual is just an oktoberfest cartoon since tampering with the moonshot tech was VERBOTTEN in 1974.

That’s a differ AFM than both of mine.
The one I tuned on was a NOS Bosch reman that was meant for a bus, so it was tuned way too lean for the 914… I guess the Porsche heads do make a difference.

The other is the part that came on my car with a FI Corp rebuild sticker on it. I haven’t messed with that one.

Regardless, both have the same guts and neither look like that.

I’ll have another look though.

yeah its going to look different to these.
these are 10 years later i think.
but pretty sure thats the temp sensor with the two phillips head screws and the two wires coming out in these later ones - and it looks like that so that it can be serviced and replaced.

in photo of yours it looks like there is something with two wires coming out of it and heading towards the interior side of the pin plug section in the same location. the wires look a lot flatter and "integrated" in yours. as if its more "built in" it will be in that spot if its there, it just won't have the ability to be removed.

i'm going to take the BMW photos and your photo to mike when i see him next.
he is going to know for sure.

Click to view attachment

i'm just curious about it @Van B .

its not the problem with Steve's car.
i've read up on what happens if the temp sensor is kaput.
and the symptoms are a lot worse than a high idle.

i wouldn't even mess with an AFM.
there is a bloke here you send them to who fixes them and knows what he is doing.
i'm tempted to contact him and ask him about the AFM on our cars and see if he knows anything about this and whether and how there is a way to service this sensor on our cars if its not easily removable. i'm betting there is.

the bit of history i have read about these things when they first came into being was that they were classified as part of the emissions system and by regulation had to be tamper proof. if there was a problem with one VW removed them from the car and just replaced the unit completely. they didn't throw them away, they were sent back to the bosch factory in germany and i'm guessing, rebuilt there. using factory facilities and manufacture techniques. so a built in temp sensor would have been a little easier for the factory itself to deal with and replace if it was faulty.

i think as the technology was taken up and the numbers got big they decentralised servicing and repair of these items and made them easier to dismantle and replace defective parts inside them.

effectively our cars and 412s are numero uno with this tech. so we probably have the version that no one is supposed to touch except the bosch factory back in germany.
hence a version that was never intended to be looked at except by factory techs.

Nope; the day after graduating college in Va. Drove to Ohio two days later for a friend’s wedding then to NJ the after getting back from OH. Broke the car in quickly!

did one xmas in NJ. in a pretty flash area. Bernardsville.
girlfriend's father lived there.

remember getting up early on xmas morning and going for a walk in the snow.
novelty value of white xmas for aussie. lots of deer wandering around.
looked like they would do more damage to your car than a kangaroo if you hit them.

also drove down to trenton once when staying in new york to check out some famous bathhouses that had been designed by the architect Louis I Kahn. the bathhouses were wrecked and in ruin.

been to cleveland OH. tough town. or it was back then in the late 80s.

never got to VA. always regretted that i never got a look at that section of the east south of new york.

was actually planning a trip to the states a couple of years ago to catch up with some old friends in LA and Chicago and was intending to finally have a look at that bit of america.
covid put paid to that. probably won't make it back there now unless i win the lottery.
or maybe sell the 914. and that ain't gonna happen if i can ever help it.

its cold down here today.
almost new jersey or maryland in winter - minus the snow.
so i am by the blow heater.

i've figured it out.
since we got the real early proto L jet.

the temp sensor is in there.
it modifies the signal the AFM sends out and does it inside the AFM before going out the door.
from there it is just one output sent to the ECU which tells it where the flap is (how much air volume) and modified for air temp (density) in the singular.

the temp sensor is absolutely necessary to L-Jet as its got to know what the density is.

the first buses with L jet (ED engines) got the same thing.
so did the first opel mantas.

neither had pin 27 on the afm plug or used it on the ECU.

then in 75 they did it differently and send two separate signals to the ECU.
one was pure position of flap info.
the other was pure air temp signal info.
the mixing got done in the ECU.

the 7th pin (27) was for that pure temp signal output from sensor I.

they must have decided it was a better way to go in 75.
possibly made testing the AFM easier as you could work out which one was screwing up, the flap and potentiometer or the temp sensor.
whereas in ours they pollute each other before it gets to the pins.
thats why there is that one simple dumb test for the AFM output for ours.
using pins 6 and 9 i think. but the test cannot distinguish it can only tell you that the cocktail signal is right or wrong, but not which bit of the cocktail is the screw up.

there was one other factor with air density and that was atmospheric pressure due to elevation above sea level. for that they could fit an altitude sensor. suspect very few cars got that. maybe if you lived way up in the rockies somewhere? denver ski resorts whatever. that was a gizmo that got screwed in somewhere down near the double relays as far as i could work out.

off to do a cold start test now and maybe drive.
it might not get this cold again and i can just about simulate north america.

@wonkipop Yep, we’re just a few miles north of Bernardsville and yes, hitting a deer (or vice versa) will do A LOT of damage. Know of quite a few cars (and deer) totaled when they met in time and space.

A key question with the temp sensor would also be HOW it adjusts things. If there’s a base temp (68F, say) then as air temp increases and density decreases the air for combustion decreases. If the system doesn’t adjust and keeps sending the same fuel there would be a rich condition, dropping the idle speed, which is what I’m experiencing. Just enough for idle to stumble but it quite cut out. A bit of throttle at stops keeps it going until I can pull over to adjust the idle screw more open, about 1-2 full turns.

At that setting when the air is cool again and I go to drive it the next time the idle is 1600-1800, so have to turn down the idle screw until the air is warm again and repeat.

It happened last summer on the way back from the NE Gathering but not in the late Fall, Winter or early Spring. Just started happening again a few weeks ago as warm Summer air invaded. Hence my suspicions about an air temp sensor issue.

Who needs those mind exercise games at my age (70) when I have a 914?

@clayperrine Would you mind chiming in on this with your LJet (74) knowledge? Thanks!

Update: Had a nice 30+ minute run today including highway. Air temp in mid 70s; a bit humid but not yet oppressive.
Car started at 1000 rpm then by 2-3 blocks rose to 1600 (my new seasonal initial preset). After 10+ min idle had dropped but only down to the desired 1000 rpm. Stayed in that mode until after highway and full time.
So, a) tightening the oil filler cap seems to have helped things from being so extreme and b) can live with a minor seasonal manual adjustments until a more permanent problem identification and solution.

In reference to the air flow meter, the air temp sensor is in both the 74 and the 75, but the wiring is different. In 74, it shared a common wire with the wiper contacts. In 75, they separated it out because of electrical interference.

Double check the ECU number is correct for the year of the car. A 75 ECU on a 74 car will fit, start and run, but the air temp sensor will not work at all. Same goes for a 74 ECU on a 75 car.

The comment about the throttle body being worn is also very good. I had fits with the idle on Betty's car for a while, until we got a 2.1 Bus throttle body. Once I did that, the idle seemed to get much more stable.


Clay

@clayperrine Thanks, Clay. It is the correct ECU, as I'm the original owner and swapped a matching number ECU several years ago. This issue just started when the weather got warmer.
Can the sensor in a 6-pin 74 AFM be replaced (even if difficult)?
Will try the throttle body test, as Van suggested.

not saying its a temp sensor issue with yours @StarBear .

I'd have to find the reference again describing what happens when it does go out.
symptoms a little more extensive.

but the temp sensor is basically making an estimate of air density (via temp) and this triggers different amounts of enrichment by the ECU. so maybe this could be something going on.

there is a test for the AFM for our 74s in the factory manual.
its not really definitive as what it is testing is that mixed (modified) signal coming out of the AFM already adjusted. but its a basic test.

looks like the master @clayperrine has confirmed what its taken me 30 years to finally learn. so a slightly misbehaving temp sensor might cause some problems. not exactly sure how since idle conditions are set off the throttle switch and the afm is out of the picture i believe (at least at idle). i'm thinking the temp sensor comes into play off idle given that any signal from the AFM about temp has to get mixed in with the AFM flapper position signal. but what would i know. maybe that temp sensor does feed in somehow even at idle.

it was 10 C (50F) in your language yesterday when i cracked the 914 out. probably one of the coldest starts i have done since recommissioning it. did the usual start. nothing out of the ordinary. fell into its little hole for 2 minutes and then came on good.
5 minute warm up and then out for a drive. roof off, heater on. into a just spectacular sunset sky here thanks to the tongan volcano last year.




as a thought, is it possible a misbehaving decel valve is causing this problem you have.
ie its getting stuck. sure fits a decel vavle high idle. but rather than being consistent and requiring adjustment its doing it intermittently? a thought.

@wonkipop @Van B @clayperrine
UPDATE #2:
1. Found the hose on the back (rear) side of the throttle body was a bit loose. Put some sealer on the nipple and put hose back on. Nice and tight now.
2. Found the lever connected to the spring (see pic) doesn't pull back all the way when not running; there's 2-3 degrees of play where the spring doesn't fully reset the lever backwards. Weak spring (replaced it only a few years ago) or sign of the worn throttle body? Shot in some silicone spray and will let soak in; maybe do another test drive this afternoon.
Onward!Click to view attachment

That sounds exactly right. I suppose after 50yrs of heat cycles, the spring steel in that AAR loses a bit of its action. Which would be most readily observed at the margins of its movement i.e. it doesn’t open all the way or close all the way without being really cold or really hot.

What I previously described is my method to see if the throttle plate is seated properly when the throttle is closed. And is best done when the car is running so you can hear a change in engine speed vs detecting 1-2mm of travel with your mits.

Springs and levers and such can get dirty and gummed up which obviously make for stick points in the travel.

But I say get the car up to temp and idling steady and then try to push the arm “more closed” and see if idle drops, even a little. If it does, it means your throttle plate and bore have worn. That’s what’s wrong with mine and will require a rebuild.

Solved, I think/hope.
Did the TB test suggested by Van. No stalling or change. Van, maybe we can chat during the upcoming Gathering to make sure I did it correctly.

The hose going into the back of the TB is nice and snug.

Drove for 20 minutes and the idle didn’t fall off from the preset slightly higher idle the entire time. So , have backed off the higher idle a bit and will retest tomorrow morning.

Still begs the answer to all our questions about if/where our 1.8s have a Temp Sensor I.

Onward!

It is built into the air flow meter. You cannot replace it.

Clay

Hmmmm and suppose no way to test it separately, either?

Resistance figures outlined in the manual are all we get.

Ok; thanks!!!

you might have fixed it @StarBear .



a throttle not fully returning would certainly do that trick.

------

re the mysteries of the L jet AFM.
nihil44 has just come on that other thread about Air Fuel Ratio for L Jet.
has posted some fantastic images of all the guts of the different AFMs.
the temp sensor is definitely there in the 74 but there is no visible wiring in.
so its some kind of direct connection that is integrated as clay perrine states and as i surmised.

http://www.914world.com/bbs2/index.php?sho...60274&st=40

its just for curiosity value @StarBear .
guarantee its not your temp sensor playing up.
did a bit more reading on that and if it was a defective temp sensor the car would be running badly all over the rev range and in extreme cases might not even start and run.
if you are the least bit worried just do the test i posted on page 2.
you cannot isolate out the temp sensor but i guess given its part of it all if the test values don't come out good it means it could be the villian and there is not much you can do about it.

i've been trying to find factory manual info on the 75 and can't find anything.
was thinking there must be a test for the sensor for that one given its wired in separate.
but nothing in the literature i have.