Looking for a little insight on the dreaded idle mysteries of a 1974 2.0 914. I've searched the archives for a couple months and tried just about everything.

Valves freshly adjusted. Compression is above 135psi and within 2% on all cylinders. Original D-jet intact with new hoses and all components tested and working. Fuel pressure tested (30-31psi from original 3 hose fuel pump). New plug wires, newly rebuilt distributor, recent fuel filter. I'm about to replace the 20 year old CHT sensor as well.

The car revs smoothly and pulls very strong. Engine runs great anywhere above 1200 rpm.

Problem is 3 or so minutes after a cold start it drops off the high idle and transitions to a rough idle, like it's running on three cylinders. When I spray carb cleaner in the air cleaner the idle smooths out immediately.

Has anyone else come across this issue and perhaps suggest a plan of action to resolve it?

Thanks and Happy New Year to All.

AAR?

Does it ever smooth out after warm-up? If so, could be the dreaded "donut hole" of running too lean until full warm-up. The CHT richens the mixture when dead cold, but ramps down too quickly, so the D-Jet runs lean until the car is fully warmed up. It's a common discussion here and elsewhere, and short of running resistors on a switch to trick the D-Jet ECU I've not seen a solution.

Cold start valve works only during cranking (and only below ~40 degrees), so that's not it.

Three minutes does sound like the AAR, but the closing of the AAR - the computer is unaware of the AAR's position - that would richen the mixture, not lean it out.

So...back to the first sentence: does it ever smooth out after warm-up? Or does it continue running like that forever?

Sounds like you could use one of Brads CHT spacers. 914LTD

I got a spacer two months ago along with a new CHT and it solved a lot of similar problems.

I've got a spacer, but haven't installed it yet because I also have the dreaded "warm start" issue where, after parking for 10 or so minutes, the heads have cooled but the rest of the engine is still warm. In this mode, D-Jet over-richens the mixture based on the CHT reading and it's hard to start and idles poorly for a few minutes until the heads warm back up.

I'd guess that the CHT spacer might worsen the warm-start problem by making the heads appear even cooler when the block is still warm and/or delay the sensed head temp coming back up. What have you experienced with yours?

I appreciate Brad as much as we all do.

Spacers that were never factory are band aids. Can we stop the band aids and get to the real problem? We need a baseline. Obviously there's something else wrong. Please don't try patching it other ways. Fix the source of the problem before you chase mysteries the rest of your car life.

I haven't warmed it up for more than maybe 8-10 minutes since discovering this issue and therefore can't answer that question, yet. I'll give that a try before installing the new CHT. I also ordered Brad's spacer and plan to test the new CHT sensor with and without the spacer. Hopefully this weekend.

About a year ago the car developed a bit of warm idle surging. After restarting the warm engine it would surge from maybe 600 rpm up to about 1500 maybe three or four times before settling down. Two years ago it didn't have this problem. I suspect that the CHT is maybe dying a slow death.

[quote name='Apexster' date='Jan 3 2020, 10:05 AM' post='2774182']
Looking for a little insight on the dreaded idle mysteries of a 1974 2.0 914. I've searched the archives for a couple months and tried just about everything.

Valves freshly adjusted. Compression is above 135psi and within 2% on all cylinders. Original D-jet intact with new hoses and all components tested and working. Fuel pressure tested (30-31psi from original 3 hose fuel pump). New plug wires, newly rebuilt distributor, recent fuel filter. I'm about to replace the 20 year old CHT sensor as well.

The car revs smoothly and pulls very strong. Engine runs great anywhere above 1200 rpm.

Problem is 3 or so minutes after a cold start it drops off the high idle and transitions to a rough idle, like it's running on three cylinders. When I spray carb cleaner in the air cleaner the idle smooths out immediately.

Has anyone else come across this issue and perhaps suggest a plan of action to resolve it?
[quote]
OP mentioned he was replacing the CHT sensor but didn't say if he had tested it or had disconnected it to achieve a different outcome.
Little point in carrying on until this is done,the sum of all the system parts work together,when one parts is down you have .
There are other factors to consider like timing or electrical which are often confused for a fuel system issue.

Fully agree with Rand.

Provide some more info: Has the car always done this or did the issue just start?

Things that immediately come to mind:
Verify the correct model year DJet parts are installed
Injector wiring harness is good
Vacuum hoses hooked up incorrectly
Vacuum can on distributor leaking
intake runner hoses
throttle body gasket
Leaking plenum (only way to do this is to remove it from the car)
bad injector seals
Idle mixture knob on the ECU is not set correctly
Bad MPS

...to name a few.

If the idle is surging I would start by adjusting the ecu knob to see if that changes anything, then test the MPS and dist vacuum canister to see if they hold vacuum. If it's not surging the MPS could still need to be adjusted.

My fluctuating idle issue (from 900-500 rpms) turned out to be a combo of a throttle body with a worn shaft, a bad vacuum canister on the distributor and a bad MPS. (And I'm probably forgetting something else.)

Just re-read the above from your first post- Does the engine rev smoothly before you hit 1200 rpm or does it bog down?

Reasonable theory. A surging idle is a direct symptom of running too lean. Your "fixing" the issue with carb clean is...a sign of running too lean. Sounds like you're on the right track.

I've often wondered (rhetorically speaking) if there wasn't another CHT we could try. The D-Jet system was used in a lot of cars...

Edit: I'm guessing you have not had your injectors tested/cleaned lately? Might not be a bad thing to do...

This issue started recently. I've been chasing a typical erratic idle for some time but this rough idle started after rebuilding the distributor and replacing the MPS. I've successfully tested almost every component and sorted all other issues with the exception of this one.

MPS is newly rebuilt, trigger points, AAR, decel valve, fuel pressure, hoses, injector boots, intake sleeves, verified '74 2.0 d-jet parts, timing, valves, etc.

Today I tested both the new (012) CHT and the 20+ year old one that was in the car. I tested both CHT's in and out of the car with interesting results. In the car the NEW CHT (012) worsened the problem. The engine dropped off the high idle in less than a minute and then idled a little worse than it did with the old CHT.

OLD CHT: 2600 ohms @ 68F, 195 ohms @ 176F, 146 ohms @ 190F
NEW CHT: 2400 ohms @ 68F, 181 ohms @ 176F, 120 ohms @ 190F

The new CHT definitely made the car run even leaner and drops off the high idle faster than the old one.

My plan of action is to verify the injection wiring harness, test the injectors and try Brad's spacer on the old CHT next. I'm wondering what else makes a 914 lean during idle?

Really want to get this little screamer back on the road asap.

If the MPS has been rebuilt you absolutely need to calibrate it for your car. I’d bet a few bucks that it’s causing your lean condition especially since the issue started after rebuilding it.
And have you verified yet if the vacuum canister on the distributor holds vacuum?

I replaced my injection wiring harness with a new one which chased a lot of gremlins.

Vacuum canister was verified prior to sending the distributor out for a professional rebuild (and full test set up on a Sun distributor tester).

I was operating under the assumption that the MPS is a sealed unit. Thanks for that heads up. This morning I jumped on Paul Anders MPS page. I have a feeling you're right about that.

After that I'll clean the injectors just for good measure.

Did you adjust the idle mixture after the engine was hot? Try adjusting the idle mix via the knob on the ECU after the car is hot say 20 minutes of driving. Set it to the "best" smoothest idle by turning the knob.

Link to Tangerine Racing's site with the tools needed to adjust if it's been opened up.
http://www.tangerineracing.com/mpsdiaphragm.htm

If the cover wasn't modified to allow for it to be adjusted fully, you'll need to do that yourself. There's probably a thread on here somewhere with pictures.

I read through Anders MPS webpage and found that the MPS controls part load and full load mixture. The TPS and ECU control idle mixture, according to what I read. Because my problem SEEMS to be limited to idle and because the old MPS I replaced was a sealed original unit that clearly had never been modified, I'm hesitant to open my rebuilt MPS and mess with it until such time that I'm positive doing so will affect idle mixture. Any thoughts of the MPS's influence on idle mixture?

The MPS is the primary fuel injection sensor to determine injector pulse width (affecting A:F mixture) based on intake vacuum. It could be no influence if it is fully functional and correctly calibrated, or it could be a contributory if not the sole cause of your issues if not.

Vacuum is what controls the MPS so if you have any vacuum leaks it throws off the MPS. Reading the thread thou and you seem to have a lean idle so adjusting that first correctly would be my first step.

Have no idea why you are rubbishing this part,Brad knows his stuff and appears these have worked for many people,read an old thread from on here and they were a factory part,they became NLA and Brad started making them,Paul Anders mentioned the one he fitted worked well....do some research man .

Thanks for that input.

My cold/dry engine compression is 137-140 across all four. Would this be considered higher than standard compression? I ask because in Ander's MPS write up he notes that higher compression increases the need for enrichment.

The PO had the engine rebuilt but he wasn't a gear head type and didn't know specifics such as whether or not the displacement had been increased any.

The color of the plugs also indicates a lean condition on all 4 cylinders.

Finally, is there anyone in the club who rents/loans/sells the test equipment for this procedure? I already have Tangerine's contact info for tools and parts.

I sent out the factory ECU brain and had all the circuitry inside re-serviced and soldered. This is a known issue.

Where did you send it? Ballpark on what it cost?