1jz: unusual high idle on cold-start

[bk_]

I am experiencing a weird issue that I cannot pin down.

Basically engine revs to ~2000 rpm on cold start, for 20 seconds or so (depending on temperature a little bit), and slowly drops down to ~1000 rpm in increments of 50-100rpm every 2 or so seconds.

On warm start, it hits 1500 rpm and quickly (within 2 seconds) drops to 800 rpm or so.

Weirdly enough, a few months ago the problem solved itself for a week then it came back out of nowhere. When it appeared fixed, coldstart idle never went beyond 1000-1100.

Some information that may or may not be relevent of the setup:
- jzx81 engine with jzz30 wiring + ecu.
- jzx81 intake plenum temperature plug is different to jzz30 plug (soldered to suit)
- stock oxygen sensor located ~36" from turbos on the collector intersection.
- custom turbo dumps: 2" into 3" collector.
- idle control valve was cleaned but appears to have not altered anything
- coolant temperature sender wire shorted itself a week or two after the idle problem re-occurred. (now wired directly to cluster)

fuel economy is around the 11L / 100km mark (measured once) so that's no problem.

There are coolant lines going to/from the throttle body - what's the purpose? related to idle control?

I'm a bit stumped what to try/test out to narrow down the issue. Any ideas?

Thanks.

-BK.

[30psi 4agte]

the coolant lines are for a wax pellet air valve. this helps the car idle faster when cold.

You say the idle control solenoid was cleaned. what you are saying sounds fairly normal to me.

could it be that since the idle solenoid has been cleaned it is working more efficently now.
Hence you may need to reset your cold idle speed.


Another is that the weather is starting to get colder in the mornings and the choke may be causing the car to idle faster cos its colder

[CrUZida]

Make sure the throttle plate is 100% closed and the TPS is in IDL at this time.

[bk_]

OK.

with ICSV disconnected, engine revs to ~2100 rpm (only tested for 5 seconds)
with ICSV connected, engine revs to ~1900 rpm and drops to 1000 rpm eventually.

How to reset cold idle speed? remove EFI fuse? I don't remember resetting the ECU after 2 months getting the car going. maybe that would be the go.

the max idle revs don't really change due to ambient temperature, it's the length of time revving at 1800 rpm that changes. in summer it was about 10-15 seconds... cold morning like yesterday it was perhaps 30 seconds.

But the revs dropping in increments - it looks consistent every 3 seconds, not erratic or random jumps, so the ecu is controlling it for sure with the solenoid of the ICSV.

I'm starting to think it may be the intaken plenum temperature sensor, that was the only different plug comparing a JZX81 engine and a JZZ30 loom, everything else matched up. JZX81 plug is round, JZZ30 one is square'ish. but then again the engine wouldn't run properly if it was different in operation. But why the different plug? heh.

-BK.

[30psi 4agte]

By cold idle i meant idle. ( my bad )

im only going from memory here as im not heaps familiar with the 1jz set up but i think there is a little vac valve attached to the throttle body and it moves out on start up to move throttle plate ????????? i sure there is something around there anyway have a look and tell what else is attached to the throttle body.

sorry to be so vague but i just cant quite remember.



Its just weird that it still revs high on start up with the ISCV disconnected something else must be doing it.

Might pay to have someone else start the car up and you be under the hood to see if you can see anything .

[bk_]

Good news.

Resetting ecu appears to have solved the problem! now she idles much lower and quieter (1200 and below == relatively quiet). and warm non-start idle is 650 rpm, instead of 800 rpm.

you were right 30psi 4agte

(are you the same 4agte that solved my starter motor issue with a one word post of "ground?!?" ? Or maybe another 4agte owner who also kick ass when it comes to problem solving

CrUZida, I did check if the throttle cable is tight - it is loose. Even if it was tight, there's about 10 degrees or so movement before the idle switch is O/C. If it was a problem, I wouldn't be able to rev past ~2000 rpm, heh. So thanks for that suggestion everything helps

off topic:
now I understand why toyota is number 1. mercedes owners are the worse people on earth - I was even accused of my car being rice by an american because it has a japanese engine instead of so called "german engineering" (starter motor _twice_ the size in each direction?!?!? burning 1L of oil per 1000 km is normal?!?!?) which is shocking when it comes to prices, performance, engines and (automatic) gearboxes.

I'll see if I can get a member's ride post going by next week - It's fair to say that I've learnt quite a lot from this forum (and the old toymods forum, especially from norbie's posts! in the end he's right that it's only right to give some knowledge back, in some form or another.

Respect to the toyota guys

-BK.

[30psi 4agte]

No prob m8
I love it when im right bummer its just not more often. lol

[ndgcpr]

i just cleaned my ISCV today and it was ideling really high as well. but yeah an ECU reset seems to fix it, maybe it compensates when the valve is sticky.

[30psi 4agte]

not sure about toyotas but i know the fords EA-ED can actually learn and compensate for things like that. Toyota being as good as they are i would say its the same!

[bk_]

hmm... I spoke too soon. problem is back next time around on cold-start.

ndgcpr, I also did notice the idle increased ~200 rpm or so after it was cleaned (few months back). So it made it even worse. heh.

Engine feels a bit... non-smooth whilst driving in the first few minutes, when the idle problem occurs. But when the ecu is reset, it felt like how it is when warmed up - smooth. I'll verify this again tomorrow or so.

If resetting the ECU helps (confirmed by ndgcpr and by me yesterday) then I think I'll have to connect the ECU that it powers down when the ignition key is removed, and vice versa.

It's a hack - but atleast it should solve the problem nicely.

I'm surprised this electronic controlled idle valve type setup is a problem on other engines aswell. Happens on a Jaguar XJ40 and on nissan r32, but the valve is closed by default (ie: hard to start).

-BK.

[CrUZida]

Make sure the throttle plate is 100% closed and the TPS is in IDL at this time.

. .

[ndgcpr]

I think I'll have to connect the ECU that it powers down when the ignition key is removed

I don't know if that would be a good idea, there is probably a reson why it needs contant power.

I started and stopped mine a few times and it seems to be all good, i will see tomorrow how it starts as i never really got it in a proper "cold" state today.

[bk_]

CrUZida, I think you are right about the TPS + IDL. My bad for doubting you.

I made a mistake on my observation before: throttle cable was pulling the throttle open by 1mm or so. (am using original mercedes throttle cable) How did I test it? I tapped on the location where it hits the factory set screw and it tapped.

I was checking pin 1 and 3, and just _slightly_ (like 0.2mm or something small) moved the throttle and it O/C.

How did I test before? just by tapping on the actual cable and not really taking any notice. BAD way to test!

Weird that my other unused throttle body (JZZ30) has more range before it O/C. Measured it to be 5mm before it does O/C. JZX81 one only takes 1.5mm or so before it does (at warm state).

I'll see if I can swap it out tomorrow and measure both properly.

Thanks CrUZida.

-BK.

[CrUZida]

Its either pins 1 and 2, or 3 and 4, depending which way is up (IDL and E2 are the 2 pins)
If you are looking at the TPS pins, and the actuator is on the left, then its the bottom 2 pins.

With the throttle plate 100% closed (make sure the stopper is backed all the way off, some people may have adjusted it to compensate for a low idle), continuity between IDL and E2 should be <100ohms, and with the throttle cracked open it should be open circuit.

[bk_]

Some measurements:

first reading is JZX81 (currently in use), second is JZZ30 (spare)

P1+P2 closed throttle: 25 ohms, 36 (max 48 then floats downwards) ohms.
P1+P2 open circuit distance to stopper: 2 mm, 6 mm
P1+P2 before open circuit: 19.9 ohms, 19.5 ohms

P3+P4 closed throttle: 3.32k ohms, 4.29k ohms
P3+P4 open throttle: 0.93k ohms, 1.24k ohms

Not much margin for error for the distance to stopper before it goes O/C for the JZX81... and that includes a slack of 1mm!

I'll see how it goes since making throttle cable a little bit loose. If there is still a problem, throttle body will be changed over to JZZ30. Makes sense since ECU is from JZZ30 aswell. (albeit not from same donor)

-BK.