home register FAQ memberlist calendar

Go Back   Turbobricks Forums > Mechanical > maintenance & nonperformance

Reply
 
Thread Tools Display Modes
Old 08-05-2018, 04:49 PM   #1
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default [Resolved] 960 Misfire + P0340 (Cam Sens Circuit), no Misfire code

Summary:

A 960 on Motronic 4.4 will run with a missing cam sensor signal, but it will not run well.

The engine will crank for a little longer than usual when you are starting the engine.

You'll experience power loss, increase in fuel use and possibly find the engine stumbles/hesitates under certain conditions.

You can test the cam sensor with this procedure. Nb, the voltage on pin #2 will be battery voltage, not 5v:
https://www.youtube.com/watch?v=cB1VEfjGz1c

-------------------

The 960 developed a misfire/cut/falter of some sort yesterday. Here's some details...

Occurs around at 3800 RPM under heavier load (75% throttle or more). I'm guessing 3800 is around peak torque for a B6304?
Also in happens in neutral, no load around 2800.
When the misfire started to happen, check engine light came on for P0340 Cam sensor circuit
There are no codes for misfires, nor has the CEL flashed which I thought was odd. (I thought M4.4 supported misfire reporting?)

Today I reseated the connector/cleaned the connector for cam sensor and the codes comes back immediately after being cleared.

Ran without the MAF incase there was something weird happening. No change.
Plugs are fairly fresh, probably less than 2K miles on them.
Planning to plug a multimeter into the pins for the cam sensor shortly.

So my first question is....
Are the cam sensor code and misfire\cut likely to be related, or am I looking at two separate issues?

(Edit) Answer: Yes cam sensor controls injection and ignition timing.

Last edited by 955IsAlive; 10-02-2018 at 04:14 PM.. Reason: updated with resolution
955IsAlive is offline   Reply With Quote
Old 08-05-2018, 05:20 PM   #2
Lankku
Board Member
 
Lankku's Avatar
 
Join Date: May 2015
Location: Finland
Default

Leave the cam sensor connector open to test if it'll work any smoother without the sensor. It'll only give a fault code.

Did you use a T40 Torx? Top one can be helped a bit by hitting with a punch to the side. No other tips than you just have to wrench those bolts open.
__________________
444 -55, 145 Express -71,
242 DL -80 (was 4.6l)sold, 245 -88 "B6284T" sold, 245 -90 "B6284T", 965 -93 B6304, V90 -98
Lankku is offline   Reply With Quote
Old 08-05-2018, 05:23 PM   #3
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Thanks Lankku. Yeah I used a T40 torx. I gave them a spray of WD40 afterwards, I'll let that seep in then clean up the heads before I try again. Punch is a good idea.

I thought disconnecting the sensor would prevent it from running. I'll give it a go.
955IsAlive is offline   Reply With Quote
Old 08-06-2018, 12:19 PM   #4
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

Torx bit in a pair of vicegrips.
__________________
No Start Thread
ZVOLV is offline   Reply With Quote
Old 09-14-2018, 05:36 PM   #5
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Well this still perplexes me. Tried it with the cam sensor disconnected a few weeks ago and it really didn't make a big difference. It's more of a "cut" than a misfire.

Shot some video today with a datalog...

https://youtu.be/T-e5ptWDfC8

Video with logging... Engine hesitates\bucks at 8% throttle around 2800 RPM and between 75-100% at 3800 RPM. It happens under those conditions only. If you look around 1:08 in the vieo, you'll see the throttle position held steady, fuel trims look good but the RPMs drip suddenly and randomly

No codes other than cam sensor P0340. Disconnected cam sensor, no change. No vacuum leaks leaks. PCV clear

Symptoms on the road are that it will sometimes buck when cruising, especially if there's a slight downjill gradient and cruise reduces throttle.

Also, between 75% and 100% throttle... bucks at 3800 and struggles to climb above unless you close the throttle just enough to pass 3800/4000. Then you can go 100% again.
955IsAlive is offline   Reply With Quote
Old 09-14-2018, 05:51 PM   #6
dl242gt
Can play the blues
 
dl242gt's Avatar
 
Join Date: Jun 2002
Location: S NJ, a suburb of Phila.
Default

How is the engine speed sensor? They intermittently fail and don't give you codes. If your car has a lot of miles it's good maintenance to replace both sensors just to have a good stage zero.
dl242gt is offline   Reply With Quote
Old 09-14-2018, 07:14 PM   #7
PCH
Board Member
 
Join Date: Oct 2008
Location: SoCal
Default

ECU will not attempt checking for misfire (won't run that monitor) if there are other problems that it already detected. Once it knows that all fuel and ignition parameters are good then it will do the misfire count. If the count is above certain threshold which may be dangerous to the Cat than the CEL will flash instead of just being ON.
PCH is offline   Reply With Quote
Old 09-14-2018, 07:41 PM   #8
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

Cam timing off may throw a cam sensor code.
ZVOLV is offline   Reply With Quote
Old 09-14-2018, 08:50 PM   #9
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Crank sensor eh? Looks cheap enough to throw on in there.
Thanks for the tip on misfire detection PCH.
I checked the cam timing when it first happened, it's good///until then I was thinking oh s**t :O
955IsAlive is offline   Reply With Quote
Old 09-17-2018, 09:18 AM   #10
Lazarus
Board Member
 
Join Date: Jan 2005
Location: Seattle, WA
Default

Too many assumptions. My opinion is that if the sensor works fine at one RPM, it is not failing at some other RPM. A sensor problem would not be related to engine load. If you have to throw darts at the problem, this sounds more like a spark coil breaking down, or a spark wire wearing out or related cross fire. What do your plugs look like?

Lazarus
Lazarus is offline   Reply With Quote
Old 09-17-2018, 02:28 PM   #11
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

Have you tested the wiring to the cam sensor yet? I imagine it's a a 5v reference, signal, and a ground. So I would expect to see 5v on one wire, a scope would show you a solid waveform on the signal wire, and ground....should have less than 10 ohms between the terminal and a good ground.
ZVOLV is offline   Reply With Quote
Old 09-24-2018, 08:01 PM   #12
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Plugs were changed about 2K miles prior to this happening. I re-checked the basic timing today in case I missed something the last time. Everything is right on the mark....

Threw a multimeter on the connector (ECU side) with the connector disconnected.

Pin 1: Ground = good
Pin 2: Reference = 12V (probably not good, should be 5v?)
Pin 3: Battery voltage = Good

VADIS\VIDA procedure calls for a bulb to be connected on pin two to test it.
Post above and a video I saw on at 850 on motronic 4.x mentioned 5v, but it's not clear if thats with the connector plugged into the CPS or not.

Reseated the ECM\ECU... no change.

I'm out of time today, will have to re-visit tomorrow.
Have not checked the sensor itself yet... but since I have 12v on the reference signal... I think I might have a bigger problem....

Last edited by 955IsAlive; 09-24-2018 at 09:26 PM..
955IsAlive is offline   Reply With Quote
Old 09-25-2018, 07:44 PM   #13
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

12v on two wires and ground on the third sounds ok.

A scan tool, especially the factory level one, would be nice to have here and look at the data. I saw info on guys using an analog volt meter to watch the needle sweep back and forth to test these sensors. A scope is a great tool also.

IDK what else to tell ya. You can do a hard brake stand to see if there is a load based misfire like a failing coil. I do it all the time.

I have access to databases where I can punch in the fault code for the car and search for bulletins and also have the diagnostic procedure written out for me. I would do more research on the code and what causes it on that model and how to diagnose it.

A nice scan tool would really help you out too. I can look at live data on exactly which cylinder is misfiring and when. I use the factory level scan tool on stuff I work on and I have used generic scan tools plenty also.

I still think a cam code like that may end up being a timing issue.
ZVOLV is offline   Reply With Quote
Old 09-25-2018, 11:31 PM   #14
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Code is EFI-314 or P0340

Motronic 4.4 disables misfire detection when a cam sensor problem is present. I followed the VIDA\VADIS procedure to check the cam timing system, it recommended to swap the sensor. I ran a crude test by measuring the voltage between pins 2 & 3 and passing metal across the hall sensor... voltage stayed steady so it seems it's screwed. Everything else checks out.

That said, I don't think the cam sensor issue is related to the drivability problems (other than a longer than usual crank to start). Apparently the CPS is used for startup, then while running it's used to help with misfire and knock detection. But otherwise it's not used for anything performance related. In the absence of a signal from the CPS, the coils fire at every turn vs only on the combustion stroke.

The "cut" only happens under two, very specific circumstances:

Neutral around 2800RPM
In gear at 75% + throttle at 3800RPM

Otherwise it's fine. Slow as hell, but fine. Calculated load value didn't exceed 65% when I tested it at full throttle last week, so something isn't right.
Once I have the new CPS in I'll hopefully have a better picture.
955IsAlive is offline   Reply With Quote
Old 10-01-2018, 11:08 PM   #15
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Well after all the messing around.... replaced the cam sensor and it's back to normal.
Hesitation is gone, gained some power back and of course it starts up a little faster.

Lessons learned:
Sometimes just throwing parts at something is the answer.
Although Motronic 4.4 falls back to to a substitute values without a cam sensor signal, it would appear that those substitute values are not so awesome and it will still run pretty ****ty.
955IsAlive is offline   Reply With Quote
Old 10-02-2018, 11:59 AM   #16
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

Nice! Also glad to hear the results of an independent study on the role of the cam sensor on driveability on these cars.

Well you did the right thing and confirmed the wiring was good first.

Replacing parts based on codes alone isn't a good idea. It's always a good practice to confirm the circuit first. I've been lazy once recently and did an oxygen sensor based on the code, and it turned out to be the wiring.

Which tools/method did you use to turn the bolts?

Last edited by ZVOLV; 10-02-2018 at 12:23 PM..
ZVOLV is offline   Reply With Quote
Old 10-02-2018, 04:07 PM   #17
955IsAlive
Board Member
 
Join Date: Apr 2018
Location: MN
Default

Quote:
Originally Posted by ZVOLV View Post
Replacing parts based on codes alone isn't a good idea. It's always a good practice to confirm the circuit first. I've been lazy once recently and did an oxygen sensor based on the code, and it turned out to be the wiring.
Agree and that sucks with the sensor. I keep telling people that codes are just clues, they aren't "hey replace this part", it's "hmm look into this". So many times I hear of someone (even garages that should know better) replacing an oxygen sensor because of a vacuum leak.

To turn the bolts.... I used a punch on the top one (like lankku recommended) which worked great, bottom one I was able to get to with the torx by manipulating the heater core pipes.
955IsAlive is offline   Reply With Quote
Old 10-02-2018, 08:01 PM   #18
ZVOLV
<Master Tech>
 
ZVOLV's Avatar
 
Join Date: Nov 2002
Location: California
Default

That's where a high end scan tool is very helpful to look at the data.

Or a scope would have shown a good or bad waveform in this case.
ZVOLV is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


All times are GMT -4. The time now is 07:26 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.