[derekj308]

Hi Everyone

This is my first post on Kia-Forums so please be kind . I own a 2008 KIA Grand Carnival EX with a G6DA 3.8L engine. I do my own repairs and I am stumped at the moment with a P0340 error. The symptoms are hard starting, rough idle, misfiring, etc. A few weeks ago it went into limp mode on my way home and I was luckily able to get it home. I put my OBD2 scan tool on and found the P0340 error and started to research what it was. Many sites said to not blindly change the CMPS so I have been doing simple testing and found that the reference voltage for both B1 and B2 CMPS as measured at the harness drops from 5V to 1.2V as soon as I plug either of the connectors for B1 or B2 into its corresponding CMPS. I am doing this with the ignition to the ON position and when either of the CMPS are connected to the harness I am back probing the connectors.

I have also tested both B1 and B2 CMPS with both connectors plugged in and when I run a metal object over the sensor the reference voltage drops from 1.2V to 0V. This leads me to believe each sensor is ok, just the reference voltage is pulled from 5V to 1.2V as soon as the reference voltage is in circuit.

I thought that in normal operation, the reference voltage was supposed to be 5V and drops to 0V when the sensor is active. Is this correct?


Any thoughts, similar experiences?


Thanks in advance

Derek

 

[bopbo]

I am changing the spark plugs today (Sunday), I will have a look at what a good reference voltage should be, But I am surprised the 5V gets pulled down I am pretty sure it should stay at 5V.
I will have a quick look and get back to you
Cheers

 

[derekj308]

Thanks bopbo. much appreciated!

 

[bopbo]

Hi derek, as expected.... 5V stays at 5v with 1 or both cps plugged in or not. there is something wrong with the 5V rail, although there are different 5v wires coming from the ecu, internally they are all connected, it may be a different component pulling the volts down.

 

[derekj308]

Thanks heaps bopbo, you are a legend. At least I now know when I think I have found the issue the signal voltages of the CMPS will stay at 5V.

 

[bopbo]

Both pinks are 5V

Brown -ve Grey signal

Black -ve Green Signal

 

[derekj308]

Same on mine. I haven't been able to find the source of the problem yet. Will post the cause when I find it.

 

[kiaguy002]

Although I'm not familiar with your vehicle, the first place I'd look for any unexplained voltage drops is the fuse box. It's not unusual to find corrosion in the fuse box of many makes and models, which can often cause resistance high enough to significantly degrade electrical flow. The corrosion can be located in the fuse box connectors, which is easy to see and clean. However, it can also form on the inner layer where much of the circuitry is located, and that can be more difficult to get to and remove, depending on how the box is constructed.

With access to a good wiring diagram, you could do some resistance testing of the wires of the problem circuits.

 

[derekj308]

Hi kiaguy002

Thanks for the reply. Interesting you mention the fuse box. My ground short testing includes the fuse box circuit as part of the issue. With the ECM A and B connectors off and the two top connectors of the fuse box off I've got around 4k ohms resistance to ground on the blue wire of the trans input speed sensor harness connector. Its a three wire sensor and I think it is hall effect so my guess is the blue lead is supposed to be 5V supply. I followed the blue wire all the way back to the fuse box. Hmmm. That same wire in the same disconnect configuration as above is also connecting to pins on the A and B connectors of the ECM somehow through the fuse box. I have the measurements written down but I got to go to work so I'll post later tonight.

Thanks for your support guys, it is much appreciated. As fun as this can be, this one is doing my head in.

 

[derekj308]

Hi kiaguy002

I got the fuse box apart and checked the connectors. They all look fine no noticeable signs of corrosion. I did get into the internals of the fuse box but didn't keep going for fear of breaking something. I assumed I would have had to pull all of the fuses and relays to be able to disconnect the board, etc. I not going that deep just yet.

I'm not so sure my approach was correct by measuring the resistance to ground through the input speed sensor harness. I had an oil leak from the oil filter housing which flooded the input speed sensor so I thought it was a good place to look and when I found a resistance I thought I was on to something. Initially what I didn't realize is that it wasn't part of the ECM connector A or B loom. Both the input and output speed sensors for the trans both get their power from the fuse box, not directly from the ECM like the two CMPS do. When the fuse box is connected to the underside 3 harness plugs I am assuming there will be a path to ground and would show a high resistance (in my case ~4k ohms), not a direct short. When those wires are out of circuit with the fuse box they test fine, no shorts to ground. I need to understand why the act of connecting the CMPS drags the signal voltage down from 5V to 1.2V and then I can begin my search for potential culprits.

Any suggestions are welcome. I am new to auto electrical fault finding and I am binging on ScannerDanner videos to try and get up to speed.

 

[kiaguy002]

… Both the input and output speed sensors for the trans both get their power from the fuse box, not directly from the ECM like the two CMPS do. ….

So you have a wiring diagram that shows the cam sensor reference wires coming directly from the ECM, or you confirmed that by tracing the wires physically? If so, that would of course eliminate the fuse box as being the issue on the power side.

Assuming you did find the cam sensors directly powered by the ECM, and have the ECM pinout, then you can do a quick resistance test from the ECM power pins to the cam sensor reference pins, just to make certain there's no issue in the wire run that's inside the bundle.

 

[derekj308]

Hi kiaguy002

I have traced the wires back from the ECM plug B and they do connect directly however I haven't done a resistance test to ground with the B plug off the ECM so that's my next task. Tomorrow job, it's 11:30pm and my brain is fried.

 

[derekj308]

Sorry I keep calling it an ECM, it is referred to in the schematic as a PCM

 

[kiaguy002]

Based on the wiring diagram that you posted, here's the resistance testing that I recommend you try. All results should be zero ohms.

For the cam sensor labeled "sensor 2 left rear of engine"; test what the resistance is from sensor pin #1 (pink) to ECM
connector pin #15; then from sensor pin #2 (gray) to ECM #24; then from sensor #3 (brown) to ECM #17.

For the cam sensor labeled "sensor 1 top rear of engine"; test what the resistance is from sensor #1 (pink) to ECM #32; then from sensor #2 (black) to ECM #18; then from sensor #3 (green) to ECM #25.

Sorry I keep calling it an ECM, it is referred to in the schematic as a PCM

Actually Kia/Hyundai calls their engine control computer the ECM, so the doc you're looking at used the wrong name. However, the names PCM, ECM, ECU, PCU (and possibly others as well) have become so interchangeable that most people are not bothered about which name is being used. This all came about because the automakers were not consistent in their choice of names, and it's nearly impossible to remember which names each maker uses for their computers.

 

[derekj308]

Hi kiaguy002

I did the tests you suggested and found 0 ohms resistance on each of the 6 wires between the ECM plug B disconnected from the ECM and the disconnected CMPS plugs. I also did a resistance test with the ECM plug B and ECM plug A connected to the ECM and both CMPS plugs were disconnected. What I found was the following;

0 ohms - Brown (GND) of CMPS B2 to Black (GND) of CMPS B1

3.3k ohms - Brown (GND) of CMPS B2 to Green (SIG) of CMPS B1

3.95k ohms - Brown (GND) of CMPS B2 to Pink (+5V) of CMPS B1

3.3k ohms - Brown (GND) of CMPS B2 to Grey (SIG) of CMPS B2

810 ohms - Brown (GND) of CMPS B2 to Pink (+5V) of CMPS B2

There is a discrepancy on the +5V line, my guess is the 810 ohms on CMPS B2 is the culprit. I'm about to go and measure the resistance directly on the pins of the ECM to ground to narrow it down to either wiring or internal to the ECM.

 

[derekj308]

I think I'm on to something now. I can make the 810 ohms change to 3.6k ohms. I measured the resistance directly on the pins of the ECM and it was 810 ohms so naturally I thought it MUST be the ECM. Plug A of the ECM was still connected so I thought I'll test with plug A disconnected and then it HAS to be the ECM. I tested again with plug A off and the resistance was 3.6k ohms. Huhhh? Not the ECM it would seem.

I then started to try and isolate the fault on the circuit that is part of plug A of the ECM. In short I have got it down to the circuit which consists of;

plug A of the ECM
top two plugs of the fuse box
the fuse box

I am able to disconnect all of the three underside plugs of the fuse box so they can be ruled out.

I'm getting closer

 

[kiaguy002]

Wow, you're doing a lot of work there, and I tip my hat for the effort you're putting into this. However, I suspect there's a validity issue with drawing any conclusions from the resistance readings you're seeing with key off, because the power circuits are not completed until the key goes to on/run and all of the required relays close.

I normally put bad computer way down on the list of possible suspects, but the testing that you have already done has unfortunately moved it right up to the top of the pile. Now it still might be possible that a supply voltage circuit coming into the ECM is compromised, so that's what I'd be wanting to test next. In order to do that, you would need the schematics for how all of the ECM supply voltage pins map back to the fuse box,

 

[bopbo]

Hi I ohmed 2 good sensors, both unplugged and removed from vehicle
-Ve on A +ve on B =sensor 1 729 sensor 2 731
-Ve on A +ve on C = both sensors open
-Ve on B +ve on C = sensor 1 1481 sensor 2 1488
+Ve on A -ve on B = both sensors open
+Ve on A -ve on C = both sensors open
Not sure if it helps but more info
Brian

 

[derekj308]

Hi kiaguy002, Hi bopbo

@kiagu002, I hope it isn't the ECM . I've dug deeper and I have narrowed it down to pin 63 of connector A which connects to the ECM. My logic is that connecting connector A to the ECM causes the resistance to change from 3.6k ohms to 810 ohms so connector A must play a part in the fault. Key off, no battery connected, I probed all of the pins of connector A and found only one pin, pin 63 connects through to ground with a resistance of initially 4.8k ohms but I have since recorded 4.3k and 4.5k ohms in subsequent repeat tests. I am at this point stumped since I can't follow the circuit diagram correctly to be able to create a list of potential candidates which could give this resistance.

I was able to follow pin 63 along a circuit diagram path through to the MIL lamp in the instrument panel but I'm unsure if that path can actually go to ground. In the process of following that path I was able to verify that the fuse box is included in that circuit and offers ~1 ohms of resistance which is the error on my DMM. I'm calling the fuse box .

So once I enter the circuit diagram of the instrument panel I am unsure where to go from here. I thought that pulling the fuse IG-1 would break the circuit to the MIL lamp circuit but it doesn't, When I pull the IG-1 fuse (under the dash, RHD car) the resistance on pin 63 changes from 4.5k ohms (see range above) to 5.3k ohms.

I do have a circuit diagram now and so far it is as per what I am finding in my vehicle. I got it free from cardiagn.com so I am assuming it is ok to post the links here. Heads up guys there are some ad links so be careful, read before you click. I tried the links I used and it was hanging at "Please Wait A Moment" so if you can't get them PM me and I'll send them directly to you.

The full set for 2008 Sedona (Carnival) linked here.

The circuit diagrams are linked here. The pages I am looking at are;

Pages 14 - 16 : Front Area Module Circuit

Pages 38 - 46 : Engine Performance Circuit

Page 70 : Instrument Cluster circuit

Page 90 : Power Distribution Circuit

For those of you reading this who haven't read this style of circuit diagrams before .... the pin numbers for each of the connectors for the ECM and the FAM (Front Area Module a.k.a. the under hood fuse box) are where the wires go into the plug. Not all pins are are labelled. You will be able to work out what is what by what is present. You need to cut the cable tie off and remove the back cover plate to see them. Also, the numbers on the circuit diagrams which have multiple pages read from left to right as if each sequential page was laid out left to right. For example, in the engine performance circuits there are 5 pages. Page 38 is the first page in this section and on it you will see that connector A which is shown as C01-1 is split into two parts, one part is pins 1 through 60 and the other part is pins 61 through 80, hence an 80 pin connector! The numbers on the right hand side of that page are interlinked with the next page on its left hand side. Likewise for consecutive pages. At the end of this section on page 46 is the B connector which is shown as C01-2 and is also split into 2 parts, pins 1 through 60 and pins 61 through 80.

Hope this helps!

@bopbo, thanks mate much appreciated . I just saw you are in Melbourne, same. Maybe we can have a drag at Calder and see if one of us can break a 16 second 1/4 mile 😂🤣😂🤣. Seriously thanks for the effort, I'm really feeling the support from this forum. I measured my CMPS and they are very different to yours. I only get 2.4k ohms between the ground wire and the signal wire on both sensors. All other combos are O/C. They are consistent so I'm assuming its a difference between '06 and '08.

 

[bopbo]

I measured good second hand sensors that i took from a vehicle while i was at pick a part or similar they have been run in my vehicle. I have no need for them I could send them to you if you wanted.
The pcm supplies the earth to light the MIL that is why the resistance is inconsistent.