Finally I can find some ppl that are going through the same problem that I'm going through, god bless you all
I have posted my issue on the g37 and some ppl didn't believe it was an issue and some never knew that this problem existed but I've done a lot of research and found out that infiniti/Nissan knows about this issue but failed to correct it.

I have change both the sending unit and fuel pump after the dealer ship told me it needed to be changed but gauge cluster is still not reading right . my car broke down couple time on the high way without the gas light notification been on.
I was wondering if now it is the same process for the 2009 g37xs or should I check the fuse too?

tks in advance.

 

been following a few threads about reflowing the cluster circuit board. before i 'cut' i want to measure 5x and ask the necessary questions for my 07 6MT coupe
* posts says NOT to remove needles from gauges. if so that means you have to drill through the cluster faceplate to get to the screws that secure the faceplate to its chasis?

* @Richie aboves says NOT to drill into the faceplate, if so that would entail removing the needles to be able to remove the faceplate to be able to get to the screws? how else can i get to the circuit board?

any input would be great. thanks in advance

 

So I have been reading all kinds of posts about possible solutions to the erratic fuel gauge issue in the 04-07-ish G35s and I have a real solution that works. The problem is in the sending units (one or both). There is a contact on the float arm that sweeps across the resistor board to designate the proper height of the fuel level. This contact is initially round and only contacts the board at a very small spot. (One increment at a time). Well over time this brass contact (dome) wears down to a flat area and now contacts more that one line of the resistor board at a time. When it contacts more than one path it increases the resistance proportionally. (Hence the erratic behavior). So the solution is to clean up the board, as many have suggested, but more than this, you must use an emery board or some sand paper to round the surface of the contact again, so the it functions as intended. I took a photo of the contact after I resurfaced it with my wife's nail board. (Shh) All is working as intended after the repair. It was very east to do, just be careful with the resistor board. To disassemble the sending unit is the same for both units. Simply remove from the car and push in the tabs on the small white cover. Then using a small flat blade screwdriver pop off the metal float arm. After the arm is off, it is easy to disconnect the wire and pry up the contact where the wire was connected. The fulcrum pivot point simply moves to one side and comes up, spring and all. Resurface the, now This is the removed contact assembly. Note the small brass dome has been rounded again to contact only one resistor line on the resistor board at a time.
flattened out contact and clean up the resistor board with alcohol and a q-tip. Reassemble and done. Saved $400.00 on parts and it should last as long as it did the first time, I guess. Hope this helps SOMEONE. Cheers!!

 

And this is exactly why you troubleshoot the problem before blindly replacing parts, the 3-80 ohm resistance change would be obviously not working when you measure across pins 2 and 5 on the fuel pump harness indicating a failed sending unit and not the more common instrument cluster failure.

Good tips on fixing the existing unit!

 

Make sense but after the reading of the entire post, my biggest question is, why those who have replaced the whole sending unit for a new one haven't solved the issue?

 

Because the problem is usually a failed combination meter but they don't know how to troubleshoot so they just throw in a new fuel pump / sending unit with their fingers crossed that it will fix the problem.

 

Thanks for the soldering locations, my gas gauge works again. The one thing everyone should know is that you have to set the needles before putting everything back. I plugged in the cluster with the front cover off and set the car into diagnostic mode which makes the needles go to a certain location, I then installed the needles at the given locations.

 

So I have been reading all kinds of posts about possible solutions to the erratic fuel gauge issue in the 04-07-ish G35s and I have a real solution that works. The problem is in the sending units (one or both). There is a contact on the float arm that sweeps across the resistor board to designate the proper height of the fuel level. This contact is initially round and only contacts the board at a very small spot. (One increment at a time). Well, over time this brass contact (dome) wears down to a flat area and now contacts more that one line of the resistor board at a time. When it contacts more than one path it increases the resistance proportionally. (Hence the erratic behavior). So the solution is to clean up the board, as many have suggested, but more than this, you must use an emery board or some sand paper to round the surface of the contact again, so the it functions as intended. I took a photo of the contact after I resurfaced it with my wife's nail board. (Shh) All is working as intended after the repair. It was very east to do, just be careful with the resistor board. To disassemble the sending unit is the same for both units. Simply remove from the car and push in the tabs on the small white cover. Then using a small flat blade screwdriver pop off the metal float arm. After the arm is off, it is easy to disconnect the wire and pry up the contact where the wire was connected. The fulcrum pivot point simply moves to one side and comes up, spring and all. Resurface the, now flattened contact point and reassemble.

 

[QUOTE=Riche;7130715]So I have used two different people on eBay (the one in NJ mentioned above and the one in Texas) to fix a cluster and both did not do a very satisfactory job /QUOTE]

I'm taking some time to thank you Richie, after doing a cluster test and trying a new cluster I realized I was having issues.

I followed your guide and MAN IT WORKED.it is a bit tricky because they use very little solder on these boards and it's very easy to mess up.

I used a bluetooth obd to calibrate the needles and I was done.

I would reccomend before you tackle the senders to try this first.

THANKS AGAIN RICHIE!

 

I posted this in the M-spot forum by accident. Hoping cleric sees this over here and chimes in.

So my gauge has been doing the always reads half tank when empty thing for a while now and I've finally gotten tired of being stressed about setting/watching the odometer for fuel level. I've read through the thread and measured both my sensors. With a close to empty tank (i filled with 17.5 gal after this test) I measured 37ohm on the fuel pump side and 40ohm on the driver side with the sensors installed. I pulled them both out to measure the resistance over the range of travel and confirmed 3ohm on both when the lever is at the top stop. The pump side measures 80ohm at the empty stop and the driver side measures 40ohm at the empty stop. These measures seem to match what is described in this thread as normal. What I find unusual is how non-linear that last gallon or two must be for the pump side and if that's normal. Does this seem right?

I guess I'll pull the cluster and check the resistors....

 

JBarnett - tks for sharing your work!! Let us know what you find out on the cluster. I have yet to fix mine but betting big that it is the cluster. Tks again!
-rh

 

I realize what I should have done during my measurements is to plug the pump side back in when it was pulled out of the tank and see what the gauge read when the float was all the way at the empty stop reading 80ohms. if the gauge would indicate empty then, my problem is in the sensor not reading the level correctly. but if it's still reading half full it's in the cluster. I think I'll do that first because the threads about pulling the cluster sound like a pita. i want to be sure i need to do it before going to that trouble.

 

I finally went back and pulled my pump side sending unit and confirmed my gauge couldn't indicate empty even with the sending unit at the empty stop. I also did the combination meter self test thing and my fuel gauge looked just like Turtle's post above with the needle below half. So I pulled the cluster and checked the resistors. The solder joints looked good with a macro lens and measured ok so I was skeptical that touching them up would fix anything. But once I retouched all the solder joints on all the resistors indicated in Riche's post, the gauge could indicate empty again. I haven't run the tank full to empty again yet but so far it looks like it's fixed.

Things I learned removing and disassembling the cluster:
1. I couldn't find a diy for an 06 model to remove the cluster and the trim was just slightly different so I used the FSM but there weren't always step by step pictures. I followed an old cluster disassembly diy on here and the pictures don't work any longer. I took some pictures to update that old thread.
2. Pulling the cluster isn't that bad. I got confused reading the FSM and was pulling on a trim piece I thought was the steering column upper cover but was actually attached to the Cluster Lid A and broke a plastic tab and bent some reinforcing bar. Just read and be sure what you're pulling on.
3. I wasn't sure how to get the needles aligned correctly during reassembly so I just put them on where it felt like the stepper motor would rest with the needles on the stops. Completely wrong. You need to set the steppers to a known position using the self test mode and then install the needles. So after installing the overlay and before needles and clear plastic cover, plug the cluster back into the car and do the self test and install the needles in the positions shown in the manual. It's even a good idea to start the car and see where your needles go before reassembling everything because you might want to adjust them again.
4. The odometer display mounted in the overlay has two rows of connector pins that you have to blind mate to the pcb when installing the overlay. This is tricky and another reason you want to test the cluster before putting things back together. If you misaligned a row of pins, the display isn't going to work and you want to know this before you reassemble everything.

 

Sorry I missed this thread being updated. The high to low resistance change should be completely smooth as you move the float arm up and down. It should be about 40 ohms at the halfway point, it shouldn't have any points where it starts to increase or decrease resistance at a different rate. It sounds like you figured that out however.

The sub unit sensor (one without the pump attached) scales in resistance from 3 ohms full to 43 ohms empty which is similar to the values you posted.

Is the problem you are having that the final gallon or so of fuel has a sharp drop in ohms? Do you have a video of the DMM readout while you moved the floats from full to empty?

It's also possible for there to be a problem with the wire or harness, you can make the same resistance measurements from the combination meter harness pin 7 as you do from pin 1 on the sub harness.

 

I had the same issue for the past 2 years, I have change both sending unit and fuel pump like 3 times after infiniti recommended me to do so, I have bought a second cluster at the scrap and i was having the same problem so I went to a local garage they check wires and everyting seems fine until the head mechanic said we will change the AMPLIFIER CONTROL AIR CONDITIONER 277601NM6D and it resolved my problem. No problem after that so check that first. I bougt that part for $15 at the scrap and I was told its almost $200-300 new.