Integra4lyfe069

1996 honda civic:

alright ron heres todays update so far. (sorry i moved this from another post to here so it would be properly named.

I took out the seats again.

took off the access plate and unpluged the 25 fuse to see if it was blown and it was fine.

put the fuse back in and turned car to on(ii) nothing happened the fuel guage is still FULL.

I then unplugged the 3 pin connector and turned the car to on(ii) hoping it would goto empty but it stayed FULL.

I then unplugged all other sensors and turned car to ON(ii) and STILL full.

I then removed the fuel sender unit from the car and proceeded to do tests on the float numbers.

ALL numbers when checked with the multi meter on the continuity setting at 200 like u showed me in my picture recently returned good, so the fuel sender unit is fine.

I put it back in and retorqued the screws and pluged in the connector to the fuel pump and the GREY 5pin connector on the metal shaft mounted on the top of the fuel sender unit.

I then with the car off tested the 3 pin connector harness with the wires that run to the guage and with the car off it returned 111 , with the car on it wouldnt read anything. also checked ito n the volt setting and still nothing.

checked fuse still not blown.

put everything back together and seats back and turned car on and drove it abit, still same FULL.

Only thing to do but retest the blk/ylw///blk wire with the correct settings is to jump the two wires with a jumper wire and see if the guage works then, but problem is it says everywhere to turn the guage off before it reaches the max but, problem is its already on full so if i jump it wont it hurt it ?

ALL pics here. (sorry theres alot for clarrifycation ) >.<













this one is the 3pin one were working on

my guess is the wires have a short in them but i took up most of the wires from the floor and pannels and they look in mint condition.
I wasnt sure if i was getting a correct reading from the connector or not so i snipped the three wires blk, yel/blk, green/yel , taped up the green yellow and tried touching the blk/yel wires even with the guage reading full andn othing happened with car on or off, and i even tried testing them again wraped around the probes of the mete and still nothing, only reads 111with car off.

Is it possible that my guage is just shot ? atm im unaware if i can jump it becasue its on full, let me know and im sure that test will determin if its gone or not because weve determined the sender unit is fine.

i also measured the resistance when it was put back in and i knew i had roughly a quartertank of gas or little more, but under half for sure, and the resistance read 51 so its working, the fuel sender unit.

seems im not getting any read from the damn 3 pin wires.

90Rexx

DUDE... wow what a waste of tons of time and work... you need a guage cluster... common on 96-00 clusters... gas guage sticks on full... get a new cluster

Former User

I responded in your original thread.

chrisfromsrq

have you played with your cluster? same thing happened to mine. check the little copper wires in the fuel gauge motor to make sure they are all intact to the screws that hold it to the cluster. if one is broken either re solder it or get a new motor for the fuel gauge

Integra4lyfe069

anyway to test an integra cluster in the civic ?

EKSiR99

having the same problem but does not get stuck on full but every time i turn on my parking light it starts to go up

Former User

Here's a way to test whether the gas gauge or cluster may be the problem. Remove the cluster (see diagram) and unplug connector B from the rear of the cluster (see second diagram). Turn the key to ON(II). The needle of the gas gauge should fall to Empty. If not, the gas gauge or the cluster itself may be bad.

90Rexx

good info! ^^^


should be put in the FAQ section!

Integra4lyfe069

alright with that said when i went out today to goto my shop where i do my work with the car.
Imediatly the fuel guage started going from full to the correct reading of gas for like minutes then it raised to full again like i mentioned it randomly does before.

I know the guage is correct reading because yesturday when i etsted the sender unit for readings it worked perfectly and when i put it back in the reading said 51 so that means just over quarter tank, today when i notied the guage working for short term i measured the resistance at the sender unit again and it was 81, so that means not empty but quarter or tank or lil less, which is where the guagewas, not empty but above, but just lil less then quarter.

So when i got to the shop I tested the battery and the battery tested perffectly 13v with ur settings..well 12.8 volt accually

then i went to work and retook out my backseats and acces plate etc unplugged the connector and tested in the harness., numbers showed up 12.4

with that said as u may know from earlier i had snipped these wires so i figured i was getting a wrong reading due to i soldered them back and tapped them up, so i snipped them again removed all garbage ends restripped the ends and tested on the wires itself...

got the same numbers 12.4

so i resoldered the wires and taped them again and retested the harness again and got 12.4 again.

so i plugged everything back in and put my seats back again.

Now I'm here with these conclusions.

->The FUEL SENDER UNIT is working. TESTS proves this.
->The Fuel Guage is NOT broken because if it was it wouldnt read correctly randomly from time to time.
->The yel/blk and blk wires show up 12.4 when tested for voltage when using the blk wire as ground. (this is wrong it should read 5-8 volts)

*Only thing i can see is testing the yel/blk wire with something else as ground to see which numbers the yel/blk wire returns?

and

*somehow figure out if the black wire is bad and not grounded properly?

which numbers would i be looking for if i went this route?

OR somewhere from the harness at the sender unit to the guage theres a short in one of them. do i need to tear everything up and replace the wire or try and fix where its broken, assuming it is broken... ?

what else could it be?

I unplugged the b connector and the guage didint drop, i unplugged C connector and the guage didint drop.
i unplugged all individual connectors 1 by one and turned car on but nothing dropped.

i unplugged both connectors that look like b and c cause i dont have A connector on mine. and the guage still didint drop.
heres a pic of my cluster. it looks a lil slightly diff.


stuck on full


back


fuel guage area on back


top left no connector A

deschlong

iTrader: (8)

This is the most enormous leap in logic in all of your postings. It really surprises me because you have already gone to such lengths to test every other piece of electrical equipment associated with the problem and ruled it out, and are clearly determined to get to the bottom of the problem.

Logic Fail: Would a PROPERLY WORKING fuel gauge "read correctly randomly from time to time"? No. If it did, then we wouldn't be having this discussion, and you wouldn't be going to great pains trying to dissect the problem.

You have been given good advice in this thread: It is highly, highly probable that it is a problem with the fuel level unit on the gauge cluster. Most likely a wire has come loose (or off). The "randomly from time to time" correct measurement of fuel likely originates with the momentary reconnection of the tiny wires that make up the fuel level gauge unit.

Simplest and fastest way to test is to swap in a known good cluster, or test one at a wreckers. These are $40 or less from them. If you're still skeptical and don't want to commit right away at the wreckers, just tell them you want to test it in your car first right there in their parking lot before you buy.

To understand really what's wrong, first off you can't see anything while it's still attached to the cluster. You have to take apart the cluster and undo the six screws for the fuel/temp gauge, three of which are seen in your picture here.


Examine the teeny tiny wires that make up the gauge, especially where they are soldered or connected. Likely, one of them is loose or broken. Swap in your known good, but FFS, be careful, as I have written before:

Alternatively, you could try reconnecting/soldering the wires back together, but as I stated above, this is really really really difficult. I've done it, but this was necessary because the nearest replacement Canadian fuel/temp cluster was thousands of kilometres away, it was February in the Yukon and I happened to have a soldering gun, some tin and tin snips to jerry-rig a connection, boat loads of time in the evenings, an equivalent amount of patience, and no GF nearby to complain about the mess(es) I made. Good luck.

Integra4lyfe069

yeh you are right, i took apart my cluster before i read your post and i did it totaly wrong rofl..

well its now safe to say i need a new fuel/temp guage.

i didint think taking out the 6 screws in the back would release the entire thing so i popped off the guage needles and it seems they dont pop off, their cracked off now,

then i realised taking out the 6screws did the trick LOL.

but i also noticed when i got it all taken out, that everything is fine, all those little ***** hair wires were intact and everything was fine.

ima run to the wreckers next week and pickup a cluster.





wanna se the most funniest ****ing **** ive ever seen in my life?

the previous owner did this to the cluster.

i always thought it was a VTEC light that was always on and i was wondering wtf why doesnt any vtec car have this.


then i saw this on the back LOL @ TAPE + computer printout on blue paper. LOLOLOL

deschlong

iTrader: (8)

Good. Well, my warning on how to reassemble it still applies. Be really careful.

Former User

Well, you clearly must replace the cluster. Don't worry, even if you had not damaged the cluster, you still would have needed to replace it, given the outcome of the test described above. This, together with the higher than expected voltage at the sending unit connector, suggest a short in the cluster, though you might also want to test for a short in the Yel/Blk wire running between the cluster connector and the sending unit connector.

deschlong

iTrader: (8)

Hey - with a new cluster you'll get rid of that cruddy mod, too. Good work.

Integra4lyfe069

alright, wanna tell me how i go about this, testing ONE wire only ? with the multi meter

where do i put the tongs //settings etc

i tried the yellow/black wire at the with the meter on 20 at the cluster cluster like i did at the harness by the sender unit and i think the numbers were 2 i beleive.

Former User

Yeah, pretty simple. Do the following test on terminal B10 (Yel/Blk wire) of cluster connector B with the cluster removed and all connectors unplugged.

--Turn the key to ON(II) and measure voltage from the B10 connector terminal to dash frame ground. The multimeter is set to measure DC volts at the 20V range again. You should basically measure 0V. Also inspect pin B10 and other pins in connector B for damage. If all is fine, then replacing the current cluster should do the trick.

Not sure, but this sounds similar to the test I suggest above. Try my test above with the key either in ON(II) or in the OFF position to compare the voltage readings.

Integra4lyfe069

you mean test them at the blue wire harness right? cause all is on the back of the guage is the rubber green matt or w/e

Former User

Test the B10 terminal in wire harness connector B that plugs into the cluster.

Integra4lyfe069

okay so it should read zero?

what if i get a number i think i got 2 last time.

also should i check the black wire also? andh ow would i do that.

Former User

Yes. Dash metal frame would be a good ground.

~0V -- You just want to verify that the unexpected high voltage (12V) measured at the fuel sending unit connector was not caused by a hot wire shorting with the sending unit power wire. I think this is unlikely, but it should be tested prior to installing a new cluster.

Yes

90Rexx

god this thread is getting annoying lol...replace the damn cluster already!

and LOL at the vtec light, but it might have been used for diagnostic purposes.... so dont call the previous owner a ricer JUST YET>>. lolz