3rd Generation Specific (1993-2002) 1993-2002 Discussion including performance modifications and Technical Support Sections.
Sponsored by:

Strange fuel level readings. What to check next?

Thread Tools
 
Search this Thread
 
Old Oct 24, 2025 | 08:43 AM
  #1  
REnaissance_Sle7in's Avatar
Thread Starter
Senior Member
Tenured Member: 20 Years
Liked
Loved
Community Favorite
iTrader: (8)
 
Joined: Jan 2002
Posts: 469
Likes: 136
From: Pittsburgh, PA
Strange fuel level readings. What to check next?

Looking for some advice on a fuel level issue I've been having. Here's the TL;DR (more detailed below):

Gauge reads when the tank is full, but after about 30 or so miles, it drops rapidly. By about 50 miles, it reads at empty (not past) and the low level light is constantly on.
1. Cluster repaired by Michael Gagne
2. Took out the old level sender and ohmed it out, was slightly out of spec - discarded
3. New fuel level sender from A.I. Designs (double checked before putting it in, and it tested good)
4. Hooked up just the level sender and watched the gauge as I moved the arm. All looked good. Read full, half, empty. (the low level light was on, but I didn't think anything of it because..well the sensor was in open air).
5. Connected the pump and dropped everything in, buttoned it up and everything seemed ok. Gauge was reading full, then quickly went to E after just a bit of driving.
6. Cleaned all grounds I'm aware of that are connected to the level/cluster (on the hanger and under the UIM) - No change

The long-winded version:
Ever since I got my car running the fuel gauge had been a bit off. It would read from full to about half tank, then would drop well below E, but it would eventually come back at about a 1/4 tank (though, I never quite trusted the reading after half). I do not believe this is a cluster issue, because it had been gone through and repaired by Michael Gagne. My research has shown me that when a level gauge reads beyond F or E, that it is generally a grounding issue. I took the level sender out of the car to bench test, and it did test slightly out of spec, as per the FSM. So, I got a new one from A.I. Designs and bench tested that one just to be sure, and it tested within spec of the FSM. One thing to note about this level sender is, it did not come with a plug, so I got pins and the plug from Corsa Technique. I plugged the new level sender in and watched the gauge as I moved the level arm (keeping the fuel pump unplugged). The gauge swept through the range just fine as I moved the arm. All seemed good to button everything up. So, I plugged in the fuel pump, connected all my lines, connected the bulk head (sidebar for more info: The fuel pump was rewired directly to the battery by IRP when they had the car, but it does still have the stock bulk-head), and off I went. Filled up the car, and the gauge was reading full (the low level light was on which was annoying, but I figured I'd dive into that more later). However, while driving to a cars and coffee, the gauge began to drop rapidly (certainly much faster than it would with the old level sender) and by the time I got back home from CnC, the gauge was reading E (not passed E now, so I don't believe it to be a grounding issue). I pulled it back out to check my connections at the plug (the new one from Corsa Technique) and everything seemed seated well, and the wires/pins didn't move when I tugged on the wires, so I seemed to have connected the new pins correctly. Right now, I'm trying to figure out a way to keep the fuel pump connected, but lift the hanger up enough to move the arm to see if the gauge will move. I have the EGI fuse pulled to kill the pump, but can't get anything in there to move the float arm.

The part I can't wrap my head around is, why does the level sender work with the fuel pump unplugged, but not when it is plugged in. To me, that would rule out: a bad sender, and a grounding issue. But I'm at a loss as to where to look next, so any advice would be appreciated.
Reply
Old Oct 24, 2025 | 11:01 AM
  #2  
Pete_89T2's Avatar
Rotorhead for life
Tenured Member: 20 Years
Liked
Loved
Community Favorite
iTrader: (4)
 
Joined: Mar 2005
Posts: 2,176
Likes: 1,252
From: Elkton, MD
Originally Posted by REnaissance_Sle7in
Looking for some advice on a fuel level issue I've been having. Here's the TL;DR (more detailed below):

Gauge reads when the tank is full, but after about 30 or so miles, it drops rapidly. By about 50 miles, it reads at empty (not past) and the low level light is constantly on.
1. Cluster repaired by Michael Gagne
2. Took out the old level sender and ohmed it out, was slightly out of spec - discarded
3. New fuel level sender from A.I. Designs (double checked before putting it in, and it tested good)
4. Hooked up just the level sender and watched the gauge as I moved the arm. All looked good. Read full, half, empty. (the low level light was on, but I didn't think anything of it because..well the sensor was in open air).
5. Connected the pump and dropped everything in, buttoned it up and everything seemed ok. Gauge was reading full, then quickly went to E after just a bit of driving.
6. Cleaned all grounds I'm aware of that are connected to the level/cluster (on the hanger and under the UIM) - No change

The long-winded version:
Ever since I got my car running the fuel gauge had been a bit off. It would read from full to about half tank, then would drop well below E, but it would eventually come back at about a 1/4 tank (though, I never quite trusted the reading after half). I do not believe this is a cluster issue, because it had been gone through and repaired by Michael Gagne. My research has shown me that when a level gauge reads beyond F or E, that it is generally a grounding issue. I took the level sender out of the car to bench test, and it did test slightly out of spec, as per the FSM. So, I got a new one from A.I. Designs and bench tested that one just to be sure, and it tested within spec of the FSM. One thing to note about this level sender is, it did not come with a plug, so I got pins and the plug from Corsa Technique. I plugged the new level sender in and watched the gauge as I moved the level arm (keeping the fuel pump unplugged). The gauge swept through the range just fine as I moved the arm. All seemed good to button everything up. So, I plugged in the fuel pump, connected all my lines, connected the bulk head (sidebar for more info: The fuel pump was rewired directly to the battery by IRP when they had the car, but it does still have the stock bulk-head), and off I went. Filled up the car, and the gauge was reading full (the low level light was on which was annoying, but I figured I'd dive into that more later). However, while driving to a cars and coffee, the gauge began to drop rapidly (certainly much faster than it would with the old level sender) and by the time I got back home from CnC, the gauge was reading E (not passed E now, so I don't believe it to be a grounding issue). I pulled it back out to check my connections at the plug (the new one from Corsa Technique) and everything seemed seated well, and the wires/pins didn't move when I tugged on the wires, so I seemed to have connected the new pins correctly. Right now, I'm trying to figure out a way to keep the fuel pump connected, but lift the hanger up enough to move the arm to see if the gauge will move. I have the EGI fuse pulled to kill the pump, but can't get anything in there to move the float arm.

The part I can't wrap my head around is, why does the level sender work with the fuel pump unplugged, but not when it is plugged in. To me, that would rule out: a bad sender, and a grounding issue. But I'm at a loss as to where to look next, so any advice would be appreciated.
Just an educated guess at this point, but if you're still using the OEM bulkhead connector (B1-06 in the FSM diagram) with the rewired (and an upgraded/higher capacity?) fuel pump, it's possible the fuel pump pins in B1-06 may have overheated, due to the extra current draw of an upgraded fuel pump. That could melt the insulation on B1-06 and cause shorts between the fuel pump and the fuel gauge/low fuel light circuit terminals. Hence that could explain why you only see the problem when the fuel pump is plugged in. Give the B1-06 connector a good visual inspection, look for signs of melted/deformed plastic around the pins on both ends.
Reply
Old Oct 24, 2025 | 12:32 PM
  #3  
REnaissance_Sle7in's Avatar
Thread Starter
Senior Member
Tenured Member: 20 Years
Liked
Loved
Community Favorite
iTrader: (8)
 
Joined: Jan 2002
Posts: 469
Likes: 136
From: Pittsburgh, PA
Thanks Pete. Yes, it is a larger pump. I did check the bulkhead, and it doesn't appear to be burnt or anything at the moment. Obviously not something I want to rely on for the long-term, but I'll probably look into over the winter. Here are some pics just in case I'm missing something


Reply
Old Oct 24, 2025 | 03:11 PM
  #4  
Cgotto6's Avatar
Rotary Freak
Tenured Member: 20 Years
iTrader: (8)
 
Joined: Nov 2003
Posts: 1,963
Likes: 29
From: Maltby, Washington
It might just be the way the picture is making the connector look but the female ports look like the metal contacts inside are a bit spead/loose. It may help to gently persuade those tighter with a small awl or similar tool. You could also depin the connector, and pinch them tighter with pliers if your familiar with disassembling connectors.
Reply
Old Oct 27, 2025 | 08:07 PM
  #5  
REnaissance_Sle7in's Avatar
Thread Starter
Senior Member
Tenured Member: 20 Years
Liked
Loved
Community Favorite
iTrader: (8)
 
Joined: Jan 2002
Posts: 469
Likes: 136
From: Pittsburgh, PA
So, I ended up finding the cause of the major issue.

I ended up pulling everything out again, and was going to try and recreate the situation from before (level working with the pump unhooked) when I noticed a strange shadow on the float. Upon further inspection, the float had filled with fuel.

Not sure if you can really see in this photo, but when I squeezed it, I could also hear air hissing out. I think this explains why it seemed to work at first, then slowly went to E after filling up. Also why it never went past E like it did before with the bad sensor.

Luckily, I hadn't thrown the original away, so I took the float off the OEM one, and put it on the aftermarket one. Level sensor seems to be working as it should now.

Unfortunately, thr fuel light is still staying illuminated, so that is something I'll have to track down this winter. But this at least got me comfortable enough to take the car to autocross last weekend, and hopefully a track day next week before the season is over.
Reply
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
JerryLH3
2nd Generation Specific (1986-1992)
2
May 3, 2021 06:18 AM
rx7f4n4tic
3rd Generation Specific (1993-2002)
5
May 29, 2006 05:26 PM
_raven2510_
1st Generation Specific (1979-1985)
2
May 16, 2006 11:20 AM
MaczPayne
2nd Generation Specific (1986-1992)
1
Jan 8, 2006 07:09 PM
shawnk
3rd Generation Specific (1993-2002)
2
Feb 2, 2003 11:46 PM




All times are GMT -5. The time now is 12:34 AM.