Fuel Gauge Calibration on the Cluster circuit board - Chevy Impala SS Forum
Reply
 
LinkBack Thread Tools Display Modes
post #1 of 11 (permalink) Old 04-21-2017, 11:45 PM Thread Starter
Senior Member
 
Z09B4U's Avatar
 
Join Date: Jul 2015
Location: Canada
Posts: 844
 
Fuel Gauge Calibration on the Cluster circuit board

Like many of you I was quickly frustrated by the gas gauge. Other vehicles also do not have linear response but the Caprice sticks on full then goes to the half then drops to zero like a rock. What this means is that the bottom half of the gauge is really one third of the tank. I have talked with other Chevy owners and we are not alone with this issue.


The short version of this post is that I claim I can recalibrate the fuel gauge without dropping the tank. I also claim I can fix the needle stuck on full issue.


First a short message. If you do not understand any repair or modification on this forum do not start it. You are responsible for your actions and the results.


Second if you have questions about how to calculate resistance values. Google ohm's law, parallel resistance, series resistance and resistance color codes. You can find on-line calculators as well. For questions on gas tanks, fuel senders, and fuel pumps there are other threads for these or start a new one please. You need soldering skills and the ability to understand circuit diagrams. The resistors needed for this mod can be found on: fleabay, maybe Radio Shack, industrial electronic supplies, hobbie shops, and old electronics.




What I hope to do with this post is show you some information so you can make an informed decision on whether or not you should replace or clean your tank sensor. I also have a method of cheating to change the way the gauge reads.


The best place to start is the hardest work. Drop the tank and replace the sender. Have a look at this link:


https://www.impalassforum.com/vBullet...stall-how.html


If you are having fuel pump problems you could clean the sender or replace it while you are doing the pump. This link covers cleaning and the practice of soldering all the in tank connections:


https://www.impalassforum.com/vBullet...nding-unit.htm



The only issue with soldering is trying to get warranty if the new pump or sender fails.


My suggestions to the tank work are these: Have a look at the socket on the car side. If you plug your new tank harness plug into a 20+ year old plug that is not weather tight anymore you will screw up your new tank harness with salt, dirt, and old grease. It may also leak in water. If the car end is not perfect replacing it may save you headaches later.


The grounding in the trunk topic is good for a separate rant. Stock from the factory this single ground wire carries 16.22 Amps when the tail, backup and brake lights are on. Then add the fuel pump amperage to the same 14 ga. ground wire! From an engineering standpoint the first thing to do at the tank is supply a separate 14 ga. ground to the fuel pump/sending unit plug. A separate 14 ga. ground to each of the Tail light lens is needed.

Z09B4U
Z09B4U is offline  
Sponsored Links
Advertisement
 
post #2 of 11 (permalink) Old 04-22-2017, 12:01 AM Thread Starter
Senior Member
 
Z09B4U's Avatar
 
Join Date: Jul 2015
Location: Canada
Posts: 844
 
Before dropping the tank to improve the accuracy of the gas gauge you should look at some of the facts about what Chevy intended when they built the system. By looking over this data you can make an informed choice on whether you should drop the tank or just leave everything alone.

To explain my modification better lets look at the stock (theoretical) design. Chevy claims that a full tank's resistance is 88 ohms and an empty tank is 0 ohms. Chevy then uses 90 ohms full and 0 ohms empty in the trouble shooting charts and the circuit diagrams?

To see if cleaning or replacing the sender is necessary lets look at the data for the fuel gauge. I could not make accurate measurements under 2 ohms as the rheostat I used would change by more than 1 ohm when moved. The gas gauge's 22 ga. wire adds 0.5 ohm, the three connectors and the connection between the plug-in fuel gauge and the cluster's circuit board also add some resistance, and the ground connections and wires add some resistance but I did not include this in my tests. The 100 ohm rheostat was connected directly to gauge with test clips.

These three samples are all non-original (replacements) to the cars I removed them from. All are 9C1 style. You would think the gauges would be the same, everyone says so. The 94 gauges were slow to move taking seconds to move from empty to full. The 96 moved so fast it made an impact noise at the minimum and maximum points. On the other hand it was almost impossible to get an resistance measurement below empty on the 94 gauges as the needle stopped just under the empty line there was more space on the 96. I now know why some owners report the gas sloshing in the tank moving the gauge needle. The 96 would twitch if I looked at it wrong.

Min |Empty| ¼ | ½ | ¾ | Full |Max
3 | 5.0 | 30 | 43 | 59 | 78 | 99 1996 week 3 NOS
0 | 3.5 | 24 | 40 | 57 | 76 | 85 1994 week19 NOS
0 | 4.5 | 26 | 43 | 59 | 77 | 88 1994 week 9 GM re manufactured
--| 1.0 | 26 | 50 | 60 | 88 | xx Chart (8A-82-11) 94 FSM
| 0 | | | | 90 | Trouble shooting charts and circuit diagram

These reading tell us what the owners manual should have said:

When the fuel gauge system is working correctly there is gas in the tank when the needle is pointing at Empty, and it is normal for the needle to go past the Full mark.
Most other vehicles do read over the full line and under Empty.

Lets look at some real world numbers from the tank senders:

Empty | Full
0.3 | 89 New FLS-1072 (courtesy Jimpala95)
0.8 | 93 Cleaned (courtesy Jimpala95)
0.6 | 96 Cleaned and modified (courtesy sherlock9c1)
4.5 | 119 Old dirty (courtesy Jimpala95)
11.4 | 130 Old dirty (courtesy sherlock9c1)

So Jimpala95's dirty gauge sender would be hard to spot as malfunctioning. Sherlock9C1's would let the tank go dry before showing empty, and stay at the maximum point too long. A new sender would pull all of my sample gauges under empty and push them over the full mark.

For me the easy test location is C400 driver's side trunk harness. When unplugged an ohm meter can measure the tank sender (black-violet). What I get from the gauge data chart is that if you measure 3 ohms or less empty or you can see that with the tank float down you read below empty and you measure 88-100 ohms full the sender is OK. In my opinion it would be a wast of time and money to work on the sender if your range is 3-100 ohms. If you want perfect go ahead and pull the tank.

See the Thumbnail for diagrams of factory Full and Empty.

Ask some one to make something and a welder will use steel and a carpenter will use wood. I have an electronics background and would rather fire up the soldering iron than drop the tank to clean or replace the sensor.

So if we want to stay out of the tank how do we change what the gauge shows? The “fuel gauge calibrator” has been around for a long time. It does what it says. By adding a parallel resistance to the fuel sender's resistance we can lower the resistance that the fuel gauge reads. This $2 part will allow you to bring the needle from pined on over full to a lower point ie just over the full mark (not pinned to the over full). This simple add on does not help with the accuracy or linearity of the gauge. The “fuel gauge calibrator” is an adjustable resistor. What I would do is get a “ten turn trimmer (potentiometer) from fleabay. The exact product is determined by your full tank resistance. You could get a 5 Kohm (5000 ohm) trimmer but it would be much harder to calibrate. A small turn of the screw would make a big change. With the correct sized trimmer you will turn the screw several times to see a change.

Here is a quick chart to show the trimmer selection process:

Sender |Trim Ohm |Calibrator type (ohms)
100 | 2000 | 2K
105 | 925 | 1K
110 | 650 | 1K
130 | 350 | 500 This will pin the gauge at 95 ohms.

If you find the on-line instructions for the commercial kit it recommends patching in the trimmer in the drivers side wire harness at the kick panel. If you do not like taking off the kick panel and working on the floor of the car you could patch the trimmer to the wires in the trunk. It is harder to adjust from the trunk but the installation is way quicker. You could jam the wires into the C400 connector (black-violet) if you just want to see what it does. I an a fan of using a normal resistor of an exact value. It is tougher than a trimmer and will not change value

What everyone wants is 23 gal full, 17.25 gal ¾, 11.5 gal ½, 5.75 ¼, and a gallon at empty to protect the fuel pump. The way to get this is to patch in a computer that is calibrated to the tank sensor and the fuel gauge. It would have to be calibrated to each individual car (minimum eight places) and would be complicated for the end user.

Since I do not care that much about the gas gauge my thoughts turned to a method that would improve the gauge, be simple to do, and of course be cheap.

The fuel gauge circuit is about balance. When the tank is empty the sensor sends full power to the “empty coil” to pull the needle away from the “full magnetic coil” When the tank is full there is little power going to the “empty coil” so the needle moves toward the “full magnetic coil”.

What if the engineers at Chevy did not care about the fuel gauge and just reused some “off the shelf” parts that were “close enough”. If they did not care they might place the axle 5/8 of an inch off center or other crazy stuff. So Empty is 9 ohms and Full is 78 ohms.

Since I do not want to pull the tank my correction method is to mess with the gauge. There is no simple way to lower the minimum resistance of the tank sensor. But what if there was a way to reduce the pull of the “Full coil”?

The way is to cut the circuit board trace to the “Full coil” and add some resistance (a “Zero resistor”)to reduce the magnetic pull of the “Full coil” What I did was break the connection between the “Full coil” and ground. You can use a dremel or a box cutter. All you are trying to do is remove enough copper to break the circuit. If you are worried about the break in the board find some nail polish and paint the cut mark. After measurements and calculations I tried a resistor. It was way out. What worked in my car with my tank sender was 2.8 ohms. An easy to find substitute would be four 10 ohms in parallel which makes 2.5 ohms. I do not think any system would work with over 5 ohms and you would not see a change at 1 ohm.

My tank also needed the “fuel gauge calibrator” so I jammed the trimmer wires in the trunk connector and “dialed it in”. I pulled the wires out and measured about 300 ohms. I soldered a 300 ohm resistor on the circuit board from the tank sensor input to ground. Now the tank is “calibrated” and can not be “die-calibrated” by the trimmer changing value do to water damage or vibration.

The thumbnails are: trimmers, “fuel gauge calibrator” and Zero resistor diagram, a picture of the back of the cluster with both resistors attached and the ground cutting point. (Ignore the purple and white wires this is a totally different modification)

This fix is targeted at tank senders that are in the 6-140 ohm range. IT WILL NOT FIX GAUGES WITH A HIGHER THAN 6 ohm EMPTY OR A SYSTEM THAT IS MALFUNCTIONING.

So what does this do for the gas gauge? First it allows a bad fuel sender to pull the gauge needle below empty. Second it corrects the needle to read just over full instead of jammed at over full. The next part is what everyone wants, the gauge needle will move from overfull to full in about 60 miles. The needle will fall faster to the half tank mark. This is because I have changed the electromagnetic properties of the gauge. I do not think it makes a difference on the last half (1/3) of the tank it still drops too fast. There are many variables in this mod. Use whatever value resistors work for your fuel system. I will experiment with the zero resistor value when the B4U is again the daily driver. I have been using this mod for two years without any problems.
Attached Thumbnails
Click image for larger version

Name:	Trimmers.jpg
Views:	136
Size:	171.5 KB
ID:	157954  

Click image for larger version

Name:	GaugeDiagram.jpg
Views:	167
Size:	180.3 KB
ID:	157962  

Click image for larger version

Name:	SenderGaugeMod.jpg
Views:	170
Size:	186.0 KB
ID:	157970  

Click image for larger version

Name:	Back1.jpg
Views:	176
Size:	163.3 KB
ID:	157978  

Click image for larger version

Name:	Back2.jpg
Views:	167
Size:	176.2 KB
ID:	157986  


Z09B4U
Z09B4U is offline  
post #3 of 11 (permalink) Old 04-22-2017, 03:47 PM
Senior Member
 
Join Date: Apr 2001
Location: Moorpark, CA
Posts: 8,441
 
Garage
FWIW several years ago I bought the NAISSO fuel gauge "fix" which has a adjustable rheostat. It has resolved the "gauge shows past full and gauge is slow for the first 1/2 to 2/3 tank then drops much quicker to empty on the last 1/3 tank" issue

The warning light does come on when needle gets to the bottom of the gauge. My estimate is there is about 1 1/2 gallons left in the tank when that light comes on....YMMV

I have my original sending unit but did replace the pump itself 10 years ago with a Walbro 255

\'96 BBB 383/T-56
BALLSS is offline  
 
post #4 of 11 (permalink) Old 04-22-2017, 07:34 PM
Senior Member
 
91ss's Avatar
 
Join Date: Dec 2008
Location: NYC
Posts: 1,351
 
Had to replace the sending unit on the 91 this past summer. It's now linear. HATE IT. In essence, the old one gives you a 'high' resolution when your tank is getting low. Now, at about the same mileage point, my pointer is in the reserve zone and panic sets in. Where before it would be at least a 1/4 or more telling me i had at least 4-5 gal.
91ss is offline  
post #5 of 11 (permalink) Old 04-22-2017, 08:07 PM
Senior Member
 
Jimpala95's Avatar
 
Join Date: Oct 2005
Location: Chicagoland
Posts: 587
 
Garage
Very nice write up and effective solution to this issue B4UZ09. Thanks for sharing.

I replaced a bad sending unit in my 94 Caprice last year.Fuel gauge stayed on full forever and would not read below half a tank. I also replaced the fuel pump as long as I was going to end up dropping the tank.

But before I dropped the tank,I wanted to confirm it was a bad sending unit. So I assembled the new pump/sending unit and tested the fuel gauge movement by plugging in the new one and slowing moving the float arm up and down.The gauge had full movement once again so I went ahead with the repair.

Then I through an ohm meter on the old sending unit and it was obvious why the gauge read the way it did. 55 to 190 lol.
The sending unit I put in was an NOS AC Delco from late 90's(my last one unfortunately).It read 0-90 deas on when bench tested prior to installing.

Thanks again appreciate you sharing your fuel gauge board resistor mod.

Jim

Last edited by Jimpala95; 08-07-2018 at 06:35 PM.
Jimpala95 is offline  
post #6 of 11 (permalink) Old 04-23-2017, 12:47 PM
Banned
 
Join Date: Oct 2009
Location: northeast
Posts: 3,437
 
Quote:
Originally Posted by 91ss View Post
Had to replace the sending unit on the 91 this past summer. It's now linear. HATE IT. In essence, the old one gives you a 'high' resolution when your tank is getting low. Now, at about the same mileage point, my pointer is in the reserve zone and panic sets in. Where before it would be at least a 1/4 or more telling me i had at least 4-5 gal.
on my 96 I did full it to the max when new @the 1/4 mark took 7 gallons..
been a long time but I think when it hits E when new it had 3 gallons left approx...

today I would never do this test . when it drops down to the 1/3 range I put in about 12 gallons . so it goes to the full or slightly under mark.. so not too much or too little in that tank so far no fuel issues all original.

adding a resistor like 300 ohms across the c400 connector wire PPL to black is cheap quick fix to show lower reading on the fuel gauge .. when this is installed fuel level indicates 1/2 tank should take about 10-12 gallons...........
j cat is offline  
post #7 of 11 (permalink) Old 04-23-2017, 03:13 PM Thread Starter
Senior Member
 
Z09B4U's Avatar
 
Join Date: Jul 2015
Location: Canada
Posts: 844
 
When the "peer reviews", criticisms, and facts pile up I hope to edit the main text.

You may have noticed I, like Jimpala95 do not recommend running under a 1/4 tank for any reason.

I mentioned C400 is a good location for two reasons: you can check the tank sender's resistance on the one side and you can do a quick gauge function check on the other.

To quick check the fuel gauge unplug C400, with the key on you should see over full.
Short out the purple and black wires to the gauge and the gauge should go below empty.
Short it with wire, plier tips, tin foil, or two keys. Any thing that can touch the copper connectors and complete the circuit.

Jaycat 300 ohms is a bad choice for most people. It will result in a gauge reading of under full if your maximum sender resistance is under 120 ohms as most will be.

What I did not express well is a basic concept of gauging and instrumentation.

Your objective is to get every movement of the sender to be a movement of the gauge needle.

If you use the "fuel gauge calibrator " to bring the needle down to the full line you remove 5% to10% of the needle's sweep (span). This results in the same information being represented by a smaller sweep. This makes every movement of the needle represent more pints of fuel.
The worst example would be if you used the "fuel gauge calibrator " to set 1/2 tank as full. You would only know full to empty with half the resolution.

The two mods get maximum sweep (span) of the needle in the gauge. This maximizes the resolution of the information the tank sender measures.

Z09B4U
Z09B4U is offline  
post #8 of 11 (permalink) Old 04-24-2017, 08:03 AM
Senior Member
 
4DoorSS's Avatar
 
Join Date: Nov 2008
Location: Beautiful Chester County, PA
Posts: 7,973
 
"Resistance Color Codes".....you brought back a very old memory from Electronics School. BBROYGBVGW....Bad Boys Rape Our Young Girls But Violet Gives Willingly.

That was a simple memory trick to keep the 10 Colors in order and how they were represented numerically. Black = 0,Brown =1, Red=2, etc.

Ahhhh....the good ole days of being in school.

BTW - Good point on the ground wire from the trunk. Next time I need to do any pump work I WILL be adding a seperate wire for the fuel pump ground and just run it in parallel to the harness from the trunk area and either splice in the new wire near the connector or crimp on a new weather pack pin to go into the connector. Crimp a new terminal on the other end and mount it in the trunk.

================================================== =============


'95 Caprice SS conversion - Bonny Buckets - 96 Column, Shifter and Console - 3.73 Gear and Herter tune - Z/28 Cluster - Blazer OH Console - Green/Pink bushings - Chromed Impala Wheels - Eibach Springs and Bilstein Shocks - Corvette Engine Dress - the list goes on and still more to do...it never ends.

Last edited by 4DoorSS; 04-24-2017 at 08:07 AM.
4DoorSS is offline  
post #9 of 11 (permalink) Old 04-25-2017, 03:24 PM
Banned
 
Join Date: Oct 2009
Location: northeast
Posts: 3,437
 
I have many resistors and I will set up a rheostat set to 300 OHMS then adjust it when tank is full. I will solder in connectors at the 2 wires . then when I get the correct reading which is different on all these but close , then install a fixed resistor across the 2 wires. since I do not go below 1/3 tank this will be more accurate than whats going on now ..

when the weather improves I will go for it..
j cat is offline  
post #10 of 11 (permalink) Old 04-25-2017, 05:42 PM
Senior Member
 
Join Date: Apr 2001
Location: Moorpark, CA
Posts: 8,441
 
Garage
Courtesy of Bradley Salemie for the common benefit of the B-body community:

If this helps anyone who is going to solder in resistors or a rheostat resistor the directions that came with my Fuel Gage Calibrator from Bradley Salemie show splicing into the smallest purple wire (there will be 3-4 purple wires depending on what year car you have) in the smaller of two taped wire bundles that are under the DS sill (you remove sill plate to access)

The ground goes to any ground screw you have available in the area you splice into this wire bundle

I have not had to adjust mine since I installed it several years ago and gauge works as it should. Doesn't stay on the full side of like 2/3 of a tank and then nose dive to empty quick like it did before and show past full mark when you fill it with gas.

\'96 BBB 383/T-56
BALLSS is offline  
Sponsored Links
Advertisement
 
Reply

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the Chevy Impala SS Forum forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in










Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page
Display Modes
Linear Mode Linear Mode



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 Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome