Repro sending unit ohm is linear correct?

[Dino]

I thought our original sending units had a non-linear potentiometer whereas the repro units are linear.  Is this wrong?  Non linear sending units are typically paired to a fuel gauge that has uneven markings yet our fuel gauges have the 1/2 mark right in the middle. 

Can somone shed some light on this please? 

[Pete in NH]

Hi Dino,

If you look at the test spec. for the fuel and other gauges you will see 10 ohms at the sending unit equals full scale, 24 ohms equals half and 73 ohms equals low. Given that the gauges are marked in a linear manner, that tells me the sending units are non-linear. The half to full range looks sort of linear, but the half to low isn't at all. If those re-pro fuel sending units are linear they may read okay in the half to full range but not below half.

[Ghoste]

Which would explain many of the complaints about their accuracy..

[Charger-Bodie]

I like to graft an original sender to a repop pick up tube. Looks new and lasts (if stainles steel).

[Dino]

I was just rereading your thread about combining old and new and that is certainly an option seeing I have the original 5/16 sender in my tank.

I'll have the new sender today so I'll check the readings and post back.  I bought the one from Right Stuff Detailing.

I thought I read on this site somehwere that the repro senders measure 41 ohm as 1/2 while the gauge would measure in the 25 range, as you said.  Yet they would not match up.

Between converting an aftermarket gauge and having two sending units to fabricate one out off, I should be able to find a way to get this done.  I really really want the fuel gauge to work again because it sucks not knowing what's left in the tank.   

[tan top]

I like to graft an original sender to a repop pick up tube. Looks new and lasts (if stainles steel).

after messing about with a couple of repops   , that what i done  in the end  , put my original sender on a repop pick up ,

[Dino]

Quick thought...

If the fuel gauge needs 0 ohm at full, 23 in the middle and 73 empty and the sending unit reads the same full and empty but middle is 41...can't we just put a resistor in the sending unit that brings 41 ohm down to 23?  I know by using the oem resistor you get the same effect but can't it be done this way as well?  For those that don't have the oem unit.

[Pete in NH]

Hi Dino,

If only things were that easy! To reduce that 41 ohm mid range resistance we would have to add another resistor across it.That is the way you reduce an already existing resistance. A little math will help us out here.

The formula for resistors in parallel is: R= R1 x R2 / R1 + R2. So, If we put a 56 ohm resistor across the sending unit, at the mid point we get 41 x 56 / 41 +56  or 23.7 ohms. Wow! that looks great, just what we need.

At the full point we get 10 X 56 / 10 +56 or 8.5 ohms, okay not too bad off by 15%.

The real trouble comes at empty when we have 73 x 56 / 73 + 56 or 32.6 ohms, our fuel gauge would tell us we have something around 3/8 of a tank when we were really empty or close to it.

The real solution to the problem if the re-pro sending units are really wound in a linear way is to wind them in the correct nonlinear way. Those resistors are wound out of resistance wire on a piece of insulating materail and a wiper arm controlled by the float slides along the wire. The spacing between the turns of the resistance wire is what is important. If you wind them with equal spacing between the turns you get a linear unit. If you wind them with a bigger spacing between the full and half tank range and the gradually decrease the spacing between the half and empty point you get the required nonlinear unit. It's very like a matter of the people making the re-pro units not using the proper spacing when they wind the resistance units. A little time spent getting the correct winding taper would fix the problem. It's not rocket science all they had to do was copy the original Chrysler units.

[maxwellwedge]

The problem is conveying that in Cantonese....

[Dino]

Hi Dino,

If only things were that easy! To reduce that 41 ohm mid range resistance we would have to add another resistor across it.That is the way you reduce an already existing resistance. A little math will help us out here.

The formula for resistors in parallel is: R= R1 x R2 / R1 + R2. So, If we put a 56 ohm resistor across the sending unit, at the mid point we get 41 x 56 / 41 +56  or 23.7 ohms. Wow! that looks great, just what we need.

At the full point we get 10 X 56 / 10 +56 or 8.5 ohms, okay not too bad off by 15%.

The real trouble comes at empty when we have 73 x 56 / 73 + 56 or 32.6 ohms, our fuel gauge would tell us we have something around 3/8 of a tank when we were really empty or close to it.

The real solution to the problem if the re-pro sending units are really wound in a linear way is to wind them in the correct nonlinear way. Those resistors are wound out of resistance wire on a piece of insulating materail and a wiper arm controlled by the float slides along the wire. The spacing between the turns of the resistance wire is what is important. If you wind them with equal spacing between the turns you get a linear unit. If you wind them with a bigger spacing between the full and half tank range and the gradually decrease the spacing between the half and empty point you get the required nonlinear unit. It's very like a matter of the people making the re-pro units not using the proper spacing when they wind the resistance units. A little time spent getting the correct winding taper would fix the problem. It's not rocket science all they had to do was copy the original Chrysler units.

Thanks for the explanation Pete!   

Looks like doing 1HotDaytona's mod is the way to go here, unless of course I use a linear fuel gauge but likely not, not much out there I can work with.

I have the sending unit at home so I'll measure it tonight to see what's going on.

The problem is conveying that in Cantonese....

Yeah that might be a problem.   

[Pete in NH]

The problem is conveying that in Cantonese....

Yes, that's very likely exactly the problem!

[Dino]

I can now confirm that the 3/8 sending unit sold by Right Stuff Detailing is linear.  41 ohm in the middle position.

[Pete in NH]

Well that's not so good! Did they get the end points right? Full should be 10 ohms and empty should be 73 ohms. I wonder what Right Stuff would have to say about the issue?

[Dino]

Well that's not so good! Did they get the end points right? Full should be 10 ohms and empty should be 73 ohms. I wonder what Right Stuff would have to say about the issue?

I'll have to redo as I forgot but it was close.  Full and empty would owrk fine but middle will be way off.  Very strange that the oem senders are non linear yet the fuel gauge has linear markings. 

I think I have a spare fuel gauge left, if not I'll find one somewhere, so I'm going to try and get it to work with the repro sending unit.  If it won't work then I'll transplant the old sender's resistors.

Now all I have to do is figure out the exact midpoint on both, I hate doing this by adding fuel to the tank and would rather do it on a bench.

[maxwellwedge]

I think the majority of issues have been that the re-pops show 3/4 when the tank is full. Makes me nervous about what is truly empty. I am sure there are other issues as well.

[Dino]

I think the majority of issues have been that the re-pops show 3/4 when the tank is full. Makes me nervous about what is truly empty. I am sure there are other issues as well.

I guess when you don't drive much then it may be acceptable to not show true full but it would be nice to know when you need to fill up again!  Driving the car as much as I do I would like to have it fairly accurate though.  And I say fairly because I know that 10 minutes after filling up, the needle probably should be around the 3/4 mark.   

[maxwellwedge]

Yes - It is important to know what's happening. It is unbelievable that they can't get this stuff right!

[Dino]

Yes - It is important to know what's happening. It is unbelievable that they can't get this stuff right!

I don't get it either, all they had to do was use a non linear potentiometer instead of linear and voila, accurate reading every time.  Mind boggling.  I'll bet you the Ford and GM guys don't have this problem.

[Pete in NH]

Hi Dino,

To answer your question about nonlinear sending units and linear markings on the gauge-

The gauge meter movements are a thermal type which move the needle by changing the current flow through the gauge to heat and measuring the heat produced. This type of gauge movement is cheap and relatively rugged, although slower to respond to changes. Being a thermal mechanism it is non linear.

Again a little math explains things. To get the heat produced in watts you need to know the gauge resistance and the amount of current flow. The power in watts = (current value x current value) x resistance . because its the current value times the current value, which is known as a square function things get nonlinear fast. For example if the current value is 1, you get 1 x 1 =1, if it's 2 you get 2 x 2 = 4,  3 would be = to 9. Note that while the current value is only going up by one in each case the result is rising rapidly. So, the heat produced within the gauge is very nonlinear. If the sender were linear this would crowd all the readings to one end of the gauge and make it more difficult to read at a glance. So, to make the gauge read in a linear manner so it would be easier to read the sender must be made nonlinear to account for the characteristics of the thermal gauge.

[Dino]

Hi Dino,

To answer your question about nonlinear sending units and linear markings on the gauge-

The gauge meter movements are a thermal type which move the needle by changing the current flow through the gauge to heat and measuring the heat produced. This type of gauge movement is cheap and relatively rugged, although slower to respond to changes. Being a thermal mechanism it is non linear.

Again a little math explains things. To get the heat produced in watts you need to know the gauge resistance and the amount of current flow. The power in watts = (current value x current value) x resistance . because its the current value times the current value, which is known as a square function things get nonlinear fast. For example if the current value is 1, you get 1 x 1 =1, if it's 2 you get 2 x 2 = 4,  3 would be = to 9. Note that while the current value is only going up by one in each case the result is rising rapidly. So, the heat produced within the gauge is very nonlinear. If the sender were linear this would crowd all the readings to one end of the gauge and make it more difficult to read at a glance. So, to make the gauge read in a linear manner so it would be easier to read the sender must be made nonlinear to account for the characteristics of the thermal gauge.

Pete that makes perfect sense, thanks for that explanation!