11 replies to this topic

[CI 0308]

I have spent the last 2 days fitting a new 90 litre Brown Davis drop tank to the hatch and whilst at it I thought id spend a bit of time calibrating the fuel gauge and sender as it has never read full.

After checking the sender resistance checks which were linear though the whole range from around 10 ohms up 80 ohms, I then lengthened the cable and sat in the drivers seat and slightly bent the arm (only the tiniest amount) so that the gauge showed full scale when at full stroke on the sender, then half way on the sender showed half on the gauge etc.

My question is why does the gauge and sender seem to work well when working from full then back to empty, but going the other direction empty up to full not so well. For instance if I put the sender up to full and leave it there it showed full on the gauge, then put the sender at half it would read half full. but if I put it down at empty and move it up to half and leave it 10 minutes it still wont read half on the gauge.

So it seems accurate from full down to empty but not so from empty up to full.

 

p.s. my dash has a new Jaycar regulator fitted and adjusted to deliver 5 volts.

[Rockoz]

There is lag built into the gauge.

If not the needle would be bouncing around all over the place when you are driving.

Perhaps the componentry in the lag circuit needs replacing.

Will be a resistor and capacitor more than likely.

Capacitors will degrade significantly with age.

[hanra]

There are no discrete components for the temp/fuel/oil/volts gauges. It is just nichrome wire wrapped around a bi-metallic strip.

Give the cluster a few taps, it will probably read ok. Remember there is a lot of vibration going through the vehicle when driving.

[S pack]

The lag in the thermal instruments is simply the time it takes for the bi-metallic strip to heat up and deflect to move the attached indicator needle.

We also need to take into consideration the constant opening and closing of the voltage regulator points (not applicable to electronic voltage regulators) which is constantly interrupting the current flow to the instruments.

 

The wires (windings) on the rheostat strip in the sender units have also been known to move back and forth slightly in contact with the contact shoe on the float arm. This condition will give variable gauge readings.

Worse case is the windings end up snapping and the sender unit needs to be replaced.

Edited by S pack, 29 December 2018 - 09:16 AM.

[Balfizar]

There could be a dampening circuit (internal to the gauge - resistor/capacitor or external in the instrument cluster)

check wiring diagram and inspect gauge. (for capacitor/resistor)

 

( I don't know torana gauges or if or when GMH moved to bi-metal strip gauges but if it is a bi-metal strip gauge the the following applies to the gauge)

If not (and more than likely) the gauge is a bi-metal strip and the gauge works on heat generated by current flow which is regulated by the variable resistor

(float moving a contact along the resistance winding) the damping effect is a product of the lag of the bi-metal strip how long it takes to heat up and deflect the needle.

That how you get a gauge that does not bounce up and down with every corner or bump.

 

Slide to resistor contact could be your problem.  The gauge is only ever full once per tank and constantly moving down from that point.

This meant there is a lot more "Bedding in" of the contact pattern in the toward empty direction.

The solution is to "Delicately" bend the contact spring to give more contact pressure between the slide and resistor.

 

The next problem and one overlooked by GMH originally is ranging.  Where the angular range of the float does not match the physical range of the drop tank.

e.g.  float is bottomed out on its travel and there is 20 litres still in the tank and or float is topped out (Submerged)and there is still 20 litres above the float.

Thats a tricky one to fix. Best to bend the float arm to contact the bottom of the tank  and ignore the full reading for the first 100 kms.

[VDO]

There is no "lag circuit" in these gauges.

 

The typically slow reacting bi-metal gauge movement which can be overly slow (faulty) depending on a couple of situations.

 

The voltage regulator should ideally be set at somewhere between 5.1 and 5.5 volts to compensate for the load when all gauges are working at the same time.

 

The windings on the bi-metal could be old or burnt.

 

The theory at play here with normal conditions is that more resistance (low 73 ohms) will incite slower heating of the windings due to lower voltage delivered and the less resistance (10 ohms full) will incite faster heating of the windings due to more voltage getting to them..

[VDO]

There could be a dampening circuit (internal to the gauge - resistor/capacitor or external in the instrument cluster)

check wiring diagram and inspect gauge. (for capacitor/resistor)

 

( I don't know torana gauges or if or when GMH moved to bi-metal strip gauges but if it is a bi-metal strip gauge the the following applies to the gauge)

If not (and more than likely) the gauge is a bi-metal strip and the gauge works on heat generated by current flow which is regulated by the variable resistor

(float moving a contact along the resistance winding) the damping effect is a product of the lag of the bi-metal strip how long it takes to heat up and deflect the needle.

That how you get a gauge that does not bounce up and down with every corner or bump.

 

Slide to resistor contact could be your problem.  The gauge is only ever full once per tank and constantly moving down from that point.

This meant there is a lot more "Bedding in" of the contact pattern in the toward empty direction.

The solution is to "Delicately" bend the contact spring to give more contact pressure between the slide and resistor.

 

The next problem and one overlooked by GMH originally is ranging.  Where the angular range of the float does not match the physical range of the drop tank.

e.g.  float is bottomed out on its travel and there is 20 litres still in the tank and or float is topped out (Submerged)and there is still 20 litres above the float.

Thats a tricky one to fix. Best to bend the float arm to contact the bottom of the tank  and ignore the full reading for the first 100 kms.

The angular range of the sender doesn't produce a continuous linear result because the winding spacings in the sender resistor are varied in their density to allow for the shape of the original fuel tank.

[Balfizar]

The angular range of the sender doesn't produce a continuous linear result because the winding spacings in the sender resistor are varied in their density to allow for the shape of the original fuel tank.

Yep I know that and that is why the shape of the bobbin compensates for the shape of the tank (sometimes).  What I was talking about   and I know because I have done it when reconditioning a 90 litre tank and seeing the angular range of the  sender not match the height or depth of the larger (90 litre)than standard tank (63 litre).  Post 80's the long range tank was just an extension down to create the larger volume, with the sender in the same (original) position the bottom 30 litres was still under a hanging float.  By bending the arm of the sender/wiper down to match the bottom of the tank you get bottom of the  tank/sender = empty. However top of the sender will still be about 63 litres and submerged when the tank is full.   If the angular range of the wiper arm  and compensating shape of the bobbin on which the rheostat - resistance wire is wound does not match the shape of the tank then you will not get an accurate reading.  With a new drop tank and an original sender/gauge you have no hope of linear "operation" of the fuel gauge.

If the Drop  tank profile is deeper than the original tank profile then the angular sender range is a mismatch.  Best you can do with the original sender is bend the wiper arm to match bottom of the tank is empty.

It does not matter where the original sender is mounted in the new/drop tank, with the angular range of a sender you will get fuel below the hanging float or above the submerged float or both in the mid mounted position.

Edited by Balfizar, 29 December 2018 - 11:58 AM.

[S pack]

Yep I know that and that is why the shape of the bobbin compensates for the shape of the tank (sometimes).  What I was talking about   and I know because I have done it when reconditioning a 90 litre tank and seeing the angular range of the  sender not match the height or depth of the larger (90 litre)than standard tank (63 litre).  Post 80's the long range tank was just an extension down to create the larger volume, with the sender in the same (original) position the bottom 30 litres was still under a hanging float.  By bending the arm of the sender/wiper down to match the bottom of the tank you get bottom of the  tank/sender = empty. However top of the sender will still be about 63 litres and submerged when the tank is full.   If the angular range of the wiper arm  and compensating shape of the bobbin on which the rheostat - resistance wire is wound does not match the shape of the tank then you will not get an accurate reading.  With a new drop tank and an original sender/gauge you have no hope of linear "operation" of the fuel gauge.

If the Drop  tank profile is deeper than the original tank profile then the angular sender range is a mismatch.  Best you can do with the original sender is bend the wiper arm to match bottom of the tank is empty.

It does not matter where the original sender is mounted in the new/drop tank, with the angular range of a sender you will get fuel below the hanging float or above the submerged float or both in the mid mounted position.

The original sender unit was only designed to suit the original 63 litre LH/LX fuel tank. The float arm needs to be lengthened to have any chance of making the sender close to accurate in a deeper tank, but that may also involve shortening the sender unit frame (pipe) so it doesn't project as far into the tank to keep the float and the pick up near the centre of the tank. If you can follow what I mean.

Edited by S pack, 29 December 2018 - 12:18 PM.

[VDO]

Some fiddling needed for sure to get the float and pickup where you want them.

 

Yes, correct re the shapes of the different tanks, that will have an effect on actual / perceived levels.

[VDO]

Did the 90 ltr tank come with a sender or was it set up to take the LH LX sender..?

[CI 0308]

The theory at play here with normal conditions is that more resistance (low 73 ohms) will incite slower heating of the windings due to lower voltage delivered and the less resistance (10 ohms full) will incite faster heating of the windings due to more voltage getting to them..

 

Thanks all for the replies. 

 

Yes Chris you use the original sender unit with Brown Davis tanks.

 

As I mentioned all my testing was done with the sender out of the tank and the wire extended with me sitting in the drivers seat looking at the gauge. (with the metal earthed via a fly lead) So the tank size etc is not applicable for tests. (I understand that the different tank proportions may be an issue.

 

I think you may be right Chris with your comment when I think about the theory, Full scale first = lower resistance and therefore gauge responds quicker due to more current.

 

I have reinstalled and am happy that empty = about 10mm above the fuel inlet. I will fill up and hopefully it indicates tank full. I will report back