• Hi Guest!
    You can help ensure that British Car Forum (BCF) continues to provide a great place to engage in the British car hobby! If you find BCF a beneficial community, please consider supporting our efforts with a subscription.
    There are some perks with a member upgrade!
    **Upgrade Now**
    (PS: Subscribers don't see this gawd-aweful banner
Tips
Tips

Spitfire Spitfire fuel gauge

Share this page

Jerseygirl

Jerseygirl

Jedi Trainee
Offline
Question for anyone and everyone... When we fill the gas tank in the Spit (visually checking that the tank is full without filling up the filler neck), my gas gauge still only reads slightly above half full. It seems to run gradually down to empty, but once there, there's still plenty of gas left in the tank.

My question is, does anyone know if this indicates a faulty fuel gauge that may need to be recalibrated (if such a thing is even possible), OR would the fault be with the sender in the fuel tank?

Thanks again in advance for putting up with yet another stupid newbie question. /ubbthreads/images/graemlins/grin.gif
 
Hi,
Well, the next time you have the tank close to empty. I would test the sender by pulling it out and putting a Ohm meter on it to see what it measures. On my spit when I first got it the gauge kinda did it own thing. Turned out that my float on the sending unit had a pin hole and filled with gas and had neutral bouyancy. I can measure mine out for you this evening to give you a range of impedence for the sending unit or someone here might know it off the top of their head. This place has some amazing folks..
Dennis
 
Hello Robyn,

a quick check for the fuel (and water temperature) gauge is to short the sender wire to earth with the ignition on and if the gauge goes full scale then that is correct and the fault lies with your sender.
Both of these gauges are fed with a voltage stabiliser at a slightly reduced voltage of about 10 volts. Do the above check to both instruments and if both fail to read full scale it is probably the voltage stabiliser. If one is full scale and the other half scale then the instrument is faulty.
Finally ensure that your tank sending unit is properly earthed as this will stop a full scale reading.

Alec
 
I agree with both Dennis and Alec about what and how to check. Another check that I have done on some cars (but not a Spit... don't know if it's possible) is to use a bent coat hanger through the filler neck to lift (and force down) the float arm of the sender unit while the ignition is on. If lifting the float arm all the way up is possible and shows the tank is full... it's likely the float has a pin hole in it.

Metal floats can be soldered closed once you locate the pin hole and drain all the trapped fuel out. If you have a plastic float, these can also be sealed once drained by carefully using a soldering iron to smooth over the area where the pin hole is.
 
I did a quick measurement of mine. If the tank is empty I had reading of 279 Ohms. If the tank were filled the reading was 19 Ohms. If you were to move the float arm like Doug mentions you'll see a reading of somewhere between 19 and 279 Ohms.
Good luck...
Dennis
 
[ QUOTE ]

Both of these gauges are fed with a voltage stabiliser at a slightly reduced voltage of about 10 volts.

[/ QUOTE ]

Not exactly. The gauges are fed by a pulsed 12 volt supply. If a constant voltage were fed to the gauges, it would need to be ~10 volts. However, in practice, a slowly pulse of 12 volts / 0 volts is sent to the the gauge circuit. The gauges are built to react slowly so you do not see the pulses.
The system works by heating a bi-metalic thermostat. When the thermostat heats up, it cuts off the power until it cools off again. Since the voltage put out by the generator changes with speed, this is a good compenstor. Voltage goes up, the pulse becomes shorter. Given the state of electronics in the late 50s when this was designed, this is a quite satisfactory method of dealing with the rather widely varying voltage in the car.
Yisrael
 
Hello yisrael,

yes that is right, but it is called a voltage stabiliser and when the voltage is measured, particularly with an analogue meter it reads about 10 volts. It's basic principle is that of a flasher relay.

Alec
 
Robyn;

Mine used to act the same as yours. I pulled the sending unit out of the tank and found the float half full of fuel. I replaced the sending unit because I have tried to repair plastic floats in the past without success.
 
Like 70herald said, the voltage stabilizer pulses the voltage to the gauges. I've never been able to measure this accurately with a digital meter and some analog meters work better at this than others. I've noticed that some stabilizers have a much faster switching rate than others. I don't know when the last LBC started using the stabilizer but it was phased in during the mid-1960s.

The measurements that Dennis took are consistent with other sending units I've measured.

I worked on voltage stabilizers a couple of years ago. I currently run a solid-state u7810 voltage regulator instead of the Smiths unit. You can see what I did and read about how it's possible to repair and recalibrate your existing stabilizer by downloading my PDF document from:
https://home.mindspring.com/~purlawson/files/SmithsVoltageStabilizer.pdf
 
Hey Doug,
Excellent document, you just saved me from recreating the wheel. I was going to look into changing mine over to a SS VR.
Thanks!
Dennis
 
Hi Doug
Nice documentation. Wish I had that about a year ago when I needed to replace my regulator. I reverse engineered the circuit at the time and put in a 7810 VR! I soldered the heat sink onto into the original case and closed it up with a pop rivet. Has worked well ever since.
Yisrael
 
I'm working on a new speedo for the Mini (VDO transplant behind the face). I decided on this one to just mount the 7810 chip to the back of the speedo body rather than try and enclose it in an old stabilizer housing. When the wires are properly tied to an anchor point this is a MUCH easier and cheaper fix than anything else. On most LBCs the stabilizer is hidden behind the dash anyway so it's never seen. The u7810 chips from Mouser are well under $1 each so it's hard to argue pricewise about using them.

There is a guy (sorry... lost his email address) on eBay who is/was selling zener diode based power supplies to take the place of the Smiths stabilizer. His argument is that his circuit is steady and very immune to voltage spikes. The u7810 chip I used is rated for normal inputs up to something like 37V and I'm sure it has some safety factor to handle voltage spikes. So far I haven't had any issues or failures with the solid-state solution. Incidentally, some of my E.E. coworkers (I'm an M.E.) encouraged me to put filter capacitors after the regulator to protect the gauges. I do not feel this is necessary since the Smiths unit is an oscillator that for years has been hammering our gauges with 12V on/off pulses. The solid state regulators are already cleaner than what was there before.

Thanks for the compliments. Hopefully the documents will be of use to more people.
 
Hi Doug
At least acording to my degree I am also an ME. All my working experience has been in the manufacturing of IC's and other silicon based things.
The battery in the car really acts like a capacitor in this case. While there may be some spikes, it is very unlikely that they would ever be big enought to kill the VR. No reason to put caps after the VR, the Smiths gauges on these cars are thermal spring designs (just like the original VR, someone in the eng department at Smiths liked bi-metalic strips!). There response rate is so slow that there is no way you could see any variation caused by the VR. Since the gauges can be shorted out (direct 12v across the gauge) without damage, obviously that is not a major worry either.
A zener diode solution will also work, however when we worked it out here, the amperage is more of a problem (with heating) besides it needed more components. The simplest, best way is the 7810.
 
Back
Top