Welcome to our resource for MG Car Information.
|
MG MGA - Fuel guage - always reading 1/2 full
My next challenge is to sort out the fuel gauge which, when the ignition is switched on, permanently reads 1/2 full. Disconnected the wire at the tank to the sender unit and it still reads 1/2 full. It would be nice to sort this out - what would be the way please to diagnose the source of the problem? Thanks. Tim |
TJ Prime |
Check out Barneys MGA GURU site for the methods of checking both your Gauge and the Tank Sender unit Tim. He has all this stuff covered in there. Cheers Colyn |
Colyn Firth |
Better half full than half empty. |
A Bowie |
Thanks folks. Will see what Barney says. Cheers. |
TJ Prime |
I started this post a few years ago shortly after I had acquired my Coupe. The gauge was always reading half full. After a few years of using a piece of dowelling to check the fuel levels I've finally got round to trying to get the fuel gauge working. Having eliminated the sender, the wiring etc I focused in on the gauge. Removing it from the car and opening it up it definitely looked in a bad way from an examination of the innards (see attached photo) and one of the coils was a bit of a mess. I sent it to a gauge specialist for repair/refurbishment and it is due back any day, hopefully Monday. Does anyone have any idea what might have caused the problem shown in the photo? I don't want to have a repeat of this with the repaired/refurbished gauge. Thanks. |
Tim Prime |
Hi Old age? Barry |
B Bridgens |
The cause is that it was incorrectly connected, and the winding has acted as a fuse for the windscreen wipers. Make sure that the wires go appropriately to the B and T terminals - they should be stamped into the rear of the casing. One is green IIRC and the other two are green with a black stripe. The green must be alone on the terminal, the two others on the other terminal. Don't mix them up. |
Dominic Clancy |
Thanks Dominic. When I removed it there were 2 green wires to the B post and a single green/black wire to the T post. 4 ground wires to the central post. Tim |
Tim Prime |
That's backward. Green with black stripe is the Tank terminal. Black stripe indicating switched to ground (through the resistive sender unit). The B terminal is Battery, holding the Green wires from the switched fuse. |
Barney Gaylord |
Thanks both. Barney - I think your message crossed with mine. The gauge seems to be have been correctly wired up when I removed it but maybe it hadn't been at some point in its life. The gauge arrived this afternoon so will try and install it this next week sometime. It was repaired/refurbished by Scott Mitchell of Classic Spares & engineering in Strathaven, Scotland and it certainly looks as new from the outside. The test will will be whether it works however! Tim |
Tim Prime |
Hi Tim I never realised connecting these wires the wrong way round could cause damage! Considering this I would double check that the right collared wires do actually go to the right place, i.e. sender unit and fuse/heater fan. After all if there was one wiring fault there may be others. Hopefully can meet again when weather warms up when we can visit other Scottish health board areas, at least I should have had vaccine by then although perhaps you might still be too young! Paul |
Paul Dean |
Barney is correct, I was writing without digging out a wiring diagram and the memory is not what it used to be.... But the gist is correct, group the wires by colour and don't mix them, Make sure the correct wires are on B and T - wiring diagram is easy to find |
Dominic Clancy |
Thanks all. I've checked the wiring diagram and as long as I put it back together the same way it was when I removed it then it should be wired okay. Tim |
Tim Prime |
Did you get your fuel gauge to work Tim?
It's been a few weeks now and we hadn't heard from you, so either your gauge now works perfectly, or it has just been too cold to go out there into your garage to try it out. If you have done all the MGA Guru checks on the gauge it still reads incorrectly, (See http://mgaguru.com/mgtech/electric/fg_10.htm ), then you need to see if your sender unit has the correct resistance range. The variable resistance should read between 0 ohms at one end of the range and 70 ohms at the other. If this measures even a few ohms out at either end of the scale, then the gauge will not settle exactly on the full or empty marks. My brand new sender unit had a range of 0.5 ohms to 72 ohms, which is pretty close to the recommended readings, but this was still enough to prevent the needle from ever dropping below 1/4 full. I eventually fitted the Gauge-Wizard gadget which let me calibrate the gauge absolutely accurately. Colyn |
Colyn Firth |
Colyn What is the Gauge Wizard and how is it used? Pete |
P Draycott |
Hi Pete,
the Gauge-Wizard is a clever electronic unit supplied by Spyder. It can be programmed to correct your fuel gauge so that the needle positions are absolutely correct and it will indicate exactly on empty when the tank is empty and exactly on full when the tank is full, with accurate readings in between the two. My original repro sender unit at 15 to 98 ohms, was miles away from the correct 0 to 70 ohms and the gauge never ever dropped below the half-full position. So I had to use the milometer to give me an idea of how much fuel I had left in the tank. I eventually became fed up with this and decided to buy a new repro sender unit, which I measured as 0.5 to 72 ohms. This gave a more accurate reading but, even with this never allowed the needle to drop below 1/4 full. The mistake I made was to begin to trust the gauge a bit more and this resulted in me running out of fuel on two occasions. So I decided to try out the Gauge Wizard after reading Edward Van Dykes article about it in Safety Fast. It has turned out brilliantly for me, it even features an optional low fuel light that illuminates when the tank level drops to 1/4 and the flashes continually when you get to 1/8th full. I have checked this and there really is only 1.25 gallons left in the tank when the light begins to flash. For most MGAs, that is 30 to 40 miles of range, but mine is a bit thirsty and I probably only have 30 miles range left at that point and so the accuracy of the gauge has been a life saver for me. Another good point is that, so long as your present fuel gauge shows a reasonable movement between the tank being full and empty, then the Gauge Wizard can probably be used to show correct needle positions. Here is the link to a discussion we had about the Gauge Wizard a while ago. https://mg-cars.org.uk/cgi-bin/or17?runprog=mgbbs&access=&mode=archiveth&subject=6&subjectar=6&thread=2018071322574923959 Cheers Colyn |
Colyn Firth |
Thanks Colyn How did you take your Ohms readings. Can this be done on the car? Pete |
P Draycott |
In theory Pete, you could measure the resistance readings of your sender unit with it still fitted into the tank.
You would have to measure the resistance between the sender terminal and earth with the tank full. Then you would have to repeat this with an empty tank. It's much easier to do this with the sender unit removed, not so easy to seal the unit though when you refit it. I used hylomar for this and it seems to have worked so far. If you do send for a new sender unit, check the resistance readings before you fit it because, even a small variation from the correct 0 to 70 ohms range can cause an inaccurate gauge reading. If the range is not very close to 0 to 70 ohms, I would send it back for a more accurate one. If your gauge already reads fairly accurately, you could use Barneys method of correcting it, see link below. http://mgaguru.com/mgtech/electric/fg_10.htm It is apparently a fairly simple procedure, but I have not yet been brave enough to try it. :^) Cheers Colyn |
Colyn Firth |
I forgot to say that I just used a simple digital test meter, set on the resistance setting. I have two or three of them and they all read pretty much the same when I use them. Colyn |
Colyn Firth |
I used one of Barney’s methods. It certainly works although it very fiddly and very sensitive to minute changes. Many mine isn’t as good as yours Colyn but it is acceptable to me although I do need car stationary and level to be consistent.
By the way I still carry my calibrated length of half inch dowel which is always perfect and isn’t in the way stowed at the side of the boot release rod. By the way that new post from me yesterday that was actually from a year ago has now been removed by Mike our web master. Paul |
Paul Dean |
Colyn. Sorry for the lack of an update on the gauge. It is because sadly the gauge refurbishment hasn't sorted out the issue as I had hoped. The gauge was definitely defective but now instead of reading permanently 1/2 full it now reads permanently full. I've checked all the wiring out and I'm pretty confident there are no faults there. Sender wiring is fine from the gauge to the sender. The sender is relatively new (less than 2 years old) but I guess it could be the problem. It was new from Moss and it replaced a non-mga sender unit that was on the car when I bought it that I believe is also used on mini-vans I've got a bit fed up with trying to get the fuel gauge working so been doing other things on the car that are a bit more productive/satisfying! Maybe I'll be sticking with the piece of dowel rod for a good while yet - we'll see. Thanks. Tim |
Tim Prime |
Tim - you have checked your sender unit is the right way up? My "A" had two previous owners and their garages who couldn't sort out a gauge that always showed full. I spotted the problem pretty quickly. The connector must be at the bottom. The equispaced holes mean it can be fitted in six positions! PeteT |
PeteT |
Hi Tim.
if your fuel gauge always reads full when you switch the ignition on, (assuming it is installed the correct way up as Pete T says), then you could have a couple of different problems. A poor or broken earth connection of the sender unit. A broken connection to the sender unit wire or the resistance of the sender unit is far too high. I originally had the same problem with my fuel gauge always showing full, this was caused by my car having rubber fuel pipe. The original metal fuel pipe was also the earth connection. Once I fitted an earth wire to my sender unit, the gauge worked more accurately. Once you know that the sender unit has a good earth, you could then check the continuity of the sender wire. If that is ok, then I would next remove your sender unit and check the resistance range. Cheers Colyn |
Colyn Firth |
Does the dowelling method utilise notches relating to fill? If so is there a measurement chart? |
P Draycott |
Attached is a photo of my dowel fuel gauge. A couple of points. The end that goes in the tank is cut at 45 degrees so it always sits in same orientation. I calibrated mine while filling a drained tank from cans. It is marked in indelible ink into 5 equal points each of approx 2 gallons. Notches may be better but ink is surviving ok. You will not the note that the lowest fifth has another line splitting it approximately in half and this is the amount fuel that has to in the tank before the pump picks up. So beware you can’t run down to empty. After use smell goes very quickly and I keep by side of boot opening rod.
I don’t use dowel very often as my gauge is working reasonably now but a lot less linear than the rod by the way. I did use Barney simplified method to set gauge but it is pretty tricky as it so sensitive to the slightest movement of the magnets. Of course calibration is a lot easier with sender unit out of tank. Paul |
Paul Dean |
Thanks Paul. What are the division measurements? |
P Draycott |
I explain measurements in message above, that is approx 2 gallons but note the extra line in lowest fifth. Paul |
Paul Dean |
For those wanting to make a calibrated stick - I use a piece of 9mm x 9mm square dowel 600 mm long with marks for 5, 10 & 15 litres at 50, 75 and 100mm from the bottom of the stick - use a hacksaw cut to mark the intervals. Cheers Mike |
Mike Ellsmore |
Thanks Mike A little confused by the 5, 10, 15 litres that equate to 1.1, 2.2 and 3.3 imperial gallons. I assume you have the standard 10 gall (Imperial) tank. |
P Draycott |
Sorry P. Us convicts went metric in 1971, not much chance of us having an Ozexit to get back to imperial measurements. Mike |
Mike Ellsmore |
Colyn In a previous post you mentioned that you bought a sender unit from Moss and this was close to the resistance settings. They show 2 types for sale. I have asked them for the ratings of each but they cannot tell me. Which one did you purchase? Pete |
P Draycott |
Pete,
I have had to look at my receipts because I didn't recall the sender unit coming from Moss, after double checking, I found that it came from NTG services. Even though the resistance range at 0.5 to 72 ohms was very close to the correct 0 to 70 ohms, the couple of ohms difference meant that the needle on the fuel gauge wouldn't drop below about 1/8th full, even when the tank was empty. (My original sender unit (from memory) had a range of something like 33 to 98 ohms) The problem was for me that I began to rely on the gauge reading to tell me that I needed to refill the tank. This caused me to run out of fuel on a couple of occasions and so I decided to fit the Gauge Wizard to make the gauge more accurate. So, if you want an accurate fuel gauge, you need to make certain that your fuel gauge is a correct original MGA gauge and also that your sender unit has pretty much exactly the correct resistance range. I were you, if you do buy a new sender unit, tell the supplier that you will only buy it with the proviso that you can send it back for exchange or a refund if the resistance range is not as it should be. Cheers Colyn |
Colyn Firth |
Over the years I have tried to replace my sender unit. The first was way out on resistance range, the next I never fitted as it had no coil at all, the next one had the six fixing holes at a bigger PCD than the tank holes! The next a 'modern' design with a plated steel sheet case worked fine for a year and failed as the single wiper had cut through the resistor coils.
So I pulled out my original sender, pulled it to bits, and got the broken internal wire connected to the terminal again. It still works perfectly. Dip-stick - with these troubles it is a must. Mine is a 1/4" dia wooden dowel simply marked at the full line - I can guess how much and I know I have something wet in there - result. PeteT |
PeteT |
My 1960 1600 fuel gauge moves erratically when moving. Currently there is about 8.5 gallons (imperial) in the tank (dipstick measurement coupled with full tank start and previously measured mpg). When I turn on the ignition the gauge shows empty and when mobile moves erratically. Here are my test results.
1. Disconnect wire from tank unit (to avoid confusion). Touch ohm meter wires together to verify it works. Should read 0-ohms (or very close like 0.2-ohms). Reads 0 Ohms 2. Connect ohm meter leads to sender terminal and sender body. Should read somewhere between 0-70 ohms (depending on how much fuel in the tank, 0=empty, 70=full). Reads 9.2 Ohms 3. Connect ohm meter leads to sender body and ground on fuel tank. Should read 0-ohms (or very close like 0.2-ohms). Reads 0 Ohms 4. Connect ohm meter leads to fuel tank and ground on chassis frame. Should read 0-ohms (or very close like 0.2-ohms). If this reads high resistance, you may have lost tank ground by cutting fuel pipe and connecting with rubber hose to fuel pump. In this case, add a grounding wire from sender unit screw to ground on frame screw nearby. Reads 0 Ohms 5. Connect ohm meter leads to sender terminal and ground on chassis frame. Should read somewhere between 0-70 ohms (depending on how much fuel in the tank, 0=empty, 70=full). Reads 9.3 Ohms 6. With ignition on, ground the sender wire, and the gauge should go exactly to "E" (not on the E" peg). Disconnect the sender wire, and the gauge should peg hard against the "F" peg. Connected - goes hard to E peg Disconnected - goes hard to F peg 7. Connect the sender wire to the sender terminal. With ignition on, the gauge should read somewhere between E" and "F" (depending on how much fuel in the tank). Reads Empty I'm wondering if these results indicate a possible float puncture problem. Any guidance would be much appreciated. |
P Draycott |
Pete (Draycott)
Your test results indicate that (AFAIK) your gauge is working but you have a problem with your sender unit. It may well be that the float isn't working, but you may as well remove the sender unit to check it and the resistance range, otherwise, it is just guesswork. (Remove the sender unit connection before you remove it.) As soon as you remove it, check to see if there is any fuel inside the float, before it has a chance to dry out. It isn't such a big deal to remove it, I drained most of the fuel from my tank and just put some duck tape over the hole. With the sender unit on the bench it is easy to check the resistance range and then you can decide to either live with it or replace it. Also, when you refit the sender, fit a dedicated earth wire to it, I connected a wire between one of the mounting screws and one of the wiring loom mounting points. Then you can be certain that your sender unit is definitely earthed. Best of luck Cheers Colyn |
Colyn Firth |
Pete (Draycott) That all makes sense, but your gauge reads empty when nearly full. Are you sure that sender unit is the right way up? i.e. terminal at the bottom. PeteT |
PeteT |
Yes Pete it's right way up. |
P Draycott |
Peter D
With your 8.5 gallons and 9.2 ohms this surely is wrong which is why I say the sender could be upside down. When full the resistance should be up in the 70 ohm region. If your terminal is at the bottom and assuming your sender is the original type I'm wondering if the bottom end of the coil has been connected to the terminal instead of the top. In theory this is possible. What happens when you are getting low on full, does the gauge give a higher reading? Then you need to remove the sender and check the resistance readings with it on the bench. Inside you should see a very thin bit of the coil wire coming from the top of the coil all the way down to the (bottom) terminal. (Be careful it breaks!) I don't think you could mount it upside down to compensate as the crank in the float arm will be upside down too. Pete T |
PeteT |
Decided to take the bull by the horns and get this sorted. So here is the potted version.
1. Built a full scale fuel tank from heavy card and calculated volumes for the curved top and bottom and the middle sections to be able to set fill levels accurately. 2. Purchased a sender unit from NTG Motor Services and measured resistance range 6-85 Ohms. 3.Purchased a Spiyda Wizard from Syida.com 4. Fitted sender unit to dummy tank. This sender is angled to the right and when fitted bottoms and tops in the curved ends. The resistance range was reduced to 8-80 Ohms. So bent the arm to set top and bottom just clear and get full range 6-85 Ohms. From the tank calculations set the depth measurements to 1/4, 1/2 and 3/4 full and operated sender arm to these settings and recorded resistances. E=6, 1/4=24, 1/2=45, 3/4=62, F=85. 5. Removed fuel gauge from car and recalibrated as per MGAguru method FG-102e. 6. Set up and calibrated Spiyda with resistance range as per the instructions. 7. Coupled up the system to mimic the car set up and programmed the gauge as per the Spiyda instructions. 8. Set the anti slosh level to 3. Functions exactly as it should with spot on accuracy and nice controlled needle movement. |
P Draycott |
Hi Pete D, I know we have already discussed this on the MG Experience Forum but I wondered if there was any advantage to building up a "dummy" fuel tank? Wouldn't it have been easier and just as accurate to follow the Spiyda instructions to drain the tank and then refill the tank in 1/4 steps to calibrate the sender unit to Gauge Wizard? Glad you have got it to work accurately now. Cheers Colyn |
Colyn Firth |
Hi Colyn Having read your last entry I had decided I would get a wizard but I am afraid they say negative earth only. Is there any reason on a positive earth car why they couldn’t just be wired in reverse if kept insulated? I hope you are staying above water level in the latest heavy rain down there? Cheers Paul |
Paul Dean |
Colyn
There were a number of reasons why I chose to use a dummy tank. I wanted to be able to bench test the complete setup in the comfort of inside without the limitations of working on/in the car. As you know it's been pretty dismal and cold hereabouts of late. It also enables easy testing of the sender movement which I can access through an inspection aperture. What it did reveal was that the new sender float was bottoming and topping before it's full range. This was due to the angle of the sender arm on the NTG sender placing the float in the curved sections of the bottom and top of the tank so falsifying readings when installed on the car. Something that could not be seen on the car. The dummy tank is also a test piece for any future work. Regards Pete |
P Draycott |
Pete D, I see what you had in mind now, it has got to be a lot safer too than having to pour in lots of fuel. One thought I had was that maybe you could set up as sender unit testing and set-up service? I would have loved to have been able to send you my sender for accurate testing, to find out if it was going to work before I installed it. Cheers Colyn |
Colyn Firth |
Is there any interest in Colyn's suggestion regarding a set up and testing facility? |
P Draycott |
Paul.
thanks for your concern, fortunately we missed the flooding this time around. I was prepared though and I had my petrol powered water pump set up just in case, my pump will remove up to 5500 gallons per hour. We were not so lucky back in 2007 when we had a foot of water through the place, fortunately my MGA was away at Bob Wests at that time. I have attached a couple of pics of that horrible experience in 2007. We have had this happen a few times since 2007, but fortunately, my flood precautions have prevented another disaster. I think your idea of using a Gauge Wizard on a positive earth car would probably work fine because the G/W unit is an all plastic box so should have no short circuit issues. But, rather than taking a chance, I would telephone Spiyda to ask them about it, just in case the electronics inside would cope with the polarity switch. Cheers Colyn |
Colyn Firth |
Another couple of pics showing the moat we had around our house back in 2007 Colyn |
Colyn Firth |
This thread was discussed between 02/11/2016 and 27/01/2021
MG MGA index
This thread is from the archive. The Live MG MGA BBS is active now.