Topics and replies.

Fuel Gauge Accuracy

Postby Peter Spurrs » 27 Feb 2017 20:05

Flushed by the success of fixing the fuel pump and seeing the fuel gauge sender unit sitting tantalisingly close to it, I've now started to think about the accuracy of the gauge. When the tank is full, the gauge reads "full". When it is about a third full, it reads "running on fumes". I can cope with it, but it would be nice for it to be a bit more realistic lower down in the range. The only reference I can find is in Rick Astley's book (P148) which states the resistance range at the extremes of the float's range. I'll take the sender unit out of the tank when it is close to empty. If it's in range, all I can think of is to bend the float arm down a bit.
All thoughts would be welcome.
Peter
Peter Spurrs
 
Posts: 63
Joined: 15 Jul 2014 08:55
Location: Derbyshire, UK

Re: Fuel Gauge Accuracy

Postby Victor Smith » 27 Feb 2017 21:34

Peter,
Mike Macartney's Rebuid report 110 included some checks he made on the sender and gauge.
http://www.v8register.net/FilesV8/17011 ... rt-110.pdf
Victor Smith
 
Posts: 478
Joined: 06 Dec 2012 15:46
Location: London SW

Re: Fuel Gauge Accuracy

Postby Peter Spurrs » 28 Feb 2017 06:47

Thank you Victor. I didn't spot that one.
Peter Spurrs
 
Posts: 63
Joined: 15 Jul 2014 08:55
Location: Derbyshire, UK

Re: Fuel Gauge Accuracy

Postby Peter Spurrs » 13 May 2017 18:09

Nearly three months later and the fuel tank is finally empty.
The sender unit looks pretty new and has the correct part number (AHU1027) on it. The resistance at the extremes of motion are within tolerance too. I then connected it up to the new solid state voltage regulator and the gauge. At full, ¾, half, ¼ and empty on the sender, the gauge was quite accurate. All looking good on the bench.
I had run the tank to dry, added 4 litres and driven 1.5 miles home, so was confident of the fuel level. When I put the sender unit back into the tank, the gauge read zero. I then added 5 litres from a can and the gauge still read zero. A total of 9 litres is two gallons in a 12 gallon tank; that’s a sixth so I would expect some gauge reading. I then took out the sender unit and bent the float arm down until the reading was about right. I then added successive cans each of five litres. 14l was just over ¼ on the gauge and 24l just short of half. 34l gave ¾ and from then on, the gauge read full.
I now have a system is inaccurate when there’s lots of fuel, but gives a reasonable indication towards empty. That’s good enough for me.
It seems to me that whilst the equipment is accurate enough on the bench, the sender unit’s range of motion is insufficient to reach the top and bottom of the tank. That’s to say that the float arm is too short. A proper fix would be a longer float arm, but that’s a more complicated job. I’ll live with my compromise.
Peter Spurrs
 
Posts: 63
Joined: 15 Jul 2014 08:55
Location: Derbyshire, UK

Re: Fuel Gauge Accuracy

Postby Mike Howlett » 13 May 2017 18:51

My gauge is pretty inaccurate too. When I built the car I measured the resistance of the sender out of the tank and asked Speedy Cables to make my new gauge to those figures. Once in the car it was hopelessly wrong, but my clever son suggested I put a 220 ohm resistor across the gauge and it was a lot better. Recently it has been very poor again, only reading a half when full, and dropping to Empty within a 100 miles or so.

So I had a look at the terminals on the tank sender unit and on pulling them off found that they were badly corroded. I cleaned up the spades with the Dremel and fitted new connectors on the wires. Using plenty of electrical grease and sealing them in as well as I can, the situation is much better. The car is just about empty at the moment and the gauge reads just a fraction above Empty. I'm hoping that once I fill up normal service will be resumed.

It's a daft idea to have the sender terminals exposed to all the water and rubbish from the road, and to have it in the side of the tank so it can't be removed without the tank being fairly empty. I suppose nothing is ever perfect!
Mike Howlett
 
Posts: 112
Joined: 15 Dec 2012 09:27

Re: Fuel Gauge Accuracy

Postby Martin Cridford » 13 May 2017 19:13

Be careful when you run your fuel level down to empty...see below what I found in my 20 year old RV8 tank! I had major fuel starvation problems because of this. A new tank and two changes of filters finally fixed it.
Note also the gauge float arm sits across the 'swirl pot' built into the tank. There are other electrical connections on the gauge that can corrode too. The black headed 'swan neck' is actually the remains of the flexible plastic return connection which feeds back into the 'swirl pot'.
Attachments
2016-08-29 17.03.08.jpg
Martin Cridford
 
Posts: 77
Joined: 20 Feb 2016 08:19
Location: Exeter, UK 1995 RV8 No:1422, 2004 MGZR 160

Re: Fuel Gauge Accuracy

Postby Mike Howlett » 13 May 2017 21:16

I don't understand why it is any more likely to pick up crud when the tank is empty. The position of the pick-up pipe doesn't change, so if there is crud at the bottom of the tank it shouldn't matter whether the tank is empty or full. The rubbish will always be there.

It's interesting to see the swirl pot inside the RV8 tank. I wondered how it was arranged. I use a standard MGB tank and have an external swirl pot mounted in the driver's side battery tray. That means I need two fuel pumps, a low pressure one to keep the pot topped up, and a high pressure one to feed the engine fuel rail. A pot within the tank means you only need the high pressure pump as the pot is automatically filled.
Mike Howlett
 
Posts: 112
Joined: 15 Dec 2012 09:27

Re: Fuel Gauge Accuracy

Postby Peter Spurrs » 14 May 2017 07:40

Martin, your tank is not a pretty sight. Luckily, my tank was fitted in the last five years and is clean inside.

I'm showing a lack of knowledge here, but why does an RV8 need a swirl pot? My limited understanding is that's it's there to ensure a steady supply of fuel under big cornering loads when racing. Are RV8 drivers an altogether more racy breed?
Peter Spurrs
 
Posts: 63
Joined: 15 Jul 2014 08:55
Location: Derbyshire, UK

Re: Fuel Gauge Accuracy

Postby Mike Howlett » 14 May 2017 08:36

Peter, all fuel injected cars need a swirl pot. With carburettors any interruption in the flow from the tank isn't noticed because the engine is running off the float chambers. But on a fuel injected car, any air bubble in the fuel feed will cause a momentary stumble and misfire as the air bubble is injected into the engine instead of fuel. When the tank is getting towards empty, the pick up pipe can easily be left high and dry by normal acceleration, braking and cornering. So the tank contains a swirl pot and the supply to the engine is taken from there. The pot should be full right up to the last drop of petrol.

I know this happens because on my car with the remote swirl pot I can hear the low pressure pump in the background and when I am low on fuel I can hear its note changing as it occasionally pulls in air. But since the high pressure pump is taking fuel from the pot, the engine keeps running smoothly. Here is a photo of my arrangement.
Attachments
Small Pumps.jpg
Mike Howlett
 
Posts: 112
Joined: 15 Dec 2012 09:27

Re: Fuel Gauge Accuracy

Postby Peter Spurrs » 14 May 2017 14:08

Mike.
A wonderfully clear and concise response.
Thank you,
Peter
Peter Spurrs
 
Posts: 63
Joined: 15 Jul 2014 08:55
Location: Derbyshire, UK

Next

Return to V8BB Messages

Who is online

Users browsing this forum: No registered users and 1 guest