2 car questions: about traction control and about fuel capacity

  1. Is there any reason to turn traction control off in good weather? For instance, does the car run a bit more efficiently without it, or is there some kind of wear and tear happening when it’s on? The owners manual is silent on the subject, other than showing where the on/off switch is.

  2. According to the manual my car (2013 Mazda 3) hold 15.9 gallons of gas. When the low gas light comes on, there’s always 2-3 gallons left in the tank, based on how much it takes to fill it. And today, the “range” estimate showed 9 miles of range left, but filling the tank only took 12-1/2 gallons. If the car actually runs until the tank empties, that means it had well over a hundred miles range left.

And here’s the question – how closely does the tank capacity match up with how much gas the engine will suck out before it dies of fuel starvation? Can I safely assume that my car has 80-100 miles or more left to run when the range indicator says 10 or 20?

The level sensor has a floating ball on an arm. When the ball is floating such that the top of the ball is hitting the top of the tank, there’s still room to pump in more fuel. And when the bottom of the ball is resting on the bottom, there’s still some fuel left in the tank.

If you want to get all empirical about it, you could 1) fill a 5 gallon can of fuel to put in your trunk, ii) note your mileage when the needle hits Empty, C) continue driving until you sputter out of fuel, and finally, 4th) do some subtraction. But you’re not going to drive the same way when you’re aware the needle is on E than you will when the needle is above F.

When my low fuel warning light comes on, it takes me around 11.25 - 11.5 gallons to fill up. But I have stretched it and it’s taken me 12.5 gallons to fill up.



  1. No. Just leave it on. Traction control works by reducing engine power if it sees one wheel spinning. If that doesn’t work, the brake on the spinning wheel could be applied by the anti-skid system on some vehicles. So if the wheels are spinning, it doesn’t do anything at all.

  2. The automakers set the low light to come on when there’s about 50-60 miles left for you to find a gas station. BTW: It’s best to not let the take get so low that the gas light comes on. The fuel pump is cooled by gasoline in the tank and having the level get so low that gas isn’t covering the tank can wear out the pump sooner. Of course this is more of a Arizona in the summer issue than Maine in the winter.

2a. The range indicator is a guess-timent and not a cold, hard fact. The only way to know is to run the tank dry and that’s not a good idea. Fill the tank when it’s down to about a quarter of a tank and you don’t have to worry very much.

The part about, “The fuel pump is cooled by gasoline in the tank …”, is mostly off. The pump is cooled by the gasoline flowing through the pump is correct. Otherwise you would see massive recalls reliability rants about failing fuel pumps and explosions from the overheating pump igniting the vapor in the tank.

Some fuel pumps are located at the top of the tank; some are external to the tank. Cooling is by the fuel flowing through the pump.

My 2015 Mazda 3 consistently behaves this way…when filled up it will say I have a ~405 mile range, which is quite accurate given I have a 13.2 gallon tank and average about 31.5 mpg.

By the time the stated range to gets down to ~10 miles, though, I’ve only driven about 320 miles per the odometer/tripmeter and the car only takes a little over 10 gallons, leaving an ~80 mile “reserve”.

When your range indicator says 20 miles you actually have the ability to drive 20+x miles before the engine coughs and sputters to a halt. The exact value of x depends on lots of factors, but I’m quite confident that x>0. I expect it would vary quite a bit from one year/make/model to another, and possibly from one car to another even if they are the same year/make/model.

FWIW, on my car, x is roughly 15% of the car’s total range. I’ve tested it a few times.

All those ‘estimated miles left until empty’ etc should absolutely be taken with a pinch of salt. Way too many variables to come close to being accurate, especially with consumer grade (read: bottom dollar) hardware.

I was involved in a project where we tried to more accurately predict and measure exact diesel fuel consumption over a fleet of off high way trucks. I am aware that diesel systems are complicated by having a significant fuel return volume back to tank, but the matter is easily enough resolved by taking in and out measurements from the Injector lift pump and subtracting.

even with extremely sensitive and expensive flow metering we just couldn’t generate any useful data that translated into the real world - data that actually had an effective outcome we could use to somehow reduce fuel consumption. and our company had a 1 billion USD per annum fuel bill (that **is **a B, not an M!) so saving fuel was big bikkies for us.

what this means to the OP’s question -

you are driving your car - you put the foot down to make that orange light, or you are speeding and slowing over a 10-15mph range because the car in front is all over the place, or there is a headwind, or the road condition is poor, or your tyres are at 20% below optimal inflection etc etc etc. All these to mean you can’t really get any better than (my WAG) +/- 20% range estimate from fuel count. even in the simpler to measure petrol engine.

Think about how the car’s computer measures the miles left. It uses an average fuel consumption and the fuel left in the tank. Fuel in the tank will not be accurate but should be fairly consistent. Average consumption is different altogether.

Our car spends most of it’s time on short local runs. We went away for Christmas and when I got onto the motorway, I could see that the miles left were actually increasing as I drove. This, of course, is because the average mpg was going up faster than the fuel was being consumed.

When it bongs to tell me that there are 50 miles left, I usually start thinking about filling it up.

Leave traction control on unless you’re driving in snow. Then you sometimes need to head the wheels spin to get down to something the tires can grip.

A spinning wheel is polishing the snow/ice slicker. To avoid being stuck take all care to avoid spinning.

Not sure about this advice. A spinning wheel (compared to an torque-interruption to that wheel) will also have a higher propensity to move sideways as the friction/traction is naturally less. This could put your car into a spin.

basically id always leave traction control on.

I don’t need advice about driving in bad weather. I want to know whether there is any cost to leaving traction control on in good weather.

No, there’s no “cost” to leaving it on - if it ever kicks in you’ll see a warning light on the dashboard come on - I don’t see why you’d ever want to turn it off in good weather (in a FWD car, anyway…)

If there was a cost, like decreased mileage or some kind of wear, that would be a good reason to leave it off when it’s not needed. Hence the question.

Leaving traction control enabled does not hinder the car’s efficiency nor does it cause more wear on the vehicle.

However, there are valid reasons to turn it off- particularly if you’re a driving enthusiast and/or have a powerful car. Many times traction control systems can be quite “invasive” in the default mode, in that they electronically close the throttle upon sensing any wheel slip whatsoever. If you have an engine that makes a lot of torque, this problem is exacerbated as it often intervenes even in normal circumstances and can become quite tiresome.

I think you bought into that WAG about the jet that crashed because of a static spark in the fuel tank!!
Have you changed out a fuel gauge sender?? the problem is usually very apparent of why it needs changing when you see the rheostat windings burnt open right there in the fuel tank, and No Fire, why? because the oxygen fuel ratio is WAY off!

The Electric submerged fuel pump will run hot if the fuel is low and besides there is return fuel that is heated up.
Your fuel pump will live longer if you refrain from running low, I bitch at my daughter about that because I don’t like working on them at all!!


2006 Pathfinder 4x4 with low range. I turn my traction control off when ascending my driveway in deep snow. I need the momentum of the vehicle to power through it. When it limits engine power because of a small wheel slip the car nearly stops and it has stuck me more than once because of it.

Low range 4x4 automatically turns traction control OFF. I either do that or depending on snow depth leave it in high range 4x4 and turn off traction control.

Turning onto a clear highway from an icy snow packed road can also be a problem with this. The last thing I want is the power to be reduced (for about 2 seconds) because of a little wheel spin.

I drive snow and ice 6 months out of the year. I expect a little wheel spin and can deal with it much better in acceleration than the TC reducing engine power.

Agree with you on the low likelihood of a spark/fire being generated in a closed fuel tank, but disagree on the cooling of the pump. The fuel pump (internal or external) is cooled by the fuel flowing through it, not by being immersed in fuel.

And yes, I changed many, both in-tank and external. The older BMW I’m restoring actually has both! An intank “helper” pump, and an external pump that does most of the real work.

Nothing I said included a remark about a spark. And the WAG you mentioned was the official report conclusion that CTers have a problem with.

Earlier comments were made about immersion in fuel cooling the pump. Hogwash, urban legend with cars in the past 50 years. Pumps are located at the top/bottom/external to fuel tanks. Even toward the bottom, it’s actually the pickup for the pump that’s lowest. I will repeat; it’s the fuel passing through the pump that provides the cooling and lubrication. Next you’ll be telling me that 3000 miles is the only recommended oil change you follow.

Over heating to the point of auto ignition is just a slower method of doing what a spark will do. Static spark or sparks caused by a contact sliding across rheostat windings? All I was really getting at is fuel level senders are certainly not intrinsically safe!!
I guess I will have to carry the
Pistol Style Digital Infrared Thermometer with for a few days and read the fuel reservoir temp, if only to prove to myself that I am wrong.