Car dies while taking corners

[TheNelson]

I recently had some issues with my '93 V6 MX-3 where it would stall/hesitate and die on me and the car was undriveable. We've found that the fuel tank was extremely rusty and the fuel pump wasn't working very well anymore.

Drained the gas out the bottom, then went in the top and scooped out as much of the sludge we could and cleaned it up a lot. Put in a new fuel pump and pre-filter.

The reason for all the rust is the car sat outside with no gas cap for over 2 years.
That fixed the problem. Filled it up with gas and I've been driving it all week flawlessly. Now that my gas gauge only reads half a tank when I take a corner swiftly (Or e-brake slide on the snow) the car dies while going through the corner.

Now the tank reads about 1/4 full and the car will die going around ANY corner at ANY speed. If I'm driving straight it will run strong and pull hard even if I'm at WOT for a bit.

What inside the tank could be malfunctioning to cause this? Is it likely that a new gas tank will solve the issue?

[mitmaks]

I'd see if prefilter is getting clogged with rust/debri

[MrMazda92]

It doesn't sound like clogging to me... I haven't seen the sending unit for our cars yet, as I haven't had reason to look at mine yet, but it sounds to me like your tank may be improperly mounted, and the applied force of a turn is either pinching your fuel lines, or the sending unit is damaged. Do you park somewhere that would allow you to monitor for small fuel leaks? If so, check for small amounts of gas on the ground/floor. Even if none are visible, this could still be your problem.

Also, if you decide to clean the tank out again, many people suggest using a length of chain to help free up rust/debris. I haven't tried this method, but it definitely would help to free debris up. There are a ton of guides on how to clean a tank online, just google it; Be careful though, I've seen many guides recommending acetone as a cleaner for the tank. If you use this route, VERY thoroughly flush the tank when you are done(You should after any cleaning anyways).

When you resolve the issue let us know what the cause was!

[Inodoro Pereyra]

I'd advise you to replace the tank.
If he tank was so badly rusted, chances are the internal baffling is gone. If that's the case, once your gas level starts going down, when you take a curve all the gas would be pushed to one side, starving the fuel pump.
here's an easy way to test it: fill the tank up again. If the problem disappears, replace your tank.

[MrMazda92]

x2 to Inodoro's advice. That hadn't occurred to me, and is a very valid point.

[Daninski]

Also, if you decide to clean the tank out again, many people suggest using a length of chain to help free up rust/debris.

A word of caution, anything that could cause a spark in your tank could result in an explosion. Even with the tank drained fuel off-gasing from the metal itself will occur and you could end up with an explosive atmosphere.

[Inodoro Pereyra]

A word of caution, anything that could cause a spark in your tank could result in an explosion.

Hmmm..., no. It can cause a fire, but as long as the tank has an opening, an explosion is not possible.

[TheNelson]

I'd advise you to replace the tank.
If he tank was so badly rusted, chances are the internal baffling is gone. If that's the case, once your gas level starts going down, when you take a curve all the gas would be pushed to one side, starving the fuel pump.
here's an easy way to test it: fill the tank up again. If the problem disappears, replace your tank.

This is exactly what I was thinking of doing, and in the mean time I wanted some other opinions!

I hadn't thought about the baffling, but it is sure possible. When I replaced the pump the gasoline looked red because of all the rust.

I'm not going to clean the tank again at all, if anything I'll spend the $100 and replace it.

[Ryan]

A word of caution, anything that could cause a spark in your tank could result in an explosion.

Hmmm..., no. It can cause a fire, but as long as the tank has an opening, an explosion is not possible.

You're partially right

An explosion requires an excess of oxygen, an almost empty tank WILL explode. A full one, not so much. Ask anyone who's tried welding a "I thought I washed it out really good" fuel tank.

[TheNelson]

On my way home tonight I got fuel and on my 12 mile drive home didn't have any problems. I tried taking the sharp corners fast and no issues.

However one odd thing happened at the pump. I had to squeeze the gas filler trigger about a dozen times before I filled up the tank. The gas would fill up the neck like something was obstructing it from flowing freely into the pump, but eventually it would seep down into the tank.

I'm ordering a new tank on Monday I think.

Thanks for the insights!

[Inodoro Pereyra]

Hmmm..., no. It can cause a fire, but as long as the tank has an opening, an explosion is not possible.

You're partially right

An explosion requires an excess of oxygen, an almost empty tank WILL explode. A full one, not so much. Ask anyone who's tried welding a "I thought I washed it out really good" fuel tank.

I have actually done so a few times, and I'm now in the process of welding 2 propane tanks together (that I cut open with a grinder, and both of which caught fire on me) to build a glycerin still.

Unless you have a purpose made explosive (like nitroglycerin, or any plastique) that will explode without the need of a container, the only way you can make a flammable substance explode is if it's enclosed (together with an oxidizer) in a container that is either airtight, or which vent(s) are unable to let the combustion gases escape fast enough to keep the pressure under the rupture pressure of the vessel. If you are going to wash a fuel tank, it's a safe bet you will do it without the fuel pump, so you will have a big hole in the tank, more than big enough to keep the pressure buildup within safe range. So, no. Tanks do catch on fire, and it's definitely no fun when they do, but they don't expode, unless you're dumb enough to seal all the vents.

Finally, if there's any doubt about it, a few years ago, in the show "Mythbusters" they tackled this specific myth. Even with the tank in the car, a few bullet holes in the tank were enough for it to catch on fire, but not to explode.

[Inodoro Pereyra]

Sorry, Nelson: the fact that the car worked properly when you added gas means that both your fuel pump and filters are good. At this point, I'm sure replacing the tank will solve your problem. Or, if it doesn't, just don't turn...

[tripleMX-3chicMS]

If you remember back in June or so... I had the same trouble with the GS... was rust in the gas tank,
and I did have to change the gas tank, fuel filter, and one temp sensor. It has been fine since....
well, except for the break line issue... rust eats everything..

[MrMazda92]

On the note of the sparks in the tank, I am with Inodoro. I personally wouldn't use the chain either way, simply because I don't believe it's worth the time. There are simpler, more effective alternatives. Inodoro hit the nail on the head though, no explosion without a pressurized atmosphere(at least for gasoline)

For the record, that Mythbuster's episode was weird. They used tracer rounds to ignite the gasoline, after having shot the tank full of holes repeatedly. This thread is literally the only time I have found the memory of that episode remotely useful.

[Ryan]

Just for the record, I wasn't assuming the scientific definition of explosion just for the sake of not being a semantic prick, but I see I could have spared myself the effort of courtesy

I thought a fast burning / flash fire is close enough to explosion for this topics sake...

Bomb calorimetry gone wrong = epic

So I agree with all of you