Fuel getting too hot after 90 min drive in warm weather

Good day folks, i have a Toyota Mr2 with a "stock" 3sgte turbo engine in, it is run off a ecumaster emu classic, and he car has been running great for 4 months now,

During a long drive today in 30°c ambient heat and medium traffic, i noticed the car running leaner via my AEM 4.9 wideband gauge in the cabin, i was able to drive home, the worst was on light load cruising when the AF ratio was lamda 1.05 and 1.1 on slight throttle opening for a few seconds,

Normally the same cell would be lamda 1. And that was what i saw when i started my commute

I stopped to I checked all the fuel lines and stuff and didn't see anything wrong,

But when I opened my gas tank Cap I noticed a lot of pressure and a lot of heat exiting the pipe. There is a walbro 255lph fuel pump in the tank but I suspect it might be overkill on light loads like today, and it is heating the fuel because it's running at 100%. The Oem 3sgte wiring harness have a fuel pump relay with a resistor in one circuit to keep the pump on a "low" speed, I didn't have that relay unfortunately and didn't think I'd need it.

Basically I'm just asking if this sounds like the problem to you guys too?

I take it that you've got the fuel system set up to return to the tank, rather than a swirl pot?

It could certainly contribute to it. In simple terms, the electrical power you put into the pump has to go somewhere - and most of this is going to be as heating, both of the pump assembly - which is primarily çooled'by the fuel - and compression of the fuel. For a 10A current and 12V drop across the pump that would be approaching 120W of heating. Whether this is the primary problem I can't say - however, the engine compartment on these cars can get very hot, and the fuel will also be picking up a lot of heat from just passing through it. I would suggest that normally the fuel lines and fuel tank are cooled by the ambient air and on hot days the air passing under, and around, the fuel tank, and the heat radiated from the road is going to be higher - IIRC, the tank is enclosed, so even less cooling.

Reducing the voltage/current to the pump should help reduce things, but how far? Other options may be to add a cooler to the return line, or buy/make a scoop that directs air to the engine compartment. Some consideration of the fuel line paths and heat shielding should also help.

The pump actually does very little heating. The bulk of the heat comes from circulating fuel through a conductive metal fuel rail that is attached to a cylinder head that is at 90°C, as well as much of the rest of the fuel system is in the hot engine bay or running alongside the exhaust underneath.

Of the 120W power consumption that Gord quotes, he missed that a vane pump has a mechanical efficiency of probably 90%, and a brushed DC motor efficiency is probably in a simlar ballpark. So, of that 120W, at a rough guess probably about 90W of it is converted directly into the pressure increase, the heat generated is more like 30W.

I can have the pump running all day in my injector flow bench which only holds ~10L fuel and it barely gets lukewarm.

Nope, I didn't miss it - basic physics, the pressure increase is a transfer of energy that heats the fuel ;-) There will be some external heat energy transfer to the surrounding air and through the mounts, but most will be heating the fuel. It's exactly the same principle as compressing air, or any other substance, and you're probably most familiar with this with forced induction.

I don't see it being significant, either, though, especially for this application, when you consider all the other factors. As you point out, the energy is easily lost to the surroundings with a small temperature rise in the tank. You may even be able to feel a small temperature difference before and after the pump, and the pressure regulator - I'd expect it to be slight when you consider the fuel mass flow and it's heat capacity?

Adiabatic heating only occurs on a compressible gas, fuel is generally considered to be a incompressible fluid.

This is turning into a good discussion thread :-) - and on that, came across this interesting discussion thread - https://cr4.globalspec.com/thread/55255/Energy-Loss-in-Vane-Pump-and-Gear-Pump

Yup, was getting back to that - posted way to late in the night, before.

The reasons "adiabatic" wasn't, and shouldn't be, used in this context is that it isn't a sealed system, with electrical energy being introduced to it, and that energy being lost throughout the rest of the fuel system back to the tank - are you thinking of it just in the 'ideal gas' context?

Yes, it was a bad example of energy being applied to compress a liquid (NOT a "fluid"), as it's clearly introduced potential for confusion (not least on my part). There's going to be a mixture of ways the electrical energy will be lost within the system, considering the pressure and velocity heads, and the inefficiencies throughout - even without the energy lost from the pump's electrical inefficiencies being partly introduced as heat to the fluid.

Perhaps I should have used hydraulics? 300 bar working pressure and much higher flow rates required large oil coolers on the return (not as efficient, but a fraction of the pressure meant much thinner wall tubing could be used for better heat transfer). Even with that cooling to the reservoir the pump, and oil, could get hot enough for burned hands, or arms, on the pressure side if one was careless, or trying to reach dropped tools :-(.

Without getting into a science discussion I probably won’t win here.

I would assume if your fuel is getting that hot then you should notice a difference in pressure. Oddly enough I’ve seen a few cases lately in the states where the ambient temps added to high performance street engines and the heat they produce was acting in combination to “super” heat the fuel in the lines. This caused the fuel to “vaporise” in the lines. It was picked up initially by a lean lambda reading. Then secondly by a low fuel pressure warning.

it was confirmed by bleeding the fuel system and witnessing bubbles. This was in easily reached fuel cell. So if you are using a stock set up this may be difficult to replicate.

hope it helps in some way.

Thankyou very much for all your replies.

This ended up in a very good discussion.

A lot of good points were made.

My biggest concern was, how bad would it eventually get?

If I had to drive another 2 hours would I get worse and worse? Or would the temperature rise rate of the fuel eventually slowed down and be kept steady by the (not so efficient) airflow under the car etc.

I had a look into my closed loop EGO feedback on the Ecumaster. And I was a bit confused when the system didn't add fuel on the non transient fuel maps when I drove 120km/h.

I gave it a bit more power and fine tuned the proportional and integral values and the integral limit. I abruptly lowered the fuel table to see how the Closed loop will catch it and it did a very good job,

So my other question would be, when the fuel temperature starts to raise, does the AFR go leaner because there is actually less energy in the fuel or because there is actually air bubbles in the fuel lines and when the injectors open up?

For the time being I removed the plastic covers underneath the fuel tank to hopefully help with airflow to the tank, because with those covers the fuel tank basically is enclosed.

I did a long drive of about 3 hours last night and I saw no lean problems, and the ego correction was very low to non existent. I don't know if it was the lower evening temperatures and airflow over the tank.

I'm also planning on adding a relay to the fuel pump with a big resistor so under low load and rpm the pump will be running at much less power to hopefully keep the flow of fuel lower, and lower the amount thats returning to the tank,

From what I’ve seen personally. The lean issues with hot fuel is a result of vaporised pockets of fuel for lack of a better word. This leads to fuel pressure drop behind the injectors and potentially no liquid fuel when the injectors open. So you obviously don’t get the right fuel mixture. The problem the ECU has is it’s not constant so it’s a difficult thing for it to learn and adjust to as its constantly changing. Opposed to a constant fuel delivery issue where it can narrow down what it needs to change. Or alert you to a direct issue.

If I was you personally I think my first concern would be narrowing down the reason the fuel is getting that hot. It may be you need a cooler or perhaps the engine bay temps are too high. Maybe try driving with some type of ambient sensor in the engine bay or try removing the rear engine cover and simply see if it makes any difference. I personally don’t think the fuel pump will be an issue. Unless there’s some kind of outside reason to cause it to heat up. I know from my experience often having 3 or more pumps running full tilt on race cars and never see fuel temps much over ambient. So perhaps some simple experiments might find you your answers. Like the removing of covers you’ve already tried.

good luck.

The primary reason for the fuel 'leaning out' at the higher temperatures is because volume=/= mass. As the fuel is heated it becomes less dense - this means there is less mass (actual) fuel for the same volume. The EFI is set to deliver specific volumes of fuel so without compensation there will be less mass of fuel being burned. There 'may'also be a tiny variation in the injector opening because the electrical resistance increases with the temperature - but I can't see that being relevant, unless you're already marginal.

Leigh has a good point about possible localised boiling if the temperature gets high enough, or the pressure low enough, but I would expect this to be more of a problem starting a hot engine, rather than a running one - certainly might be wrong, though.

As we've commented, the engine compartment, and engine, will be very hot, so insulating the fuel lines and, maybe, running a cooler (in a cool/well ventilated position) on the return, would be a good idea to reduce the heat transfer to the fuel, and help remove what there is.

There is quite a lot of physics involved, but under light fuel demand, and especially with a pump that uses the fuel for cooling, almost all the electrical energy is going to be ultimately transferred to the fuel, proportional to the use:return*. The greater the proportional fuel demand, the less fuel carrying the energy back to the tank. However, while I don't see this as too much of an issue compared to the other heat sources, it will contribute.

Normally, the heat input to the fuel is dissipated through the fuel lines and the surface of the fuel tank, and as you had covers the transfer to the surrounding air was compromised and that may be expected to be why removing the covers and exposing the tank surface has lowered temperatures.

*if anyone's actually interested, I can give a 101.

This is a very interesting thread actually. I don't really have any issues, but when my car is fully up to temp, my idle afr's are a certain figure, and then after a longer drive, my idle afr's then become leaner as heat soak starts to take effect?!

If it was to do with high intake temps at idle due to heat soak, this would actually make the afr's richer wouldn't it? due to a less dense air entering the chamber?

No, the airflow is cooler than the manifold, the perceived temperature is higher than actual charge temp, so fuel is injected for less than actual air mass.

ah ok, so due to air intake correction being wrong basically, you do end up with a lean mixture?

Yes