Engine shuts off with half tank of fuel left

[Murray_Calavera]

Hi all,

So I have an issue where the car appears to run out of fuel with half a tank left. Basically everything is perfect, then when fuel drops to about half a tank, fuel pressure rapidly drops to 0 and the car stalls. Once the tank is filled everything is 100% perfect again. This is not a fuel gauge/level sender issue, when refilling the tank to full it only took 28 litres. 

I had a look around inside the tank today and confirmed the fuel pump is exactly where it should be and I didn't see any other issues. 

My thought process now is, if the fuel pump is unable to supply fuel while being completely submerged, the fuel tank must be building a vacuum as the tank empties.

Does this sound correct? I'm thinking a check valve has failed somewhere in the EVAP system that is allowing a vacuum to slowly build while the car is running.

I'm thinking the next step in diagnosis is to run the car until it stalls again, then disconnect the charcoal canister at the line that connects to the fuel tank and see if that removes the vacuum. This is only a guess though so would appreciate any thoughts on this matter!

[r32-25t]

Has the sock fallen off the bottom of the fuel pump 

[Murray_Calavera]

I really didn't want to pull the hanger out if I didn't have to, so I haven't checked to see if the sock has fallen off. 

Would it be possible for the pump to be fully submerged and yet unable to supply fuel because the sock had fallen off? I can't picture why it would work with a full tank and not at half tank when in both conditions the pump is fully submerged.

[r32-25t]

I’m just assuming it’s a 33 or 34 and the pump sits halfway up and tank and the sock goes to the bottom of the tank 

[Murray_Calavera]

That does make the most sense, I'll pull it out to confirm. 

If the sock has fallen off, it will go back in with a Hydramat setup.

[blind_elk]

I would expect if it was a vacuum problem, it would be a gradual deterioration of performance, not a sudden shut down. But, if it's a vacuum problem, then why not simply remove the fuel cap and see if the engine will run.

[Duncan]

Per above, it is the cap that is meant to vent air into the tank, so just undo it to eliminate that option. 

[Murray_Calavera]

So of course the strainer had fallen off and I was just overthinking everything =/

Really not impressed with Deatchwerks. 

* Advertises the fuel pump kit to suit "93-98 Nissan Skyline R33" https://deatschwerks.com/products/9-401-1043?_pos=18&_sid=5047b1750&_ss=r - of course it doesn't fit the R33 stock bracket. Nor do their instructions mention anything to do with the R33. The kit is obviously generic and not vehicle specific. 

* No mention of the fact that it doesn't have an internal check valve - really enjoyed tuning my cold start with that, thanks. 

* Wiring on the supplied pigtail was garbage. The wire insulation melted while applying heat shrink over the open barrel splice connector I used to connect the wires. 

* Strainer for the fuel pump is held in with hopes and dreams only. Lasted about 2 years before falling off. 

They can't get any of this right but they go to the effort to remove the Bosch part number from the fuel pump housing.

So I bought a 525 LPH Aeroflow so the next strainer I use has a better chance of hanging on. Fingers crossed this time. 

[GTSBoy]

I was going to ask if we'd fixed this yet. The suspense was killing me.

[tridentt150v]

Most aftermarket pumps are not a direct replacement...they all work once correctly mounted, but this may involve some 'customisation' - on a small scale.

The same can be said for many aftermarket products.....they are designed for cars, and made for profit......not skylines!!!

[Murray_Calavera]

100% appreciate what your saying about fitment of aftermarket parts in general. Deatschwerks marketing however make some pretty specific claims, this is copy paste from there website - 

https://deatschwerks.com/collections/pumps

APPLICATION SPECIFIC FITMENT

Application specific product development is a complex process, but DW has made it easy for YOU… all you have to do is choose your year, make, and model.

DeatschWerks’ Product Development Team carefully engineers each pump application fitment to optimize pump format + flow combinations; each kit is a carefully curated assortment of electrical connectors, filters, hose, clamps, and spacers, and completed with step-by-step installation instructions.

Their claims are full of shit. Even if you only grab the low hanging fruit, they say there are step-by-step instructions, well no, there isn't. Their instructions for the R33 are the instructions for the 300ZX.

I don't think I'm asking too much in expecting the company selling a product not to lie to me about what the product does and what supporting documentation will accompany the product. 

[BK]

I agree and get what you're saying, but only Nismo 280lph will ever be exact direct replacement.

[Murray_Calavera]

So interestingly, the place I went to buy the Hydramat from talked me out of buying it. We spent about 30 min playing with various Hydramats, adaptors and fittings, they even called a big name performance workshop and we had a chat about options. Was quite impressed with them trying to help me rather then focusing on a sale. 

The short version was, they strongly advocated to try a setup with the fuel pump mounted at the very bottom of the tank first. If I still wasn't happy, they thought the money going on a Hydramet should be put towards a surge tank setup. 

So this is my setup based on their advice. Fuel pump mounted at the very bottom of the tank then I chose a strainer that extends forwards along the bottom of the tank. 

Please excuse the fabrication skills (no one can see it inside the tank lol). Also, I'm sure it's not 100% perfect however it's the best I could get it using a shitty bore scope and my limited patience. (I really wish I had a fuel tank to sacrifice buy chopping out the side so I could clearly how everything was sitting inside). 

If someone wanted to use a similar setup, the advice I would give them is - the angle of the fuel pump is very important, it really wants to run parallel along the bottom arm as I've indicated with the green lines. In general, the clearances are tight, however where I've circled in red was the biggest challenge I had with clearances to allow the bracket to completely slide down. 

The parts I used are Raceworks fuel strainer FPS-067 and Aeroflow fuel pump AF49-1057.

Ignore the below. I went googling for these details and found nothing, hopefully it helps at least one person in future. 

R34 skyline fuel pump bracket length. R33 skyline fuel pump bracket length. R33 skyline fuel pump bracket extension.

[niZmO_Man]

Thank you for sharing. No doubt this will help people in the future.

[Murray_Calavera]

Just some more data from this setup as I wanted to change from e85 back to 98. 

Drove until there was about 15 litres left in the tank. Parked next to the fuel bowser and used the fuel pump to empty the tank (I'm glad the petrol station staff were cool, I checked with them that this was OK before hogging the bowser). 

Once the fuel pump stopped flowing fuel this was the result - 

56 litres to fill the tank.

Mixture went from e80 to e10.

Slightly disappointed it couldn't draw up the last 9 or so litres in the tank, but it is what it is.

[Murray_Calavera]

So on my previous attempt to drain the tank, I must have had those 9 litres in the other saddle tank. 

Turns out if the car is driven normally and the venturi is working as intended, this setup can run the car empty. 

I was trying to run the car low to calibrate my fuel level gauge and in the end ran the tank down to 5 litres remaining. I was driving around relatively quickly and got zero fuel surge. 

So in the end I am thoroughly impressed with how well this setup works. I can't believe that it's possible to have 5 litres in the tank and not experience any fuel surge!

[joshuaho96]

On 11/7/2022 at 9:42 PM, Murray_Calavera said:

So on my previous attempt to drain the tank, I must have had those 9 litres in the other saddle tank. 

Turns out if the car is driven normally and the venturi is working as intended, this setup can run the car empty. 

I was trying to run the car low to calibrate my fuel level gauge and in the end ran the tank down to 5 litres remaining. I was driving around relatively quickly and got zero fuel surge. 

So in the end I am thoroughly impressed with how well this setup works. I can't believe that it's possible to have 5 litres in the tank and not experience any fuel surge!

The real test as with anything is going to be running on track pulling high Gs. I’ve heard the R32s especially have problems with fuel starvation if you run the tank too low. 

[Dose Pipe Sutututu]

1 hour ago, joshuaho96 said:

The real test as with anything is going to be running on track pulling high Gs. I’ve heard the R32s especially have problems with fuel starvation if you run the tank too low. 

hence those in-tank twin pump kits for R32s or S13s don't work well for track use.

[Murray_Calavera]

2 hours ago, joshuaho96 said:

The real test as with anything is going to be running on track pulling high Gs. I’ve heard the R32s especially have problems with fuel starvation if you run the tank too low. 

I was on semi slicks, nothing crazy, but spirited street driving. With only 5 litres in the tank I was very impressed that there was zero surge. 

I'm not saying that this is the solution to all fuel issues for the track, more so that all the hours of frustration with the bore scope and making tiny adjustments to the angle of the bracket over and over again really felt like it payed off. 

Where the stock bracket mounts, having the fuel pump sit halfway down in the tank, this setup is a vast improvement over that.

It will probably be a couple of months before I can get on track, but next time I do I'll log fuel pressure and run the tank down and see when fuel surge starts to be an issue.

[joshuaho96]

8 hours ago, Murray_Calavera said:

I was on semi slicks, nothing crazy, but spirited street driving. With only 5 litres in the tank I was very impressed that there was zero surge. 

I'm not saying that this is the solution to all fuel issues for the track, more so that all the hours of frustration with the bore scope and making tiny adjustments to the angle of the bracket over and over again really felt like it payed off. 

Where the stock bracket mounts, having the fuel pump sit halfway down in the tank, this setup is a vast improvement over that.

It will probably be a couple of months before I can get on track, but next time I do I'll log fuel pressure and run the tank down and see when fuel surge starts to be an issue.

I’m definitely curious about going a similar route in my R33 once I have time to start messing with the fuel system. I really want to try running 997.1 Turbo EV14 injectors and a 440 lph brushless pump for an E85 conversion. I figure my comically tiny turbos should be enough of a limiting factor to get by on ~630cc injectors if I bump the fuel pressure enough.