Fuel pressure dropping off with increased boost

Hi Guys,

Sorry if this is perhaps sounds a bit amateurish but I am a stuck...

I have an R32 GTR with a hard wired Walbro 255, stock fuel lines and rail, 1000cc Bosch injectors and a Turbosmart FPR800 (previously Nismo). Now, when I get into positive boost, my fuel pressure is not meeting the expected fuel pressure. As you can see from the log, the amount of error is proportional to the amount of positive manifold pressure. What I would like to ask is; is this normal? Are the injectors still getting the correct base pressure?

I initially thought is was to do with the pressure slope of the FPR. I changed from a Nismo to the Turbosmart and the results are exactly the same..

What do I do?

Many thanks in advance!

Nick Cox

Hi Nick, it isn't too much of an issue and something I see fairly often, and in most cases, unless I was having an issue with the fuel pressure dropping off I wouldn't waste too much time chasing as I would work around it knowing it is what it is.

But if I was looking to try and fix it then I would be looking at things that affect the fuel flow like pump voltage and good ground under load, checking the fuel flow returning to the tank at that load ( it should have approx 3 liters per minute returning to the tank)

A very common problem is people not upgrading ALL the power supply wiring to the pump, and the earth/ground, and this can drastically reduce the ability of the pump to operate to the expected standard. "High performance" pumps will normally have a much higher current draw than OEM, and this will in turn increase the voltage drops in the wiring and reduce the voltage across the pump and, remember, as a rule of thumb the power is proportional to the square of the voltage - if you have a 20% drop in the wiring, you will only have ~64% of the potential power the pump could provide - in extreme cases, you could have less flow than the OEM pump.

I'd start by checking the voltage across the pump terminals when it's operating (or as close to it as practical) and compare that to the battery voltage - you should be able to get it to within a half volt without too much hassle, and closer would be preferable.

Hi all,

Thank you so much for getting back to me!

Before I posted yesterday, I back pinned and hooked up my multimeter to the fuel pump terminals in top of the hat. I took it for a drive and recorded a very slight voltage drop. It was averaging 13.5v at light load and around 13.38v at higher load. Would this amount account for the for the fuel pressure drop I am getting?

My wiring is gauged between 12 and 14awg all the way from the battery, relay and ground. I had to splice into the original top hat connector wires with the pump, into the stock in tank wires. The ground is bolted to the ATTESA mounting bracket. I have a feeling this could be the possible issue?

Big thanks again for your time! I am busy learning all about this. Your feedback is very welcomed!

Cheers,

Nick

Ah, cool you've already tackled the wiring, you can't do much better than that across the pump!

Double checked, and it's claimed to only support up to (approximately) 500WHP - which is easy for even a relatively mild build of that engine to exceed, and especially if running a high nominal rail pressure. This may be the root of your problems.

If you're happy the pump spec's match, or exceed your fuel demands - it will also drop drastically with pressure increases - next thing would be to check the filters are large enough and aren't partially blocked - this includes and in-tank pickup as well as in-line.

that is pretty good voltage but check that earth even add an extra to it for testing, but make sure you measure the fuel returning to the tank it is most important

checking the fuel flow returning to the tank at that load ( it should have approx 3 liters per minute returning to the tank)

Hi Guys,

I feel the pump should be within my spec for the time being as around 400whp is my goal at the moment. I set the base pressure to 3 bar (300kpa) as per the injector spec sheet. In terms of the filters, I have recently replaced the stock inline and installed a new sock when I upgraded the pump. I am pretty sure there isn’t a massive restriction.

Unfortunately, I don’t have the facilities to measure fuel return at the moment. However, I shall look into my wiring set up again and see if there are any bottlenecks in the system. I shall also extend the ground to (hopefully!!) a better point in the chassis and see what happens!

Big thanks again! Watch this space! 👍

Nick

Do you have a manifold pressure sensor and a boost pressure sensor by any chance? I noticed your comparative channel is fuel pressure and expected fuel pressure, how is this being generated?

Usually if it is a fuel pump running out of flow, you end up with an unsteady pressure reading, as the inability to maintain volume and pressure in the rail, makes the pressure reading far more prone to injector pulsations, and you end up with an unsteady trace (unless the channel is logged slowly and the sensor voltage has a heavy filter on it, which may be the case with Haltech logging.)

Being that its an R32GTR, which port did you use for the fuel pressure reference? The front port on the case balance pipe above the injectors? Is the manifold pressure sensor reading from the same place? The hose on the back of that cast unit is still definitely attached? Do you have any other devices attached to that port? I'm going to take a stab and say that the fuel pressure expected is calculated from a base fuel pressure value that you have entered, and it uses the manifold pressure signal to calculate it.

Variance in manifold pressure signal or inaccuracy of the reference pressure used in the expected fuel pressure (was the 300kpa determined priming the pump, or with the engine running and the reference line removed from the reg?) are more likely reasons for a consistent and linear discrepancy.

Hi Nathan,

I am running a 4 bar map sensor T’d into the rear vacuum/boost line that goes to the stock boost gauge sensor - see pic. The FPR takes its reference from the front port of the balance bar as it does stock. You are correct that the expected fuel pressure is worked out from the map sensor creating a the expected differential fuel pressure.

I worked out the base fuel pressure by disconnecting the vacuum line from the FPR with the engine running, 3 bar/300kpa. Connecting the vacuum line back on, the fuel immediately went to expected fuel pressure.

There are nothing else connected to the balance bar.

What are your thoughts? I shall double check the sensor calibration also..

Thanks dude!

Nick

Is the fuel pressure just lagging behind the target and will eventually catch up if you run at steady state or is there always an error in fuel pressure? Can you post a log that includes RPM and Manifold Pressure also? Understanding exactly what the engine is doing in these logs would help.

How is the fuel pressure control setup? (electronic/mechanical; 2D target table referencing manifold pressure with PID adjusting for error?) I'm not familiar with GTRs but this appears to be a control issue, not an undersized pump or plumbing issue.

Is the estimated fuel flow graph showing per cylinder fuel flow or total fuel system fuel flow? My initial assumption was per cylinder which raises a red flag because it would mean your 1000cc injectors are running at or near 100% duty cycle to achieve the logged 1046cc/min estimated flow. If you post another log, include injector duty cycle also. This shouldn't cause the fuel pressure target error you are seeing but might indicate there is a configuration issue in the software.

One other thing to check for a leak in the manifold pressure reference line. Make sure it's not plumbed inline with a boost controller that is bleeding off air to control the wastegate.

Hi all,

I have tried pretty much everything... I have double checked the ECU sensor calibrations, re-checked voltage and ground - all great, fuel filter is good, changed the pump strainer sock, plumbed in the FPR into the same vacuum line as the map sensor, changed FPR and still the same result!

The only thing that I feel could be creating this issue now, is the diameter of the vacuum line. Its a 4mm line from the vacuum bar (rear end) to the MAP sensor and a 3mm line (front end) to the FPR. Would this be a reason? My instinct says not...

Its all rather frustrating to say the least. I just want it running mint before I take it to the dyno.

DynoDoug - Please find attached another screenshot including injector duty cycle - looks good to me.

Many thanks again guys!

Nick

Again have you checked the basics as I stated before. Return fuel flow ! It is one of the most basic checks to do and one of the over looked. If the flow is as stated above move on to something else. The target flow is a calculated target it isn’t always achievable, don’t sweat the little things if the return flow is good then it is getting the required fuel to the injectors.

Hi Ross, thanks for your reply! Unfortunately I don’t have the facility to do this at present. I am just road testing. Like you said, I am probably overthinking it a bit… I just want it to be spot on - the OCD in me! 😂 I just can’t think why it wouldn’t be right… I have done everything bar measuring the fuel. I’ll try and measure it when its at the dyno. 👍

Nick,

Thank you for posting another log with injector duty cycle and manifold pressure. I am able to determine a few things from this:

1. The fuel system is not undersized or restricted in any way based on the maximum fuel flow achieved during these test. The fuel pump is operating well, no dirty filters, no undersized fuel lines, etc. This can be determined because when maximum boost is reached (roughly 20:12-20:15) and injector duty continues to raise, the fuel pressure holds constant. If there was a flow issue with the fuel system, then the actual fuel pressure would drop.

2. The ECU is not controlling the fuel pressure in any way, it is just calculating Fuel Pressure Expected by taking you base fuel pressure (300kPa) and adding the Manifold Pressure (65kPa at max boost) for an expected 365kPa. This may seem obvious but was considering that you may have electronic fuel pressure control.

I assume you know, but just in case, the reason the fuel pressure raises is to maintain a constant pressure differential across the injectors as manifold pressure varies. This is a nice to have feature but not absolutely necessary for the injectors to deliver the proper fuel. It's the fuel pressure regulators job to adjust the fuel pressure and it does this mechanically via a diaphragm, spring and vacuum line. That part is very important to understand. This mechanical device does a good job, but not a perfect job. (Think of it as a carburetor in the sense that it does it's job well but you can't ask too much from it in terms of accuracy) If you were doing R&D studying the effects of fuel pressure on combustion efficiency, then this error would be a concern and electronic fuel pressure control would be required. That's not the case though, you just need to meet your target AFR.

The good news is that the fuel pressure regulator is doing it's job as well as it can be expected to. When manifold pressure raised, so does the fuel pressure, roughly proportional. As Ross mentioned, this error is something that can be worked around. When the fuel maps are tuned it will essentially just require a small amount of injector duty that grows proportionally with boost. Just by hitting your target AFRs, you will be compensating for the fuel pressure without even realizing it.

One thing you can do if you want to minimize the fuel pressure error as much as possible is just to make sure there is no leak in the vacuum lines during boost. There's no easy way to really test this but putting zip ties on any fitting that are just slip fit will help. If they are rubber lines that are 20+ years old then consider replacing them with new rubber or silicon ones. You only need to worry about lines that are on the same plumbing circuit as the fuel pressure regulator. Again, it's not really an issue though and is just to make yourself happy, not to make the car happy.

In summary, nothing is wrong. It's just the accuracy limitation of a mechanical device. Enjoy your time at the dyno! (I wouldn't waste any of it checking fuel flow)

Are you using gauge pressure to work out your expected pressure or absolute?

Hi Doug, thank you very much for your detailed reply!

And many thanks go to everyone else to. Myself and my OCD feel much better now! 😂 The car will be booked for the dyno, hopefully within a few weeks. Very much looking forward to it and putting all what I have learned through the various lockdowns to good use!

Chris - Its working off gauge pressure, std 3bar/300kpa fuel pressure.

Cheers, Nick

Hi Guys,

I just thought I'd give you an update on this issue... I went to a local dyno, tuned it as was and enjoyed the car throughout the summer. It drove well! However, I still knew this little issue needed resolving..

So, over the last week I checked everything again, pump voltage, potential restrictions and I even put a bigger pump in! Still scratching my head... Then I thought, 'I wonder if its the fitting that holds the fuel pressure sensor??'. After a bit of re-jigging, I put the fuel pressure sensor into the port on the FPR800. PROBLEM SOVLED!! The original fitting was obviously the issue, causing the restriction. see pic.

Moral of the story, don't use one of this type!

Thanks for all your input!

Cheers,

Nick

I'm glad you figured it out, Nick! Do you think the fitting was just restricting quick pressure changes to the sensor or was it actually restricting fuel flow to the injectors? In other words, are both pathways through the tee restricted or just the one going to the sensor port?

Doug

Yup, it seems its both pathways! It was confirmed as I’m now running rich everywhere… 🤣 I initially thought there was a kind of Venturi effect on the sensor, but nope. All interesting stuff! 👍

Cheers Doug,

Nick

Ah good on you for finding it. You can see now the importance of checking the fuel return flow,

Yes indeed Ross! 👍 thanks a lot!

Nick

Just curious, where in the fuel system was the sensor tee'd into previously? I was surprised this restriction didn't present itself in the logs the way I expected. I initially thought the sensor was on the fuel rail but doesn't seem to be the case!

Hi Doug,

It was placed in the fuel line between the fuel filter and the fuel rail. I didn't particularly want to make serious modifications to anything... See pic. The only thing I can think of is that it was causing some kind of turbulence/cavitation within the moving flow - increasing with demanded flow/pressure, linearly increasing the pressure drop? Anyway, I wont be using one again! :-)

Nick