Angrey's custom Fuel System Thread

[Angrey]

So it's been a long time coming, but I've finally come to rest on my fuel setup and ready to report and discuss. There's a lot to unpack, so I'll lay out the big picture in this first post and then we can expand and discuss further as we go.

A little background, I planned on creating a fuel system for my GT350, that would accommodate boosted power levels, but I also wanted it to continue to provide features and capabilities of the GT350 and not just a drag car, but a daily driver with all the refinement and acumen associated with typical use (i.e. long trips). I also wanted E85, which significantly reduces the typical range of a full tank. I also wanted it to be capable for 4 digit power, so returnless systems (currently) were not really an option.

Although there are many tried and true return style systems on the market, I couldn't get past the limitations that most guys running high horsepower just accept as life in the big city. I didn't want a fuel system that had to be above 1/2 tank in order to be assured the pumps wouldn't starve under acceleration. Combined with the E85 range referenced above, this would mean that on long trips, I'd get maybe 75 miles of driving before I'd either have to refill OR drive the rest of the tank out like an old woman (what if I wanted to stomp the car with 1/3 of a tank remaining?).

Additionally, even the OEM system has inherent limitations/flaws that I wanted to correct. The current generation of mustang with IRS necessitated a mid-car fuel tank location which leads to a "saddle" style tank, which is essentially two separate reservoirs separated/combined by an elevated connection between them over the drive shaft tunnel. Ford isn't alone here, the GM/Corvette crowd deals with the same frustrating setup (more on that later). This saddle style system means that either a siphon system or a separate lift pump is required to move fuel from the "passive" side over to the active side where the fuel pumps are located. It also means that during certain driving conditions (hard left turns) fuel can and does flow up and over the saddle into the passive side (lowering the level of the active side and increasing the concern of naked fuel pumps and fuel starvation).

Given that I would be spending a large amount of money on building the power train, reliable and unfailing fuel delivery was a top priority and I didn't like the typical limitations of current aftermarket offerings.

I also didn't like the idea that driving the car hard in left turns would also exacerbate fuel delivery safety concerns.

I also didn't like the current tradeoffs with staged pumps (risking a failed Hobbs switch or running multiple pumps at 100% duty while heating the fuel and straining the electrical system for long periods of time on long trips).

I looked very hard at surge tank options which would essentially eliminate virtually all of the starvation concerns. But I couldn't find a suitable place under the car/firewall to mount available surge tanks and I wasn't crazy about having an additional fuel assembly inside the cabin (or losing the functional space in the trunk).

Additionally, I was very attracted to the new(er) brushless style pumps that had many advantages over the old brush style pumps. From durability, to true variable capability to reduced electrical draw and efficiency.

It seemed one of the first pioneers of an in-tank, brushless fuel pump return style system, capable of 4 digit power fuel delivery was Fuelab. They offered a 1200 hp capable in tank/bucket brushless pump for a very reasonable price. The problem is, they stopped making them because no one bought them. They were figuratively a decade ahead of the market. After speaking with them, they put a lot of R/D into that system/kit and virtually no one bought it. So they weren't interested in the effort to replicate the system on an S550 bucket.

So I set out to replicate my own version of their setup, and use their brilliant electronic fuel pressure regulation. Fuelab has a patent on their electronic FPR which indexes pump duty based upon return line pressure. There are many ways to increase the fuel delivery on a variable system, you could reference RPM, you could reference a hobbs/pressure switch, etc. But to have true variability that's incremental (and not just some stepped effort), the best solution on a return style setup is to monitor the return line. When the return line is flowing too much, the system can reduce pump duty. When the return line is flowing too little, the system knows the motor is consuming more fuel and can increase pump duty.

I also set out to resolve the "left turn" issue and issues surrounding passive/active separate tanks. Many guys over on Trackmustangsonline have tried everything from surge tanks (not a preferred option) to hydramat in the saddle to fuel foam surrounding the primary pump buckets. I wasn't a fan of the siphon system, both from a flow standpoint (it's more a trickle than a flow) or the pressure/flow bleed off the primary pumps required to power it.

The result is a set of low pressure lift pumps on the passive side which actively pump fuel from the passenger side over to the active side. Not only do they flow about 400 liters/hour combined, but now the primary pumps can focus on providing engine fuel and not bleeding flow off on a siphon setup.

These low level pumps draw about 3.5 amps each. Combined with the efficiencies of brushless pumps, this means at full power draw/flow, the entire system would draw about the same power as a pair of TI285 pumps, and at idle/low load, would result in the current draw just a little more than a single 285 pump running full power.

So I had the roadmap to a system that would deliver gobs of power on E85, would run much cooler at idle/cruising, would only increase heat/power draw at WOT and spirited conditions, would be more reliable and resistant to pump wear/failure, and would continuously move fuel from the passive to the active side, without reducing the primary pump output.

Combined with the Radium bucket, which acts as a miniature surge tank (that has a one way baffle at the bottom to allow fuel in, but not out during acceleration or turning), the whole system is NEARLY perfect for what I wanted and not markedly more expensive than a top of the line triple pump return style system with controllers, regulator, lines, etc.

Here's a list of the final setup:

1) Radium bucket with integral 8 electrical posts for pumps and 2 electrical posts for level sender.
2) 2 each DW 440 brushless fuel pumps with variable PWM controllers.
3) Fuelab electronic pressure regulator (for PWM signal), with vacuum line boost reference
4) 2 each DW micro lift pumps for transferring fuel from passive to active side
5) The rest are pretty common wiring and fuel line arrangements. 8 A/N feed line/fittings. Typical dual pump wiring kit with relays and fuses. The exception is that I repurposed the OEM feed line to use as my return line (equivalent to a 6 AN)
6) Radium Microglass in line fuel filter.
7) Whipple fuel rails (I researched and tried to figure out how to run aftermarket on a whipple Gen5 but gave up. The whipple rails aren't that big of a restriction).

I tried to feature fuel pulse dampeners (Radium) but they need to be close to the rails to be fully effective and there just isn't room to plumb them close to the rails without making the engine bay look crowded and tacky).

The last issue I've been trying to resolve (frustratingly) is the issue of running the passive lift pumps dry. I found an Autometer Elite Pro Control Fuel level gauge that IN THEORY will control the pumps and ask them to run when the fuel level on the passenger side is there, but shutoff automatically when it's empty. After much brain damage trying to get an autometer fuel level gauge to work with Ford's overly complex fuel sender design, I was able to get the fuel gauge to work, however not without disconnecting the passenger side sending from the BCM/PCM. Additionally, even after 3 calls with Autometer, I can't get their product to work as advertised. In the end, I punted and installed a toggle switch that I can manually turn the lift pumps on/off. The autometer gauge triggers them when the level is high, but it turns off at about 3/4 level (instead of 1/8 or lower where I want it). That's what I've been trying to debug for the last few weeks.

So in the end, I now have a fuel system that will support over 1100 rwhp (verified) on an 8 AN Feed, 6 AN return with no booster. It runs low draw (about 7 amps each DW and 3.5 amps each lift pump) at idle/cruising and draws about 35 amps at full tilt (less than two TI285s at full boogie). I don't have to worry about fuel starvation even at very low fuel levels as the bucket surrounds the pumps and the lift pumps are supplying the bucket as well as the one way check valve. I don't have to lose any space in the trunk (or have a fuel tank inside the firewall with me).

So now I have a car that's detuned to 1000 rwhp that I can turn or stomp whenever I want down to very low fuel levels and not worry about leaning out and destroying my motor. I'm going to continue working on the automated setup for the lift pumps with the Autometer gauge.

Sponsored

 

[Cory S]

Love the radium buckets. I run the internal return style setup with the regulator in the hat. So simple with many advantages in addition to not needing an actual return line.

 

[Angrey]

Some additional note and lessons learned:

1) I had to run an additional relay to use the OEM power from the PCM as my trigger source. I tried using the yellow wire with gray tracer that comes from the FPDM as the trigger to the Fuelab electronic FPR, but it did NOT like that at all. I then wired using the Violet-Green power wire which works, but then the pumps and controllers run with accessory or key on. In order to retain the initial prime and off feature like the OEM setup, I had to install an additional relay that triggered based on the Yellow-Gray, but powered using the Green Violet. This both resolved the power supply to the FPR but also operates just like stock (initial prime the off, and run when the car is running). The Yellow-Gray tracer does provide enough current and voltage to operate a typical relay, but because it's PWM and variable, the FPR wasn't happy. Initially we had some tuning challenges surrounding the WOT enrichment tables and the Magnum XL signal for torque calcs, so we could have tried to simply adjust the tune fuel tables to give full effort, but rather than introduce a new unknown I just elected to physically wire around the issue.

2) I modified the fuel pickup assembly on the passenger side by cutting material on the bottom of the pickup to make room for mounting the lift pumps. I then used simple plumbing T connection and an adapter to repurpose the OEM siphon line as my feed going over to the primary bucket.

3) When we were having problems with the FPR and the trigger signal, I eliminated the factory fuel filter in the now return line. A simple set of quick connect adapters was a better solution than just turning the filter around.

4) When we were having problems with the FPR, I considered swapping out the FPR spring assembly on the return line fitting. The Fuelab FPR comes with a little spring loaded diaphragm that gives the FPR it's return line pressure readings. I had bought (but didn't end up using) a simple gate valve fitting that I was going to try to manually adjust in order to set the "idle" duty. We didn't end up using the gate valve once we solved the issue from the trigger source, but I do think it would work if you want the system to flow more at idle/base.

5) I used the DW bulkhead adapter on the passenger side. it's a drill through the top of the plastic hat and sandwich through a top and bottom electrical connector but it leaked when you fill the car up all the way to the top of the filler neck. I had to reinstall it using JB tank weld to seal it and so far it's worked to eliminate the leaks.

6) The OEM fuel sender from Ford is ridiculously complicated. It features a powered signal and return system rather than a typical voltage - resistance - ground setup that the rest of the automotive world uses. Luckily the Autometer gauge has a feature where you can "calibrate" the range from full to empty. However, it would not work wired up to the OEM connections with it. So I had to remove the OE connection and convert the sender to a more traditional setup. This means my factory fuel gauge in the dash now reads just over half tank for fuel (only the driver's side) and then my autometer gauge gives me the information for the passenger side. This allows me to see when the driver's side starts to drop and either turn on the toggle switch to transfer fuel (or ideally have the autometer run and control it automatically). When the car is full, the trigger works and the lift pumps continuously send fuel over. But once the passenger side drops to about 3/4 for that side, they shut off for some reason. I haven't been able to figure that out with Autometer. As a fall back, if I see the driver's side dropping more than I'd like, I simply flip the toggle switch and fill it back up with the lift pumps. Not as elegant or seamless as I'd like, but it's working for now. I mounted the toggle switch in the back of the center console compartment where it's easy to access while driving, but out of the way and not inadvertently activated by bumping it with something.

7) Despite my frustrations at the time, no one offered a bucket that would accommodate the BKS1000 brushless pumps (commonly referred to as the Veyron pump). The Radium bucket can now be configured to run the BKS 1000 pump which might be a more economical route than twin DW440's and paired with eliminating the siphon system, would probably be good for around 1000 rwhp on E85 but definitely way more with a typical booster added. I had legit concerns the Fuelab regulator would "talk" to the DW controllers properly but DW assured me as long as the PWM signal fell within an extremely wide frequency range, it would work. I'm betting the BKS1000 pump would also work and might be a simpler (and cheaper) setup than twin 440's. Fuelab also offers their own in tank brushless pumps now (they didn't at the time) and while they're pricey as well (all brushless pumps are expensive due to the additional controller) it's another option that would be interesting to pursue for others wanting this same style setup.

 

[Angrey]

 

[Angrey]

 

[Decible]

So it's been a long time coming, but I've finally come to rest on my fuel setup and ready to report and discuss. There's a lot to unpack, so I'll lay out the big picture in this first post and then we can expand and discuss further as we go.

A little background, I planned on creating a fuel system for my GT350, that would accommodate boosted power levels, but I also wanted it to continue to provide features and capabilities of the GT350 and not just a drag car, but a daily driver with all the refinement and acumen associated with typical use (i.e. long trips). I also wanted E85, which significantly reduces the typical range of a full tank. I also wanted it to be capable for 4 digit power, so returnless systems (currently) were not really an option.

Although there are many tried and true return style systems on the market, I couldn't get past the limitations that most guys running high horsepower just accept as life in the big city. I didn't want a fuel system that had to be above 1/2 tank in order to be assured the pumps wouldn't starve under acceleration. Combined with the E85 range referenced above, this would mean that on long trips, I'd get maybe 75 miles of driving before I'd either have to refill OR drive the rest of the tank out like an old woman (what if I wanted to stomp the car with 1/3 of a tank remaining?).

Additionally, even the OEM system has inherent limitations/flaws that I wanted to correct. The current generation of mustang with IRS necessitated a mid-car fuel tank location which leads to a "saddle" style tank, which is essentially two separate reservoirs separated/combined by an elevated connection between them over the drive shaft tunnel. Ford isn't alone here, the GM/Corvette crowd deals with the same frustrating setup (more on that later). This saddle style system means that either a siphon system or a separate lift pump is required to move fuel from the "passive" side over to the active side where the fuel pumps are located. It also means that during certain driving conditions (hard left turns) fuel can and does flow up and over the saddle into the passive side (lowering the level of the active side and increasing the concern of naked fuel pumps and fuel starvation).

Given that I would be spending a large amount of money on building the power train, reliable and unfailing fuel delivery was a top priority and I didn't like the typical limitations of current aftermarket offerings.

I also didn't like the idea that driving the car hard in left turns would also exacerbate fuel delivery safety concerns.

I also didn't like the current tradeoffs with staged pumps (risking a failed Hobbs switch or running multiple pumps at 100% duty while heating the fuel and straining the electrical system for long periods of time on long trips).

I looked very hard at surge tank options which would essentially eliminate virtually all of the starvation concerns. But I couldn't find a suitable place under the car/firewall to mount available surge tanks and I wasn't crazy about having an additional fuel assembly inside the cabin (or losing the functional space in the trunk).

Additionally, I was very attracted to the new(er) brushless style pumps that had many advantages over the old brush style pumps. From durability, to true variable capability to reduced electrical draw and efficiency.

It seemed one of the first pioneers of an in-tank, brushless fuel pump return style system, capable of 4 digit power fuel delivery was Fuelab. They offered a 1200 hp capable in tank/bucket brushless pump for a very reasonable price. The problem is, they stopped making them because no one bought them. They were figuratively a decade ahead of the market. After speaking with them, they put a lot of R/D into that system/kit and virtually no one bought it. So they weren't interested in the effort to replicate the system on an S550 bucket.

So I set out to replicate my own version of their setup, and use their brilliant electronic fuel pressure regulation. Fuelab has a patent on their electronic FPR which indexes pump duty based upon return line pressure. There are many ways to increase the fuel delivery on a variable system, you could reference RPM, you could reference a hobbs/pressure switch, etc. But to have true variability that's incremental (and not just some stepped effort), the best solution on a return style setup is to monitor the return line. When the return line is flowing too much, the system can reduce pump duty. When the return line is flowing too little, the system knows the motor is consuming more fuel and can increase pump duty.

I also set out to resolve the "left turn" issue and issues surrounding passive/active separate tanks. Many guys over on Trackmustangsonline have tried everything from surge tanks (not a preferred option) to hydramat in the saddle to fuel foam surrounding the primary pump buckets. I wasn't a fan of the siphon system, both from a flow standpoint (it's more a trickle than a flow) or the pressure/flow bleed off the primary pumps required to power it.

The result is a set of low pressure lift pumps on the passive side which actively pump fuel from the passenger side over to the active side. Not only do they flow about 400 liters/hour combined, but now the primary pumps can focus on providing engine fuel and not bleeding flow off on a siphon setup.

These low level pumps draw about 3.5 amps each. Combined with the efficiencies of brushless pumps, this means at full power draw/flow, the entire system would draw about the same power as a pair of TI285 pumps, and at idle/low load, would result in the current draw just a little more than a single 285 pump running full power.

So I had the roadmap to a system that would deliver gobs of power on E85, would run much cooler at idle/cruising, would only increase heat/power draw at WOT and spirited conditions, would be more reliable and resistant to pump wear/failure, and would continuously move fuel from the passive to the active side, without reducing the primary pump output.

Combined with the Radium bucket, which acts as a miniature surge tank (that has a one way baffle at the bottom to allow fuel in, but not out during acceleration or turning), the whole system is NEARLY perfect for what I wanted and not markedly more expensive than a top of the line triple pump return style system with controllers, regulator, lines, etc.

Here's a list of the final setup:

1) Radium bucket with integral 8 electrical posts for pumps and 2 electrical posts for level sender.
2) 2 each DW 440 brushless fuel pumps with variable PWM controllers.
3) Fuelab electronic pressure regulator (for PWM signal), with vacuum line boost reference
4) 2 each DW micro lift pumps for transferring fuel from passive to active side
5) The rest are pretty common wiring and fuel line arrangements. 8 A/N feed line/fittings. Typical dual pump wiring kit with relays and fuses. The exception is that I repurposed the OEM feed line to use as my return line (equivalent to a 6 AN)
6) Radium Microglass in line fuel filter.
7) Whipple fuel rails (I researched and tried to figure out how to run aftermarket on a whipple Gen5 but gave up. The whipple rails aren't that big of a restriction).

I tried to feature fuel pulse dampeners (Radium) but they need to be close to the rails to be fully effective and there just isn't room to plumb them close to the rails without making the engine bay look crowded and tacky).

The last issue I've been trying to resolve (frustratingly) is the issue of running the passive lift pumps dry. I found an Autometer Elite Pro Control Fuel level gauge that IN THEORY will control the pumps and ask them to run when the fuel level on the passenger side is there, but shutoff automatically when it's empty. After much brain damage trying to get an autometer fuel level gauge to work with Ford's overly complex fuel sender design, I was able to get the fuel gauge to work, however not without disconnecting the passenger side sending from the BCM/PCM. Additionally, even after 3 calls with Autometer, I can't get their product to work as advertised. In the end, I punted and installed a toggle switch that I can manually turn the lift pumps on/off. The autometer gauge triggers them when the level is high, but it turns off at about 3/4 level (instead of 1/8 or lower where I want it). That's what I've been trying to debug for the last few weeks.

So in the end, I now have a fuel system that will support over 1100 rwhp (verified) on an 8 AN Feed, 6 AN return with no booster. It runs low draw (about 7 amps each DW and 3.5 amps each lift pump) at idle/cruising and draws about 35 amps at full tilt (less than two TI285s at full boogie). I don't have to worry about fuel starvation even at very low fuel levels as the bucket surrounds the pumps and the lift pumps are supplying the bucket as well as the one way check valve. I don't have to lose any space in the trunk (or have a fuel tank inside the firewall with me).

So now I have a car that's detuned to 1000 rwhp that I can turn or stomp whenever I want down to very low fuel levels and not worry about leaning out and destroying my motor. I'm going to continue working on the automated setup for the lift pumps with the Autometer gauge.

Angry,

I was think of doing what you were doing except using 2 radium buckets. One in each saddle with a DW440. Did you look into this and if you did is it a bad idea?

 

[Angrey]

Angry,

I was think of doing what you were doing except using 2 radium buckets. One in each saddle with a DW440. Did you look into this and if you did is it a bad idea?

I know a guy with a Holley setup that has a primary pump and bucket on each side. The problem is there's no ability to balance the two sides then in the event of fuel shift for turning or if there's any significant disparity between the flow/consumption of each side. And without a way to monitor each side, you really have no idea whether there's a fuel level disparity between the two.

Just for some perspective, now that I have a dedicated gauge for the passenger side, I can see the fuel move from one side to the other through gauge level shift when turning hard out of a left or right turn or a long sweeping turn on the highway.

At one point I even considered a leveling tube/pipe underneath the drive shaft to balance the tanks, but then decided that was inherently unsafe for a number of reasons.

However, a modified version of what you're proposing is to run a bucket in each side and use the passenger side as a lift bucket to send fuel over to the primary side. Radium has actually done this and the fuel bucket will fit flipped orientation.

 

[Decible]

I know a guy with a Holley setup that has a primary pump and bucket on each side. The problem is there's no ability to balance the two sides then in the event of fuel shift for turning or if there's any significant disparity between the flow/consumption of each side. And without a way to monitor each side, you really have no idea whether there's a fuel level disparity between the two.

Just for some perspective, now that I have a dedicated gauge for the passenger side, I can see the fuel move from one side to the other through gauge level shift when turning hard out of a left or right turn or a long sweeping turn on the highway.

At one point I even considered a leveling tube/pipe underneath the drive shaft to balance the tanks, but then decided that was inherently unsafe for a number of reasons.

However, a modified version of what you're proposing is to run a bucket in each side and use the passenger side as a lift bucket to send fuel over to the primary side. Radium has actually done this and the fuel bucket will fit flipped orientation.

So you are saying is use the second radium bucket in the passenger side to lift fuel and dump it in the driver side? I could see that working with a y-block and tapping the return line….haven’t thought of that. I was going to y-block the feed line from each bucket into the main feed line.

 

[Angrey]

So you are saying is use the second radium bucket in the passenger side to lift fuel and dump it in the driver side? I could see that working with a y-block and tapping the return line….haven’t thought of that. I was going to y-block the feed line from each bucket into the main feed line.

Yeah, I think he has each side feeding a bank (so essentially in parallel).

There's a zillion ways to deliver the fuel to the motor, it's just how to deal with the 2 saddles and the pros/cons of each arrangement.

I'm certain the approach you're suggesting will work, and it's probably a small chance of issue with one side being higher than the other. Admittedly, I tried to eliminate or greatly reduce even miniscule/rare issues, from an OCD standpoint.

 

[Decible]

@tj@steeda Tj it just made me think….how are you all controlling the fuel slosh from the saddles in the yellow and red track cars? Surge tanks?

 

[gimmie11s]

Very cool. Radium makes some quality shit for these cars!!

 

[illtal]

I have a comment and a question.
1: fuelab is still having trouble procuring my controllers from its' distributors. Been waiting since FEB! but I have a radium hat that I will be testing E85 on a returnless 10 AN line setup so I hope this works out fine (the dual regulator setup is probably not needed).

2. @Angrey can you not run a float to "break" the circuit on a relay when you have a low fuel level on the passenger side saddle? That would ensure that the primary saddle would always be full of fuel for street or track applications and prevent dry burnout.

 

[illtal]

@tj@steeda Tj it just made me think….how are you all controlling the fuel slosh from the saddles in the yellow and red track cars? Surge tanks?

I'd bet they are using surge tanks. Open flow to a surge the stock pump can handle.
I've seen setups that use a PNR tank as a surge so imagine having 7 extra gallons of fuel to burn thru so you can get back to the pit or where ever needed.

I wouldn't want to lose the spare on a street car though.

 

[Angrey]

I'd bet they are using surge tanks. Open flow to a surge the stock pump can handle.
I've seen setups that use a PNR tank as a surge so imagine having 7 extra gallons of fuel to burn thru so you can get back to the pit or where ever needed.

I wouldn't want to lose the spare on a street car though.

I was going to speculate that it's a surge tank. If not, most die hard track guys don't run the fuel lower than 1/3-1/2 tank. It's well documented for loss of power out of long left turns.

 

[Angrey]

I have a comment and a question.
1: fuelab is still having trouble procuring my controllers from its' distributors. Been waiting since FEB! but I have a radium hat that I will be testing E85 on a returnless 10 AN line setup so I hope this works out fine (the dual regulator setup is probably not needed).

2. @Angrey can you not run a float to "break" the circuit on a relay when you have a low fuel level on the passenger side saddle? That would ensure that the primary saddle would always be full of fuel for street or track applications and prevent dry burnout.

1) I bought my regulator from a distributor/vendor (not direct through fuelab) which is why it's Fushia colored (haha) but because I routed the regulator and vaccum block to my battery well, it's out of sight.

2) I looked at Radiums fuel float but didn't want the brain damage of how to fab it up in the OEM saddle tank. In theory, the autometer works off the same principle using the factory swing arm float. Just haven't been able to get it to work properly. When the fuel float is high, it activates the pumps, but then stops them after it drops to about 3/4 instead of keeping it active all the way down to around 1/8. Ford didn't make it easy. The factory fuel gauge/sending unit isn't the normal resister type. Ford uses a two wire system. I have it wired now with the traditional single wire resistance setup, but it took using the gauges custom calibration to make that work (and of course means I can't run it with the 2 wire feeback back to the BCM).

On a positive note, when the lift pumps are running, it works like a charm. I can watch the autometer gauge dropping. The factory gauge (the 1/2 remaining on the driver's side) has much more of a time lag to it. But the passenger side, the lift pumps can suck it down in just a few short minutes. It's not optimal right now, but I have a lit toggle switch and so if the fuel level is lower and I'm going to stomp the car, I just turn it on. Any fuel that's in the driver's side well enters the bucket through the check valve and any fuel that's in the passenger side is being picked up and pumped into the bucket.

Sponsored