View Full Version : Converting to return fuel system - question


demon
06-27-2006, 04:43 PM
Guys,

As mentioned in a previous post I am running a T04R boosted 2 rotor Cosmo engine in my RX8. I currently have converted the returnless system to a return system by returning to the opposite side of the tank to the fuel pump; the RX8 fuel tank is kind of saddle in shape.

The problem - I run out of fuel (or start having fuel pressure issues) as soon as I get about half full. I am using the stock pump and as mentioned above the return line returns to the opposite side of the tank. Would returning to the fuel pump side fix this problem? Any ideas why this might be happening? I would have guessed that some sort of mechanism would balance out the two sides of the tank somehow - does such a mechanism only work from the pump side to the other side and not vice versa?

Thanks in advance...

rotarygod
06-27-2006, 05:59 PM
The tank already has a small pump which pumps fuel between sides. You would be absolutely shocked at how fast the tank can empty from the fuel pump. Even though the engine is only using so much, fuel is constantly being returned to the tank. You could empty your entire tank in a matter of minutes. You are probably completely draining one side of the tank and filling the other side from the return line. I would put the return on the same side so this doesn't happen and then just let the stock pump that transfers between sides do it's job at it's own pace. You are probably exceeding it's pumping ability.

demon
06-27-2006, 06:00 PM
The tank already has a small pump which pumps fuel between sides. You would be absolutely shocked at how fast the tank can empty from the fuel pump. Even though the engine is only using so much, fuel is constantly being returned to the tank. You could empty your entire tank in a matter of minutes. You are probably completely draining one side of the tank and filling the other side from the return line. I would put the return on the same side so this doesn't happen and then just let the stock pump that transfers between sides do it's job at it's own pace. You are probably exceeding it's pumping ability.

This was close to my thought also, thank you sir!

RX8PR
06-27-2006, 06:54 PM
Demon,

We (rgonza and I) resolved this problem with
different changes,

The problem with the intank pumps is that you would be pumping
hot fuel (specially the returned fuel) and you would notice problems of fuel pressure specially in long trips.

I recommend that you install and external fuel pump,
first you would need to make some changes to the fuel pump canister (you would need to remove the factory pump from there) make the necessary changes to put the returned gasoline to fill the canister. The canister always would be full of gasoline no matter if you have less than a 1/4 of fuel tank it was designed that way.
The gasoline that is transfered from the other side of the tank is transfered directly to the canister that is the factory design and you would not change that.

Once we installed the external fuel pump we note that in long trip the primary pump begins to make noise, we resolve that installing a Aeromotive FPSC (fuel pump speed controller) this Sensing engine RPM, the FPSC kicks your fuel flow into high gear when you need it, but reduces pump speed and flow to keep things cool when you don't. It's like an automatic transmission for your electric fuel pump, you can find more info here :http://www.aeromotiveinc.com/pdetail.php?prod=31

but then in more long trips we note that at high boost we begin to see some fuel pressure loss, we decided to install fuel pump cooler before the gasoline return to the tank , you can buy it at Summitt or you can make one from a oil cooler.

Our setting use two fuel pumps the primary and the secondary that enters at 3,000 rpms,
plus we have auto meter nexus fuel pressure gauge installed inside the cars to ALWAYS know our fuel pressure.

the picture is not tthe actual set up but is very similar.

Manuel

dannobre
06-27-2006, 06:58 PM
Actually the "pump" between sides of the fuel tank is a venturi driven suction from the outflow of the fuel pressure regulator on the drivers side that transfers fuel from the passenger side. I works well....at about 1/3 tank it has pulled almost the fuel to the drivers side of the tank....the only problem is that fuel transfers across with "g" loading in corners

demon
06-27-2006, 07:02 PM
Actually the "pump" between sides of the fuel tank is a venturi driven suction from the outflow of the fuel pressure regulator on the drivers side that transfers fuel from the passenger side. I works well....at about 1/3 tank it has pulled almost the fuel to the drivers side of the tank....the only problem is that fuel transfers across with "g" loading in corners

If this is the case, why is the drivers side tank empty at (what the gauge says is) half full? I am returning fuel to the passenger side, so surely this mechanism should still work correctly and pump fuel over to the drivers side(no modifications have been made to the drivers side fuel pump assembly other than to wire it directly to the haltech)?

demon
06-27-2006, 07:04 PM
Demon,

We (rgonza and I) resolved this problem with
different changes,

The problem with the intank pumps is that you would be pumping
hot fuel (specially the returned fuel) and you would notice problems of fuel pressure specially in long trips.

I recommend that you install and external fuel pump,
first you would need to make some changes to the fuel pump canister (you would need to remove the factory pump from there) make the necessary changes to put the returned gasoline to fill the canister. The canister always would be full of gasoline no matter if you have less than a 1/4 of fuel tank it was designed that way.
The gasoline that is transfered from the other side of the tank is transfered directly to the canister that is the factory design and you would not change that.

Once we installed the external fuel pump we note that in long trip the primary pump begins to make noise, we resolve that installing a Aeromotive FPSC (fuel pump speed controller) this Sensing engine RPM, the FPSC kicks your fuel flow into high gear when you need it, but reduces pump speed and flow to keep things cool when you don't. It's like an automatic transmission for your electric fuel pump, you can find more info here :http://www.aeromotiveinc.com/pdetail.php?prod=31

but then in more long trips we note that at high boost we begin to see some fuel pressure loss, we decided to install fuel pump cooler before the gasoline return to the tank , you can buy it at Summitt or you can make one from a oil cooler.

Our setting use two fuel pumps the primary and the secondary that enters at 3,000 rpms,
plus we have auto meter nexus fuel pressure gauge installed inside the cars to ALWAYS know our fuel pressure.

the picture is not tthe actual set up but is very similar.

Manuel

Thanks for the post Manuel, I'm a tiny bit confused though; have you completely removed the stock pump or are you still running it and have an external pump in line as well? Also in your system, surely the passenger side of the tank is completely unused as you are only returning fuel to the drivers side and have no mechanism to transfer fuel between both sides of the tank (if you have removed the stock pump).

dannobre
06-27-2006, 08:22 PM
If there is no fuel to pump through the pump..and thus the venturi...no suction...and no fuel transfer from the passenger side. You would be better to return the flow to the pump unit on the drivers side like OEM.

demon
06-27-2006, 08:42 PM
If there is no fuel to pump through the pump..and thus the venturi...no suction...and no fuel transfer from the passenger side. You would be better to return the flow to the pump unit on the drivers side like OEM.

Surely there is only no fuel once the drivers side is empty though; surely, for example, if the drivers side is only half full it should still be pulling enough fuel in from the passenger side and as a result never really allow itself to be completely empty?

dannobre
06-27-2006, 11:52 PM
Here's my take on the problem I have seen. At high fuel usage...ie track or possibly turbo under boost with higher flowing injectors..the fuel pump doesn't have a lot of extra capacity to dump out the pressure regulator/venturi...so the siphon slows down...

In LH turns the fuel gets thrown to the passenger side......and makes the problem worse

If it can't fill the cup the fuel pump sits in..it runs out of fuel....and then the car dies...usually at a really inconvenient time.

demon
06-28-2006, 12:01 AM
Here's my take on the problem I have seen. At high fuel usage...ie track or possibly turbo under boost with higher flowing injectors..the fuel pump doesn't have a lot of extra capacity to dump out the pressure regulator/venturi...so the siphon slows down...

In LH turns the fuel gets thrown to the passenger side......and makes the problem worse

If it can't fill the cup the fuel pump sits in..it runs out of fuel....and then the car dies...usually at a really inconvenient time.

That's an interesting theory about the capacity; surely if I return to the drivers side, I will still have a similar problem as the passenger side will now never be used and I am again likely to run out of fuel at half tank, no?

dannobre
06-28-2006, 12:10 AM
If you return to the top of the unit like the picture above...it will spill into the cup around the pump...it won't contribute to the venturi effect...but it will keep the pump in fuel like the OEM regulator return.

I have added a second pump to the passenger side...but I won't know if it solves my problem till my next track day on July 28th :(

demon
06-28-2006, 10:28 AM
If you return to the top of the unit like the picture above...it will spill into the cup around the pump...it won't contribute to the venturi effect...but it will keep the pump in fuel like the OEM regulator return.

I have added a second pump to the passenger side...but I won't know if it solves my problem till my next track day on July 28th :(

I was thinking of a second pump too; it seems even if you return to the top of the unit the passenger side fuel will never get used and thus, you will still run out of fuel at half tank?

RX8PR
06-28-2006, 01:45 PM
Demon,

to clarify the situation we eliminate the factory pump from inside the tank but we keep the ventury system working, we also use external fuel pumps, our primary fuel pump suck fuel from the tank (only from the driver side of the tank).

We don't like inside tank fuel pump because the heat generated. But you can make that way, In the picture that I sended yesterday, we converted the factory fuel line to a return line and the bigger brider line it was the fuel line. That picture is old and that system was with the inside fuel tank we replaced the factory one with a walbro 255 and fits ok on the factory canister.

Now with the 3 rotor we used bigger fuel line and return lines and installed inside the canister a piece of pipe that is near 1/2 inch of the bottom of the tank, we can run our car to "E" (of empty) without problem.

Manuel

rgonza
06-28-2006, 07:28 PM
Surely there is only no fuel once the drivers side is empty though; surely, for example, if the drivers side is only half full it should still be pulling enough fuel in from the passenger side and as a result never really allow itself to be completely empty?

Hi Demon, This fuel system works like the factory ones , it's take the fuel from the pasenger tank with its factory ventury that's we install in the fuel return line this unit keeps some pressure diferential to keeps the ventury working it's in same order like factory. If you,re looking for 400 rwhp + i suggess to you to install to in line fuel pumps remenber turbocharged rotarys engine it specifict brake comsuption is around .65 to .70 you're need a lot of fuel to keep the engine happy and safe with afr 10.8.

Regards
Rgonza :)

demon
06-12-2007, 12:17 PM
I just want to bring this up from the dead as I finally got this running.

I'm still having problems with fuel pressure fluctuations at half-full tanks of gas. I think I've totally ignored the venturi system which is creating the problem.

Here's what I've done, maybe someone can help me:

I removed the stock fuel pump and filtersock from the stock canister. I then drilled a bulkhead fitting into the top of the canister (just like the picture above) to use as my feed line. From this fitting, INSIDE the canister, I connected a short piece (maybe 3" long) of -6 hose from the bulkhead fitting down into the canister so that it would be pulling fuel from roughly the same spot as the stock fuel pump inlet which I removed.

I am then using the stock feed line as my return line. I left the little plastic tube connected to the other side of the tank connected to the canister. With this setup though I am encountering problems in the half-full range.

Can someone explain to me what is involved in getting the venturi system working, or how others have done this in their setups?

dannobre
06-12-2007, 08:01 PM
If your not running the fuel return from the FP regulator into the venturi...you are getting no fuel return from the passenger side. The venturi needs fuel flow to create suction on the black tube inside the tank that pulls fuel from the passenger side.

demon
06-12-2007, 08:33 PM
So, using the stock fuel feed line as my new return, how would I plumb this? I currently have it from my FPR -> stock feed line -> canister

dannobre
06-12-2007, 08:54 PM
The FP regulator discharge needs to be plumbed to the venturi where the stock regulator discharged.. that's it...

GTAW
06-12-2007, 10:10 PM
If you left the reservoir intact inside the fuel pump assembly, you could connect the return to the venturi (as mentioned), To give you a visual, granted the barbs are different sizes...
http://img373.imageshack.us/img373/4788/fuelpumps004qd6.th.jpg (http://img373.imageshack.us/my.php?image=fuelpumps004qd6.jpg)
http://img373.imageshack.us/img373/8760/fuelpumps005ic0.th.jpg (http://img373.imageshack.us/my.php?image=fuelpumps005ic0.jpg)
I also wonder how much pressure would build up routing it this way. It seems like the venturi would work well, but what about the function of the return line?