Alfa Romeo Forums banner

1 - 20 of 31 Posts

·
Registered
Joined
·
4,501 Posts
Discussion Starter #1 (Edited)
I have seen this on some American cars, but never considered it for the Alfa...

Now that we are at the stage where we need to plumb the ATL fuel cell to the fuel-rails (with everything in between,) we need to lay it out and thought to consider this arrangement; (number of pumps, left-side versus right-side versus combined pick-ups and other semantics aside), what are the collective thoughts on the following:

In the conventional layout there are TWO fuel lines running to and from the tank ALL of the way to and from the two fuel-rails. The pump(s) and filter(s) are somewhere at the rear and there is a fuel pressure regulator somewhere at the front - with or without a gauge - usually plumbed in to the return-line.

In the alternative arrangement there is only ONE fuel line running from the rear to the front with the FPR mounted at the rear (close to the tank) on a T off of the sole line running to the front (with a loop from the FPR back to the tank) and dual lines only for that short distance between the tank and the FPR right next to it!

Somebody can maybe sketch this up for us but in essence you would eliminate the second line back to the front and the 2nd fuel-rail would in fact terminate at it's end (fuel comes in the one rail, loops around over to the other rail and then just ends there...)

Fuel pressure would still be regulated (and in our case adjustable) and bleed-off would still loop back to the tank via the SHORT loop at the rear (and now the overall setup will still be maintained at the given set pressure!)
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #4 (Edited)
We're not running vacuum-directed variable fuel pressure on these motors... (Dawie has suggested that we set our FPRs to 3 BAR and call it good!)

Also, keep in mind that we are not running MAP on these motors at all in fact - we are running TPS! (There is no vacuum in those plenums!) With the 6 individual throttle-bodies down there at the base of the intakes (right at the top of the intake ports), all of the vacuum happens BELOW that point and it is all throttle-plate and then intake port from there on down...! (Not much room for a vacuum-port there!

So, back to the question - any other thoughts on the plumbing arrangement?
 

·
Registered
Joined
·
4,090 Posts
You need a fuel feed and a return line! I don't understand how you want just one fuel line from rear to front. The pump puts out excessive fuel and the regulator returns fuel to maintain pressure at rail. So, regulator sits AFTER rail, and fuel pressure gauge needs to sit at fuel rail side of the regulator and NOT on the return line (but on feed line)! Maybe your custom regulator allows you to run single line...??? Are you sure it is constructed for that configuration? I suppose I should know since I picked them... my excuse is that it is a couple of years ago... :rolleyes: Maybe take a quick look at the instructions for the FPRs?
Jes
 

·
Registered
Joined
·
4,090 Posts
JJ,
Some rough pointers:
1. coarse fuel filter between cell and pump(s),
2. fine fuel filter between pump(s) and engine,
3. check valve if pump doesn't have build-in check valve (I think pumps for street cars typically do, but some, maybe most?, race pumps don't),
4. don't T two pick-ups together into one pump (my understanding is that it won't do what you want - you need one pump per pick-up and T afterwards)! Watch pressure ratings of fuel lines and dimension according to pressure x SAFETY_FACTOR.
Jes
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #7 (Edited)
Yeah-yeah - no we're good on the safety concerns, primary and secondary filters, dual (but separate) pick-ups etc etc.

I wanna talk about the idea in principal. (This is important so let's make sure to understand it before we dismiss it.) I'm not trying to make a case one way or another (except that it is simpler, lighter, costs less and it's easier to install and it is still effective!) GM happens to use it (not that they are the Alpha and the Omega here or anything), but it deserves a second look!

The set-up IS pressure regulated! Think of it this way; you have a direct line from the tank, to the filter(s), to the pump, to the first fuel-rail, over to the second fuel-rail and then you terminate and cap off after rail # 2!

Yes, now you have a problem - no regulator...

However, NOW you create a T off of that single line (right after the filter, still in the trunk RIGHT by the fuel cell as a "leak" - it bleeds-off pressure! That T loops back to the cell and feeds excess pressure and fuel FROM THE MAIN FEED back to the cell! Simple and brilliant!

In-line (on that T-loop back to the cell) is the FPR!

That T (and loop back to the tank) is essentially a "leak" (a bleed-off of the pressure that would otherwise ALL head on down-stream to the fuel-rails...)

Except now you have that T that bleeds off any excess pressure... ;)
 

·
Registered
Joined
·
4,090 Posts
I suppose that may work! Do it, and double check by having a separate fuel pressure gauge on the dead-end fuel line going to the fuel rail. Just mock up a system power up the pump and watch you fuel pressure gauges (one on the FPR and one on the dead-end, or skip the one we usually have on the FPR since what matters is the fuel rail pressure)! BTW, you didn't show the pump, but I assume it sits between the two filters, and the primary fine filter needs to sit before the T.

Anyway, looking at your drawing, I think it may work :D

Do it, and learn.

Jes
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #10 (Edited)
Sorry, yeah - I was revamping the drawing as you were typing - check it now... I'll change it some more to add the 2nd pump...

So, are we saying then that we would need the two pick-ups activated and that they each run to their own respective coarse filter - so TWO coarse filters (one for each pick-up line before each of the TWO pumps?) And then after the two pumps the line t's together and THEN one fine pump and then off to the front?
 

·
1966-2013
Joined
·
13,741 Posts
GM also happens to have a lot of failures with that system, or at least the garages I go to seem to have to replace a lot of them :shrug: (when you're talking at least 3-4 a week every week, well.......)

In general it works well for carburetted engines because the floats need lower pressure to have the ability to cut off the fuel flow. (meaning that pump pressure is still slightly higher than what is required for operation so that the bowls stay constantly full, even under heavy engine load/fuel consumption. ie: the pressure is just high enough to meet those demands, but just low enough to where the floats can cut off the flow so you don't flood out all the time)

I could see one feed line with both rails being serviced off it, then another tie together with a regulator post rail(s) to maintain pressure (with or without vacuum) and a return line to deal with the excess pressure.

Otherwise you're likely to end up with well metered pressure at low load and idle, and starvation under high load. (since you've set pressure at idle, it has no way to get more fuel to the rail when it's demanded)

Conversely, if you set pressure under high load/consumption, you'll end up running sickly rich at low low load and idle as the regulator wouldn't know there was less demand for fuel as it's doing it's thing well before fuel even hits the rail.

Put a regulator in post rail, with return line, and pressure should remain relatively steady throughout the rpm range and load levels as it can now 'sense' when the pressure drops under load, and it'll not allow that as it's been set to maintain X pressure so as long as the pump(s) can put out a decent excess of volume to keep up, pressure should remain constant.

Same goes for low load/rpm usage: the post rail regulator would simply bypass everything that was too much.

As to vacuum activating the regulator, it's not really a super neccesity as long as the pumps can maintain volume and pressure.

I've tinkered around with the stock Bosch regulator as found on the L-Jet systems by removing the vacuum line and threading in a long screw with locknut down through that port with good success. Tighten the screw, pressure goes up, loosen the screw, presure goes down, and it stays pretty consisitant pressure-wise across the load and RPM band if the pump(s) can produce adiquate pressure and volume.

It simply 'knows' that you want X pressure and will sustain it for as long as the pumps can keep pushing hard and fast enough.

Granted, you may experience a small flat spot upon initial hard acceleration, but that is more relevant to the pumps than it is to the regulator proper. (if they aren't redundantly powerful, they can't keep up with the very rapid pressure drops when the injectors suddenly go into garden hose mode, but they will catch up pretty rapidly)

If you really want to vacuum actuate the regulator, you'll have to make a sort of 'balance chamber' that feeds off as many cylinders as you can tap with nipples.

I've found a 1 1/4" diameter X 11 1/2" long pipe (plugged at both ends obviously) is adiquate enough to supply vacuum to a regulator and MAP sensor while still producing 11-12 inches of mercury -ive pressure at idle on a 4 cylinder. (most engines set up in the 'normal' fashion produce 15-20 inches of mercury -ive pressure)

The nice thing about building a chamber is that it can go anywhere you can reach hose to, so it doesn't neccisarily have to even be on the engine or manifold.

Granted, in your situation you're not likely to even be able to get nipples in to tap off, which is understandable. (however if you 'wanted' to do it, it could be done. mabe by making a spacer plate that fits between the TB's and head just thick enough to put nipples in. mabe with some creative angular drilling and angled nipples brazed in like throttle bypass adjuster ports)

I think you'll find setting up using TP rather than MAP will create a few difficulties AFA how the vehicle gets driven.

Certainly you can set it about as accurately as MAP, but the downside is that it won't have clue one on how to widen or narrow the fuel pulse based on engine demand like MAP can.

Net result is something that runs like a scalded dog when you are hammering on it, and gets about 4mpg under reuglar driving situations, or, gets 25-30mpg when puttering about, but that can barely get out of it's own way when you hammer it.

That's the downside of TP sensing: you can't have both.

Well, you 'could' set low level TPs for economy and high level for power, but then you end up restricting your driveability ranges to specific pedal positions/rpm ranges, which may or may not work all that well depending on the variety and speeds/rpms you intend to drive. (t'won't do no good to set it up to run economically at X rpm and powerfully at Y rpm, then suddenly go out on a road trip that lets you run just fast enough to be in Y, not X)
 

·
Registered
Joined
·
4,090 Posts
JJ,

Fuel filters need to be on opposite sides of the fuel pump. Both filters right after each does little for you. You usually would put a coarse between the cell and pump (secondary), and a fine after pump (primary). The latter would correspond to the one under the Milano at the right rear wheel. The Milano doesn't have the coarse filter. The idea is to have a coarse filter prevent whatever "pieces" of the foam in you cell that "washes" off from going into the pump. The fine fuel filter is your traditional fuel filter.

You may also need check valves - not absolutely necessary, but they will prevent fuel from flowing "backwards" when you shut down car, and hence ease starting.

Jes
 

·
Registered
Joined
·
2,618 Posts
I think a lot of Japanese cars also run this returnless fuel system? Well not completely returnless, but there was an article on autospeed.com about this. The more vacuum or fuel hose I can get rid from the engine bay, the better.
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #14
I could see one feed line with both rails being serviced off it, then another tie together with a regulator post rail(s) to maintain pressure (with or without vacuum) and a return line to deal with the excess pressure.
Thanks for that detailed brain dump! (This particular description though is how about a ga-zillion cars work any way - we understand it.)

Keep a few things in mind;
1) None of this (as far as what we are building) was done with an eye on practicality or efficiency.
2) These are full race cars - we want them to run like raped apes at WOT
3) We are NOT running MAP - not by choice rather the ITBs are right ON the intake port, but like I said - it works well.
4) We don't have the option to go MAP here - the positioning of the ITBs is a forgone conclusion - they are right on top of the intake port on the head - about half an inch away!
5) The first motor (expected/rated around 375 at the crank made 300-plus at the wheels here on pump gas. 10.7:1 compression and with 90% of the torque available at 2500 rpm!)
6) These are running, dyno'ed fully functional motors already!
7) One already ran here and was dyno'ed locally using the more conventional fuel-line routing (like what you described here...) We have video of the FPR and the gauge showing a near ZERO drop in pressure under full load on the dyno - it ALWAYS sat right on 3 BAR (idle or WOT - single stock pump!)
8) The mapping of the EFI is EXPECTED to take care of the variances in fuel demand (controls the duration of the injectors in milliseconds) and the injectors in turn are EXPECTED to do their job - only let through what is behind you as directed by the FPR!

The point of the fully-programmable system is just that! (This is NOT American hot-rod style where you cheat and bump the fuel-pressure to gain more fueling - now effectively pushing the entire fuel curve up STATICALLY across all rpms and across ALL load levels leaving you rich at idle and lean at WOT...)

This IS programmable and you DON'T want vacuum making your mind up for you on a race car. Again, the "prescription" was; LEAVE IT AT 3 BAR ALL OF THE TIME with plenty of pump in reserve to meet an increased demand from the FPR when it commands it (once the ECU allows it!)

With that in mind the FPR (and a single pump) did a GREAT job in maintaining 3 BAR for us at idle and under full load as monitored on the dyno!
 

·
Registered
Joined
·
2,984 Posts
This is a few things to think about..

when the injectors close the fuel flowing down the line will have to stop.
The fuel has mass.
So with no were to go it will hammer. hard on the injectors etc.
with a normal setup the fuel will just keep on going and not hammer the system to death.

If there is any air in the system (hiting a bump/ hard turn etc) the air will get traped and will have to be injected.

a normal system will purge the air.


If the motor gets hot and shut down the fuel in the rail will get hot could even be at a vapor state.

normal system purges the hot fuel as soon as you hit the key.
 

·
1966-2013
Joined
·
13,741 Posts
Re-re-reading: 'alt tank'? As in seperate auxillary totally seperate or baffled off from the main so as to be used in a sort of reserve tank or dual tank configuration as found on some pickup trucks?

Silly question, but is there any reason you couldn't put some form of float switch in the main tank so that when fuel level in it gets to a certain level, it activates a pump in the aux tank, which in turn cross pumps fuel from one to the other? (of course there'd have to be an interupter in the aux tank pump system so the pump would shut down once that tank was devoid of fuel)

It should save a lot of the hassle with T-ing off and such, as all the aux tank would need a course filter and pump with a line going into the main tank, then the main tank can deal with all the outgoing (and incoming if you go with a return) lines in a singular fashion.

Or, not knowing the arrangement from one tank to the other, just a balance pipe and no pump in the aux tank?

If the aux tank bottom is even an inch higher than the bottom of the main tank, fuel will flow downhill into the main tank, and the main tank will pump it til the tanks run dry. (the nicety of that is filling one will fill both simultaniously)
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #17 (Edited)
Jes, maybe just read my quote again - just to be sure... I AM saying a coarse filter BEFORE the pump (drawing has it backwards - actually it is right - what I am referring to as "Primary" is as in the first one - the coarse one - and by "Secondary" I am referring to the fine filter after the pumps - understanding FULLY that the finer one is the more important one - hence you were calling it the "primary" filter...)

So, are we saying then that we would need the two pick-ups activated and that they each run to their own respective coarse filter - so TWO coarse filters (one for each pick-up line before each of the TWO pumps?) And then after the two pumps the lines t together and THEN one fine filter (after that) and then off to the front?
My question was that since we now activate TWO pick-ups with a pump for EACH - and since we want "coarse" filtering for the pump(s) - does that mean that we now need TWO of the coarse (primary/secondary - whatever) filters - one for EACH pump since we want a corse BEFORE the pumps and there are two and we are NOT t-ing the pick-ups together BEFORE the pumps...?
 

·
Registered
Joined
·
4,501 Posts
Discussion Starter #19
Hhhmmm... Just went back and saw the other posts (while I was drawing...) I think that Slyalfa just gave me the definitive answer NOT to do it that I was looking for...
 

·
1966-2013
Joined
·
13,741 Posts
1) None of this (as far as what we are building) was done with an eye on practicality or efficiency.

2) These are full race cars - we want them to run like raped apes at WOT

3) We are NOT running MAP - not by choice rather the ITBs are right ON the intake port, but like I said - it works well.
Never said it wouldn't work, just that it won't work well in a daily driver. (you hadn't clarified the point that you were dealing with a race only engine 'til this moment that I can see)

4) We don't have the option to go MAP here - the positioning of the ITBs is a forgone conclusion - they are right on top of the intake port on the head - about half an inch away!
If you can't fit a vacuum nipple in a space as large as 1/2" you're not trying hard enough.... Moot point, as you're set up for TP already and thus got the hard part out of the way and are just trying to find alternative feed methods.

6) These are running, dyno'ed fully functional motors already!

7) One already ran here and was dyno'ed locally using the more conventional fuel-line routing (like what you described here...) We have video of the FPR and the gauge showing a near ZERO drop in pressure under full load on the dyno - it ALWAYS sat right on 3 BAR (idle or WOT - single stock pump!)
I sorta anticipated that had been tried, but that you were reaching for a bit of weight savings or ease of maintenance, not building a race car.

8) The mapping of the EFI is EXPECTED to take care of the variances in fuel demand (controls the duration of the injectors in milliseconds) and the injectors in turn are EXPECTED to do their job - only let through what is behind you as directed by the FPR!
'Expecting' and 'actual results' can sometimes be drastically different. Not saying that's the case here, but I've been 'expecting' to win the lottery for over a decade, though 'actual results' have certainly proven otherwise.

If you're pulling adiquate pressure post-rail with no drops or surges, then cool, you've got it licked and I'm happy for you.

The point of the fully-programmable system is just that! (This is NOT American hot-rod style where you cheat and bump the fuel-pressure to gain more fueling - now effectively pushing the entire fuel curve up STATICALLY across all rpms and across ALL load levels leaving you rich at idle and lean at WOT...)
Easy big dog. You're not the only person in the world working with programmable EFI.

This IS programmable and you DON'T want vacuum making your mind up for you on a race car. Again, the "prescription" was; LEAVE IT AT 3 BAR ALL OF THE TIME with plenty of pump in reserve to meet an increased demand from the FPR when it commands it (once the ECU allows it!)
You might not want vacuum and that's fine. But properly programmed MAP can be of use, or at least stands a better chance of not barfing the pipes off when unburned fuel ignites in them under engine braking as it can be programmed (there's that word again) to cut injector flow alltogether when the vac reaches whatever limit you set.

Surprisingly enough, economy is also considered a performance aspect, (unless you like pitting for fuel more than everyone else), if you're doing any kind of distance racing as opposed to sprint races. (and if you are just sprinting, why even bother with an aux tank at all?)

With that in mind the FPR (and a single pump) did a GREAT job in maintaining 3 BAR for us at idle and under full load as monitored on the dyno!
If you'd have said it was for full race, I wouldn't have bothered replying in regard to MAP or TP at all as it wouldn't have been pertinant, but since you didn't, I went with a summary and an idea or two in regard to them both.

I didn't realize that your link in the other section really meant that you didn't actually want anyone questioning the methodology based on what facts you didn't bother to give up front.

My apologies for interrupting the topic, as it sounds like you've got it all worked out anyway.
 
1 - 20 of 31 Posts
Top