First, a little background.... My fuel system so far starts with an aluminum, sumped tank with a -12 AN feed through a 74 micron prefilter to a Magnafuel MP-4301 pump, then through a 24 micron filter. Then it goes to 1/2" aluminum hard line to the firewall.

Lately, I have been putting a great deal of thought into how I will run my fuel system once it reaches the rails. I've been doing a lot of reading on this site as well as others on the pros and cons of series versus parallel configurations.

With that said, i'm not buying into the benefit/cost of parallel configurations. Don't get me wrong, I believe it is a great setup. However, I don't see how it is really any better than a serial setup, especially when you consider the added cost of the fittings.

My intention is to plumb an -8 AN line from my 1/2" hardline to the back of one of the rails. Replace the crossover tube with a prebent section of the same 1/2" hard line tig'd in place. And run another -8 AN line from the other rail to my SX regulator and 1/2" hard line back to the tank.

The only way that I could see a problem with injector starvation (which is the main arguement i've seen for parallel setups) is if I didn't have a sufficient fuel pump to keep up with the demand of the injectors. That would result in a pressure drop in the rails and I would expect all injectors to be affected. But the way I see it, that would cause a problem with either a series or parallel setup.

Long story short, am I missing something here? I would love to have an intelligent debate on this. All thoughts and opinions are welcome.

Thanks,
Marty

 

The problem with putting high flow through the rails in series is that the pressure drops the farther along the path you are. The last injector in the series sees a lower fuel pressure than the first one in the series. Think about it.... 8X the fuel demand for an injector (+ recycle flow) enters the end of the inlet rail, and drops to 7X, 6X.... etc. as it passes each injector. Since pressure is controlled by the FPR on the outlet end, the last injector will see a lower pressure than the first.

Unless you have an ECU that can trim the fuel to each cylinder, you now have to tune the entire engine for the leanest cylinder, which means 7 of them are going to be running richer. You're giving away power.

The parallel path means that only 4X fuel (+1/2 the recycle) enters the end of each rail, so the pressure loss in the rails is 1/4 of what it was at 8X flow (pressure loss is proportional to the square of the velocity), and there are only 4 injectors having the pressure controlled by the FPR on the outlet, and the pressure difference between the first and last injectors in the series is an order of magnitude less.

As far as using the hard line from the tank to the engine, NHRA safety rules require that all metallic tubing must be enclosed in a 16" length of 1/8" wall thickness steel tubing, where the line passes the "flywheel/bellhousing" area. Fuel lines can not be run in the DS tunnel. Braided S/S lines do not require the protective steel tubing.

 

Thanks for the reply Fred, very thought provoking. I see what you are saying, but in all fairness, i'm having a hard time believing that the fuel pressure differential from the beginning to the end would be drastic enough to notice a difference. With the stock system in place on my old bolt-on motor I never saw any difference in the fuel trims for either side. I would think that would be a good frame of reference, no?

Last night I was thumbing through the latest Hot Rod magazine while sitting on the, uhhhhh.....library! And I came across a tech tidbit on a 3,500HP motor for a 10.5W car that ran in the mid 6's at 230+. It was a twin turboed, 600+ cube motor with huge 240lbs injectors. In looking at the photos, I noticed it was running a series fuel arrangement.

I'm not arguing that parallel wouldn't be better, but i'm thinking that if series works at a 3,500HP level then my 800HP level would possibly be overkill and an unnecessary expense.

As far as the fuel line goes, does that rule apply even if i'm not running the line anywhere near the trans tunnel?

 

My point in explaining the reasoning was not to try and change your mind. The point was to explain that this is not something that is only applicable when you "didn't have a sufficient fuel pump to keep up with the demand of the injectors." It isn't really addressing "injector starvation". Its insuring that a well designed supply system is matched with a well designed distribution system. I didn't expect to change your mind - it was clear from your post that you weren't going to change it.

You can't extrapolate "....the stock system in place on my old bolt-on motor..." to a high performance application. Double or triple the fuel demand and you have 4X - 9X the pressure loss. Simple fluid dynamics. And, there's no way you can tell how much fueling difference there is between the first and last cylinder, unless you do a fairly thorough examination of all the plugs, shutting down the engine immediately after the pull, and having started with a brand new set of plugs, and doing it for a wide range of operating conditions. That's the advantage of doing the tuning, at least to establish the cylinder-to-cylinder variances, on an engine dyno. Easy to pull the plugs. It gets expensive using a new set for a large number of pulls, but the longevity of the engine makes up for it.

But, then I have no idea what sort of HP you are trying to fuel. Its apparently a lot more than what you show in your signature, because there is nothing there to justify the system you are designing. Apparently you are looking at some sort of extreme NA build, or a huge power adder.

I found it beneficial for insuring uniform pressure on an engine putting 800 HP of fuel through the stock rails, with a BSFC on nitrous well above 0.50 #/HR/HP. Also found it beneficial to run the system at 58psi, to minimize the impact of the pressure differences between the cylinders and pressure variations elsewhere in the system. I also found it beneficial to limit the supply and return lines to -6AN, to limit the loss of pressure on launch, due to G-forces acting on the mass of fuel in the lines.

I didn't just dream this up based on my technical training and work experience in designing fluid flow systems. It was a joint effort with some really smart engine builders, who had a lot more experience with similar systems. Its also pretty much in line with the technical material presented in the Kinsler Fuel Injection design manual.

The importance of this is related to the magnitude of the fuel demand, and the way the fuel is being used. In an engine running a 300-shot of dry nitrous, precise fuel delivery is extremely critical. We could have simply used the individual cylinder fuel trims in my MoTeC ECU to compensate for the different fuel pressures from cylinder to cylinder, as well as the nitrous and air distribution variances from cylinder to cylinder. By carefully managing the fuel pressure, and putting as much effort onto minimizing the flow distribution issues with the nitrous in a ported LT4 manifold, we were able to minimize cylinder-to-cylinder differences.

 

Wow, that is a lot to absorb. Thanks again for helping me have all of the facts before moving forward. Truth be known, I haven't decided yet. I'm just looking for some clarification on the pros and cons of each and so far you have provided excellent food for thought. I hope I didn't come across as confrontational, I just like to throw my thoughts and theories out there and see if they hold water (or in this case, fuel pressure ). Fluid dynamics is something quite foreign to me, I run data centers for a living and serial and parallel have a completely different end result when it comes to electronics.

I probably should have given a little more info on the current build. As you've guessed, this isn't for the combo in the sig. I'm building an all forged 383 at 9.1:1 compression with a single Innovative GT80-DBB R-trim through a 3 core air to water intercooler. Heads are old school AFR 190's that used to belong to Dave Diluca. I'm hoping to make somewhere in the 8's for horsepower.

Thanks again for you input.

Marty