Fuel Pressure Relating To Hose Size 1

[StephT]

I'd really appreciate some help with the design of a fuel supply for a fuel injection conversion of an engine currently fuelled by twin carburettors. The engine is a 4 cylinder 1.2 litre Honda from the 1970s. It currently has 6mm ID fuel lines and hoses feeding twin 34mm CV carburettors. For the conversion I will be fitting 4 x 38mm individual throttle bodies that were sourced from a 2006 Honda CBR600 motorcycle. The use of these types of throttle body assemblies has become commonplace on the later series 4 cylinder Honda engines, but to my knowledge only one person has done such a conversion on the 1.2 litre engine, and with great success. Fortunately I have a good relationship with this person, but his engine was developed for full race conditions alone, whereas my application will predominantly be for the street.

My problem is the selection of fuel line and hoses. I was hoping to retain the original fuel line from the fuel tank to where it enters the engine bay. From that point on I plan on using a 6mm ID to 8mm ID fuel line adapter, and run 8mm ID hoses from there. In the first instance it will feed into a large FRAM HPG-1 fuel filter, which is often used in this application as it doubles as a surge tank, and from there the fuel line will lead directly to the fuel rail attached to the throttle bodies and injectors. My question is, how much of a restriction will the 6mm fuel line be to fuel pressure supplied to the throttle body rail? The volume of fuel is not an issue, as the engine itself will not use much more fuel than it did with the twin carburettors fitted, it's whether the pump will be able to supply a minimum of 50psi to the rail. The fuel pump is to be located next to the fuel tank, at the lowest point possible, while still safely positioned, where it will push the fuel up to the fuel filter and fuel rail. There will also be a small pre-filter immediately before the pump to protect it from 60 micron and above sized particles, whereas the aforementioned FRAM HPG-1 filters down to 10 microns.

Hopefully this makes sense. It'd be ideal if there was a formula, or some way to calculate the pressure reduction that occurs due to a reduction in fuel line size? For example, if a pump were capable of supplying 90psi through a 10mm ID pipe, what pressure would it be capable of supplying through a 6mm ID pipe, or an 8mm ID pipe?

Thanks,
Steph

 

[patprimmer]

http://www.grc.nasa.gov/

http://WWW/K-12/WindTunnel/Activities/Pascals_principle.html

You are asking the wrong question and also not supplying enough data to answer the question you are asking.

With no flow, the pressure will equalise no matter what the diameter of the pipe. Once flow starts, the calculation for pressure drop from flow rate is complex.

The question you should be asking is, will the flexible portions of the hose designed for use on a carburetor system that probably never exceeds 10psi still be suitable and safe at 50psi.

How are you going to accurately measure MAP to tune the thing for part throttle cruise.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[StephT]

OK, I would have thought the type of hose was an obvious choice (high pressure hose suited to the application), but I guess nothing is obvious where no prior knowledge of the poster exists. I'm completely replacing all the hoses in the engine bay with larger ID (8mm) hose. Just to clarify, it's only the 6mm ID steel fuel line between the fuel tank and the engine bay I wish to retain. All hoses will be high pressure with a braided nylon sheath, with the choice of the nylon braided exterior purely for aesthetic purposes, although I do believe it helps minimise hose-bulge, and increase the pressure rating. 350psi (working) in this case).

In essence, my question is, will the diameter of the hose effect pressure, or as I suspect, just the volume alone. I think I already know the answer, but I was looking for some reassurance, and perhaps a formula that I would have imagined included the area of a circle as a part of it. Intuitively, I'd expect 6mm ID hose to only flow 56% of what 8mm ID hose is capable of at the same pressure, based purely on cross sectional area alone. So the pressure should remain the same, and volume alone varies. Correct?

As far as individual throttle bodies go, setting up a MAP sensor with individual throttle bodies is a bit more of a challenge, because the transition between a low-load and full-load reading has the potential to occur very quickly with only the slightest movement of the throttle. As far as part throttle cruise goes, it's still early days in the research side of things, but I haven't come across anything making specific reference to this issue, but I suspect it one and the same to what I just described. Can you further enlighten me? For reference, I'm employing a MegaSquirt 2 as the engine management system for this project.

 

[thruthefence]

The Toyota 20valve "blacktop" uses individual throttle bodies and a MAP sensor. They use a small accumulator plumbed into each throttle body to give a 'smooth' manifold pressure signal.

 

[LionelHutz]

I would expect 6mm (almost 1/4") steel line would feed a 1.2l engine. 3/8" can feed a mild 5.7l engine.

The Fram filter will be completely useless as a surge tank. If this idea came from the expert then you need to find a better expert...

 

[StephT]

Well the experts are quite well known, and they've been published as well, in fact that's where I got the idea from, out of their book 'Performance Fuel Injection Systems'. I'd be surprised if you knew more than they do. They also own and run DIY Auto Tune. The way the filter is used, is you mount it so the tube that goes to the bottom of the canister is the pickup for the high pressure pump. Again, these guys do this routinely for a living, so I'm expecting they know what they're talking about.

 

[patprimmer]

Provided the fuel pump has sufficient reserve capacity for the pressure and volume required:-

1) A steel tank that always remains full and has no pressurised bladder cannot change volume beyond the stretch in the steel so it cannot act as an effective accumulator or surge tank.

2) EFI with a good pump and pressure regulator on the fuel rail needs no surge tank.

3) Pressure drop in the fuel line does not matter as the pressure in the rail is what counts and that will be controlled by the regulator.

A Walbro 255 HP will have more than enough fuel supply for your engine.



Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[LionelHutz]

Wow, you post that the pump is at the tank and this filter is an inline before the fuel rails. Now, you suddenly have another pump between the filter and the fuel rails. Get your story straight.

You have no idea what I know. I do know you didn't both reading the rules or obeying them when you posted. I know lots of people that do something for a living yet I wouldn't trust them to do anything for me. So, what's your point besides trying to insult.

I will stand by what I posted. Using your description of plumbing from the tank to the pump to the filter to the fuel rails means the filter will not work as a surge tank. Any expert that has told you it will has proven they're not an expert on that topic.

 

[patprimmer]

Lionel.

I completely agree.

I made an error of judgement when I replied. I will fix it.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[NormPeterson]

You're also going to have to do something to cope with fuel slosh in the tank as the fuel level in it drops. I rather doubt that any 1970's-era fuel tank for carbureted applications has anywhere near enough baffling, and EFI is not particularly tolerant of having its pump lose suction.

You can fix this without opening up the tank, but it'll cost you at least another pump and a separate tank.


Norm

 

[patprimmer]

Norm

You mean a real surge tank.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[thruthefence]

It's not really that difficult to enter the tank and install an "in-tank pump & filter", along with a baffled box. It's a lot less noisy, as well.

Of course, you probably wont notice the pump because of the wonderful intake noise of the four throttle bodies!

 

[thruthefence]

"Wow, you post that the pump is at the tank and this filter is an inline before the fuel rails. Now, you suddenly have another pump between the filter and the fuel rails. Get your story straight."

Sir, I think you need to read the original post.

The poster says, and I'm paraphrasing,: a "coarse filter" between the inlet of the electric pump and the fuel tank, and a "fine (fram/(surge tank)filter" at, or near the fuel rail.

Tank - coarse filter - fuel pump - fuel line to firewall - fine filter- fuel rail.

Seems perfectly reasonable to me, altho I agree the "surge tank" business is a bit of a stretch.

And he's going to need a return line to the tank from the fuel rail regulator, I would think.

 

[NormPeterson]

Using the necessary parts usually does beat wishful thinking . . .


Norm

 

[LionelHutz]

thruthefence - the filter in question is the FRAM fine filter that was claimed to also be working as a surge tank. You might have missed the original of the edited post too.

 

[thruthefence]

Not to belabor the point, but from my last post.

"Seems perfectly reasonable to me, altho I agree the "surge tank" business is a bit of a stretch."

again:

"surge tank" business is a bit of a stretch."

If we eliminate this surge tank business,(while keeping the Fram filter to do it's (filtering) job, where is the fatal flaw in this set up?

 

[patprimmer]

Lionel, you did misinterpret the layout of fuel pump and filters a bit however you are in my opinion correct in all other details.

The biggest flaw in this project is the OR has virtually zero knowledge of fluid dynamics to the extent he has no clue as to how to evaluate the advice given.

There is a fatal flaw in the logic that Pascals Principle does not apply when the system is static.

There is a fatal flaw in not understanding that pressure losses vary with flow rate when dynamic.

There is a fatal flaw in not understanding that if the pressure is regulated at the exit to the fuel rail, the injectors will see the pressure set by the regulator or very close to it so long as the fuel pump is big enough to supply an excess so as to overcome the pressure losses as they occur and so supply enough for the regulator to control the system.

There is a fatal flaw in the concept that a fuel filter will function as a surge tank.

There is a fatal flaw in giving ANY credibility to anyone who tells you that a filter is a surge tank.

There is even a flaw in the concept that this needs a surge tank. There are millions of late 80s to early 90s EFI OEM Hondas out there with no surge tank and many thousands of them have been modified to over 250hp on stock steel fuel lines and quite a few to over 400hp all without surge tanks.

As for damage to the pump from surge in the main tank, sure a surge tank supplying a remote pump will protect the remote tank from running dry, but what protects the pump that feeds the surge tank from running dry.

I know that running an electric pump typically used for EFI systems dry is reputedly harmful, however I and several family members have or have had quite a few EFI D series Hondas of about 1990 vintage that have never ever had a fuel pump fail and in every case they where at least 250,000kms old and up to 400,000+ old. I regularly push the limit on range and run the tanks very low, sometimes to he point of an engine stutter during hard corners. Still never a pump failure.

There is a flaw in thinking that journalists who write technical books and get published automatically know what they are talking about.

There is a fatal flaw in the concept that individual throttle bodies will perform better on a daily driver than tuned length runners from a plenum with a single throttle body. It can be done to get better response, but at considerable expense in other regards unless you have the knowledge of a team of good factory engineers and a lot of support facilities. It is VERY obvious that this won't be the case in this instance.

The biggest hurdle will be a stable but responsive MAP signal.

Another hurdle is the disruption to airflow by each throttle body in the heavily pulsating high peak speed airstream in the runner as opposed to the more constant lower peak speed airflow of a throttle body feeding a plenum.

The proximity of the throttles to the inlet valve do potentially give a slightly quicker response, but in reality it is a few turns of the engine, probably in the magnitude of 0.01 or less of a second difference.

Bottom line is almost negligibly better air flow response for loss of total airflow and a significant loss if engine management system response to changes in fuel requirements.

A Walbro 255HP will flow enough fuel to support a 300+hp engine through the stock Honda 6mm steel fuel lines. They are cheap and the normal high performance replacement for a Honda fuel pump on engines up to about 400hp.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[NormPeterson]

Pat - A low pressure electric pump designed for carbureted engines is likely more tolerant of the occasional loss of suction than the high pressure EFI pump.

Anything with OE EFI would already have a baffled tank.

Doing hot work on a 35 year old tank sounds a mite iffy. I wouldn't/didn't.

It is still possible through an extreme combination of low main tank fuel level and sustained hard maneuvering to empty a smallish surge tank if you're supporting a bit more power than 1.2 liters worth.


Norm

 

[thruthefence]

Pat, with respect; you are missing the point of this whole enterprise, that is; the "wank factor".

Open the hood, four individual throttles, topped with gold anodized velocity stacks,and one of these:

http://www.nerace.com/mm5/graphics/00000001/hpg1.jpg

Chromed & mounted on the firewall with stainless steel internal wrenching fasteners.

Lots of faux aeroquip hoses, & a "billet" oil filler cap.

It's a hobby car!

 

[alister1370]

Hi there,

The ratio of air to fuel is IMP
You have to calculate Lambda factor
Lambda air=amount of air transferred to cylinder/(the amount of air fit to displacement volume*normalised pressure)
and Lambda fuel is similar and it is ratio of relative fuel supply

then you you find lambda=lair/lfuel
l<1 is a rich operation and mixture

Regards

http://www.adhengineering.co.uk