Full Blown Dual Pump System - Things You Should Know

[Ted B]

Originally Posted by mr.alex

I've tested various drill sizes for the eductor, indeed 3mm was not enough, i had to go 3,5mm but just touching the unit with a drill caused the suction from the right tank-side to drop dramatically.

The suction should drop, but is likely greater than what it is with the factory configuration. I say this because at 3,5mm, I am just barely able to control the FP in the rail at idle. I would have no such situation with the factory eductor size and fuel pump.

This tells me there is higher pressure at the eductor than with the factory setup, and therefore higher velocity through it than what we have with a factory setup. The Bernoulli effect should not be diminished as compared to the factory setup.

[mr.alex]

Instead of drilling, why not heat up an edgy tool or something and try to stretch the hole? this would spare the lenght of the nozzle inside the eductor and give better suction.

[l2r99gst]

Ted B,

Before running the dual in-tank pumps (parallel), had you considered running two pumps in series? Two pumps in series gets rid of the extremely high fuel pressure at idle problem, yet maintains fuel flow capabilities out at high pressures much better than the parallel in-tank pumps. Actually, it outflows dual intank pumps at fuel pressures of 90psi and higher (I know that's more boost than most will likely run). AMS has a great graph of their testing that shows this.


Eric

[diambo4life]

Originally Posted by l2r99gst

Ted B,

Before running the dual in-tank pumps (parallel), had you considered running two pumps in series? Two pumps in series gets rid of the extremely high fuel pressure at idle problem, yet maintains fuel flow capabilities out at high pressures much better than the parallel in-tank pumps. Actually, it outflows dual intank pumps at fuel pressures of 90psi and higher (I know that's more boost than most will likely run). AMS has a great graph of their testing that shows this.


Eric

That is incorrect. As a matter of fact, AMS testing showed the EXACT opposite of what you are saying there Eric. Parallel plumbed fuel pumps will outflow pumps setup in series any day.
Chart

[Ted B]

Originally Posted by mr.alex

Instead of drilling, why not heat up an edgy tool or something and try to stretch the hole? this would spare the lenght of the nozzle inside the eductor and give better suction.

Like I said, I don't think any of this necessary - I'm not experiencing any issues. I expect that with both pumps running, the suction is greater with the drilled eductor than it is with the factory setup. Why? Because with both pumps running, the drilled eductor is just barely able to keep up.

As for running the pumps in series, there would be nothing wrong with doing it that way. It's just that the pumps have a little more overall capacity when running in parallel, and that's why most choose to do it that way.

[l2r99gst]

Originally Posted by diambo4life

That is incorrect. As a matter of fact, AMS testing showed the EXACT opposite of what you are saying there Eric. Parallel plumbed fuel pumps will outflow pumps setup in series any day.

Nothing I said was necessarily incorrect. But, I was referring to a slightly different chart from AMS than yours. Your chart does show the duals holding flow longer, but my general statement holds true. The pumps in series maintain a flatter flow, without the huge overrun at idle. And, at a certain pressure, the pumps in series outflow the pumps in parallel (although it is at a high pressure where most won't reach...boost pressures of about 45psi or greater, depending on your base fuel pressure).



You can see in the graph above, the light blue are the 255s in series and the pink is the 255s in parallel (intank dual setups). The duals in parallel give you too much flow where you don't need it, like idle, and the series configuration is much lower, like you would want it. And, at higher fuel pressures, the series doesn't dramatically drop off like the parallel config. Yes, I do agree that the parallel outflows the series at lower boost pressures, but that also comes with the issue of way too much flow at even lower boost and non boost fuel pressures.

All I am saying is that the series configuration makes more sense to me in terms of the amount of fuel you need in different situations. Ideally, we would want the least amount of fuel at low fuel pressures, like idle, and the most at highest fuel pressures, like high boost. The series configuration matches this ideal much better than the parallel. At the very least, you wouldn't have to worry about hobbs switches, triggering the pump, adjustable fuel pressure regulators, etc.

However, if you will never run boost pressures high enough to get to the area where the series config outperforms in terms of flow and the series config doesn't provide enough fuel flow at your intended boost, then of course, you have to use the parallel. But, even doing some quick math at 30 psi (73 psi fuel pressure), the series config will flow enough for roughly 90lb/min airflow at 11.5:1 AFR. That's more than enough for any turbo at that boost pressure on our engines. So, all I am saying is that the series seems to fit the ideal much better and should flow plenty of fuel in boost without all of the issues at idle and non boost.


Eric

[Ted B]

The series configuration sees steadier pressure at low usage rates because one pump is providing the volume, while the second pump reduces the pressure step.

The potential issue I see with this lies in the use of a large turbo and E85 (which is exactly my intention). The series configuration appears not to have any reserve capacity (given my calculations) at the expected boost pressures to give any margin of comfort.

It's a matter of individual application. It would work for some, probably not for others.

[diambo4life]

Originally Posted by l2r99gst

Nothing I said was necessarily incorrect. But, I was referring to a slightly different chart from AMS than yours. Your chart does show the duals holding flow longer, but my general statement holds true. The pumps in series maintain a flatter flow, without the huge overrun at idle. And, at a certain pressure, the pumps in series outflow the pumps in parallel (although it is at a high pressure where most won't reach...boost pressures of about 45psi or greater, depending on your base fuel pressure).

http://forums.evolutionm.net/attachm...hmentid=112816

You can see in the graph above, the light blue are the 255s in series and the pink is the 255s in parallel (intank dual setups). The duals in parallel give you too much flow where you don't need it, like idle, and the series configuration is much lower, like you would want it. And, at higher fuel pressures, the series doesn't dramatically drop off like the parallel config. Yes, I do agree that the parallel outflows the series at lower boost pressures, but that also comes with the issue of way too much flow at even lower boost and non boost fuel pressures.

All I am saying is that the series configuration makes more sense to me in terms of the amount of fuel you need in different situations. Ideally, we would want the least amount of fuel at low fuel pressures, like idle, and the most at highest fuel pressures, like high boost. The series configuration matches this ideal much better than the parallel. At the very least, you wouldn't have to worry about hobbs switches, triggering the pump, adjustable fuel pressure regulators, etc.

However, if you will never run boost pressures high enough to get to the area where the series config outperforms in terms of flow and the series config doesn't provide enough fuel flow at your intended boost, then of course, you have to use the parallel. But, even doing some quick math at 30 psi (73 psi fuel pressure), the series config will flow enough for roughly 90lb/min airflow at 11.5:1 AFR. That's more than enough for any turbo at that boost pressure on our engines. So, all I am saying is that the series seems to fit the ideal much better and should flow plenty of fuel in boost without all of the issues at idle and non boost.


Eric

I see what you are saying Eric but when you are playing with E85, the dynamic will change drastically. Series setup just doesn't have the flow required to sustain big turbo demands.

-Mike.

[fostytou]

This may sound slightly stupid, but is the overrun only at idle or would it exist at light throttle before/after the Ted B fix? If it is only at idle could we use an idlesw output (I believe this is still on the ECU, even though it was removed from the throttle body itself) to control the second pump's on/off?

[Matthew333]

why not just run 2 kenny bell boost a pumps to control each fuel pump???? You will also get increased voltage to the pumps!

[fostytou]

Originally Posted by Matthew333

why not just run 2 kenny bell boost a pumps to control each fuel pump???? You will also get increased voltage to the pumps!

I'm not totally familiar with that product, but it looks similar to the hobbs solution... just adding a bit more voltage to get more out of the pumps. The problem here is that it is very difficult to tune for something like this. It is very possible that in one gear/rpm/etc combination you are at 80 load and in another you are at 120 load.

The question is how do you tune for the extra fuel when you don't know when it will start flowing (based on your tunable parameters)?

[Johnboy1065]

I run the full blown dual walbro setup and activate the 2nd pump with a pressure switch and havent had any issues. I do have an adjustable Fuel pressure regulator though

[rodent]

Sorry to revive an old thread but I ran into issues last Friday during a dyno tune where the hobbs switch wasn't kicking in all the time and it was also causing a rich condition at 16psi that's difficult to tune out. I'm also running twin fuel lines from the tank into -8AN wye which goes into the Aeromotive fuel rail and uses the stock return line. I think the twin/larger lines are causing the Aeromotive regulator to spike in too much pressure at 16psi and a larger return line should be installed. The car will be running well over 700whp on E85.

I think I'm going to get rid of the hobbs switch and run both pumps all the time. I'll definitely need to do the larger return line. I was thinking of keeping the stock return line in place and run another 1/4 line down to the tank. So basically it will be -6AN at the regulator and wye into two -4AN (1/4") lines. Another hole will have to drilled into the top of the pump housing for the 2nd line. This way the stock syphon tube won't have to be modified with a drill bit and should still work and still give plenty of return.

Thoughts?

[WHTEVO]

I think the only issue you might have is that the fuel will return down whichever line has the least resistance more. Which may or may not have a effect on the tee on the pump housing. i tried making my own tee so i could use a single larger return line, and use one large external pump, and lets just say it didn't go so well. It has to be damn near perfect for it to work. I don't know what calculations Mitsu did to come up with that orifice size to create that bernoulli effect. Good luck man.

Aaron

[rodent]

Update- I really didn't want to mess with the stock siphon tube and wanted something larger than the stock return line size (1/4") so I ran another 1/4" line for the return and everything is working perfectly. So its basically a -6AN into a wye, then two 1/4" lines coming back to the tank. I never had to touch my regulator with the two pumps running at idle. I also drove around for a week with just 1 pump to make sure the siphon tube was working and no problem there. For the 2nd return line, I found a good spot on top of the fuel pump assembly and drilled and tapped it for the 90 degree fitting. I also used some JB Weld around the fitting for added strength. Anyway, just wanted to post my results with the dual return line. I'll be posting pictures and a quick write up on my dual feed/return line setup later this week.