IT RUNS!!! - My 500whp build - Element motor and Moore Performance Twin Scroll Kit! by dexterous

13 minute(s) of a 312 minute read

Have you guys looked at P&L's single scroll setup? This is what makes me want to down size the turbo a bit. This car has 450ftlbs @3800 RPM which is the stuff the hopes and dreams are made of. I have seen graphs of other mid sized Moore kits (verses my large sized) and the results are even better than the P&L setup.


P&L Proto type Comp turbo kit 450ftlbs@3800rpms, full boost 3300rpm 500whp!

The DOM3 is no joke. I dont have a dyno graph of my Dom3 but I do have a graph of my XTR. This graph is from my April 2010 tune at Element with a 100% stock long block and a Dom3 XTR. I adjusted the correction factor on the dyno after the tune because it was set a bit optimistically. For reference, during the tune the dyno read 470whp peak. I think a stock 2005 STI put down 230ish on this dyno if that helps.

The whole reason why I decided to move to the Element Pro Comp motor can be seen in this graph. 1500 RPM worth of rev range above 400wtq just sucks. Regardless of whatever the power numbers are on my new setup, I fully expect it to hold on very well through 8000 RPM. I would really like to see 3500 RPM worth of rev range above 400wtq.

8-25-2012

WTF Happened in here?

8-26-2012

Well, here's what I will add to this mess:

The J&S Dyno is currently the ONLY AWD dyno in western PA that is opened to the general public (the Wyotech dyno really doesn't count since its not available for public use). Like so many other people in Western PA, I count on J&S to continue to provide a valuable and much needed service to the entire motor sports community (not just Subarus). As a business owner, I have an in-depth understanding of the commitment required to operate a business, especially one that requires such a massive financial investment in equipment such as a dyno. I have talked to the owners at J&S on multiple occasions and they have always struck me as really stand up guys. I sincerely want nothing but success for J&S and their ownership. After all, if anything happens to J&S or their dyno it would be a devastating loss for our local motor sports community.

That being said, over the past two years I have heard concerns from many people (some who dont know each other) regarding their perception of the accuracy/calibration of the Mustang 500 series dyno at J&S. I would be lying if I said that I was not somewhat concerned about the situation myself. This isn't a new feeling and the content of this thread has had ZERO influence on my thoughts regarding the J&S dyno. The way I see it, a dyno is a sensitive instrument which is in constant need of maintenance and calibration. Regardless of whats happened the past, the Dyno is being moved to a new location fairly soon and Jason has assured us that the dyno will serviced and re-calibrated by Mustang at the time of the move. I look forward to getting this service completed so that the community can collectively be assured that the dyno is on target. Once this is done I will take my STI to their new location to see what the new graph looks like.

Now let’s change the subject.

So in other news…

In addition to everything else thats been going on, I have been fighting a really annoying issue with my fuel pump setup for the last month. As you guys know, I have two fuel pumps, the factory location Walbro 255 and a Aeromotive A1000 in the trunk.

Most people who have a fuel system like mine run both the in-tank and external pumps using simple relays. A relay only setup results in the pumps either running at 100% on or being 100% off with nothing in between. Running the pumps at 100% all the time has three negative side effects:

-Excessive noise (an A1000 at 100% sounds like a jack hammer in the trunk)

-Reduced pump life

-Most importantly, HOT fuel.

In typical "Tom style" I decided that I could do better. My in-tank Walbro pump is controlled by the stock fuel pump controller which is programmable via the AccessPort. I have the AP map setup so that the pump runs at 33.3% at idle and light cruse then quickly ramps up to 100% on throttle. The primary objective here is to always keep the 2.2L surge tank 100% topped off. This isn't hard to do on 93oct but I am guessing that the system will need to work harder when I switch to E85. Either way, my in-tank pump setup works pretty well with both the Hydra and the stock ECU so no worries there.

Unfortunately, my A1000 pump has been a bit more challenging. First off, the A1000's power requirements are massive so I ran a 10 gauge wire directly from the battery to the Aeromotive PWM based pump controller. To keep things safe I threw in an under the hood circuit breaker with an amp rating appropriate for the A1000 (cant remember the size). The Aeromotive controller allows the A1000 to operate at various speeds using pulse width modulation. This is much better than reducing the voltage using a rheostat (resistor) because it allows the pump to operate at full voltage thus avoiding damage.

While reading the install instructions, I quickly learned that the Aeromotive controller is really designed for old school naturally aspirated muscle cars, not forced induction imports. The controller was designed by Aeromotive to use a tach signal input (RPM) to calculate an appropriate pump duty output. Basically, as the motor speed increases so does the pump speed. This is fine if fuel consumption directly relates to engine speed (think NA motor) but fairly inappropriate for forced induction applications. Fortunately, the Aeromotive controller is also capable of switching between a programmable “low duty” and 100% duty using an override input. This allows the controller to easily be adapted for alternative applications such as forced induction imports. 

In years past I had the Aeromotive controller setup to receive control input form the Hydra ECU so that I could easily control the pump speed based on a 2D map boost based map. This made the pump very easy and reliable to control since the Hydra basically did everything for me. Unfortunately, there's no easy way to recreate this function using the stock ECU / AccessPort. This forced me to come up with an entirely new way of controlling the A1000 using non-ECU connected electronics. Since the Hydra successfully controlled the pump duty based on boost pressure I decided to replicate this using a pressure switch. I purchased a 2psi pressure switch off of ebay for around $10 shipped and hooked it up to the same hose that I had been using for the Hydra’s map sensor. I saw many other posts with links to a similar switch being used to control a second fuel pump so I figured that it would be a good fit.

Unfortunately, I immediately had issues with the new setup. At part throttle the controller would go into "default" or error mode which causes the pump output to be shut off until ignition power was cycled. I hooked up a test LED and discovered that at my cheap pressure switch would rapidly flutter for just a moment if boost was floating around 2PSI. I theorized that this rapid fluctuation was responsible for the fuel pump controller error. Assuming that my theory was correct I then blamed the “low quality” ebay boost switch and immediately removed it. I then purchased a very high quality Honeywell brand pressure switch from http://www.lsxtune.com/shop/ for around $40 shipped. The Honeywell switch is capable of withstanding 250F temps which allowed me to place it under the hood (closer to the pressure source). I ended up putting it on the intake manifold using the same mounting spot that the evap solenoid would normally be at.

While the new switch seemed to be better, I still had nearly the exact same issue with the fuel pump going into default mode at part throttle. I continued to think that the issue was caused by input flutter and began looking for a fix. It seemed obvious that I needed a few seconds of delay between input cycles to prevent the controller from erroring out. I knew that such a behavior could easily be achieved using an off-the-shelf time-delay relay with a delay upon coil release. This would allow the system to be instantly activated at any time but would not release the relay for around two-three seconds after the power was cut. The end result would be a minimum input cycle length of around three seconds which I believed would never upset the Aeromotive controller.

Knowing exactly what I wanted I searched the internet for several nights for a reasonably priced time delay relay with a delay on the power off cycle. Unfortunately, I couldn't find one with the specs that I needed at a reasonable price in North America. Apparently a 12v coil, DPDT time-delay relay with a 1-3 second off delay is a rare animal. I found lots of higher voltage stuff and many 30 second delay units but nothing with the exact combo that I needed. Granger and McMaster Carr did however have several $140 solid state relays that were programmable to my exact specs but I wasn’t going to spend that much on something so simple. I was also afraid of importing a super cheap ebay relay from China. After four nights of looking I decided that I would just build my own setup. One quick trip to Radio Shack later and I had a prototype built on my bench:

This circuit design uses a capacitor, a diode and a transistor to energize a relay coil for exactly 3 seconds after the power is cut to the transistor’s base input. After bread-boarding it and fine tuning everything I built it into a printed circuit board complete with fused inputs.

I dipped the board into liquid electrical tape to seal everything up and let it dry. A few hours later I had the board installed in the car and working perfectly. Every time I hit the gas the pump runs at full blast for three seconds. A quick drive on a back country road told me that the pump controller is now behaving.

No, I haven't. What I did was set up the pump duty with enough flow to maintain base fuel pressure until around 5-6psi. If the pump duty didn't increase by 5-6psi the fuel pressure started to dip. From there I monitored the average voltage output of the controller (its PWM so all you get is an average over time) and made sure that it was above minimum voltage stated in the A1000 manual (9v if I recall correctly).

The fuel setup has been this way (less the time delay) for around two years. While I cant really monitor the wear on the pumps, I can tell you that the in-tank fuel temp sensor tells me that the fuel is significantly cooler under typical conditions (20-30F). I can also tell you that the in-cabin noise is tremendously quieter with the pumps both in low duty.

Just curious, what would you have set it to?

9-2-2012

This was the next video to come up. LOL:

9-4-2012

Back when I started this thread I had no idea that two years and nearly 2200 posts later it would still be going strong. What a ride! Its also nice to be talking about header designs and materials selection verses failed motors and stuff. Its a nice improvement form this time last year.

9-20-2012

Phil,

Have you taken apart one of the new motors yet? Any thoughts you are willing to share with us? I am most interested in your thoughts on its ability to be built up into a decent forced induction platform. It seems that all of the FRS/BRZ cars that are involved in motorsports are running swapped EJs so that makes me wonder.

From a business perspective, I worry that Subaru lied to us regarding their intention to bring out a turbo version. If people buy the first gen cars then trade them on a second gen turbo version it allows Subaru to "double dip". It may not be true but its an easy assumption that a factory FI platform would be a better starting point for you and your race car verses its NA counterpart. Dumping a ton of R&D cash into the NA motor platform may prove to be fruitless sooner than later. My point here being that there's lots of room for a small business to get screwed.

One more question, now that you have a real suspension and some seat time on your FRS, how does it handle on the track in comparison to the STI?