- Fast Intentions Twin Turbo Kit
- Fast Intentions Dual 3″ Exhaust
- Twin HTA 2868 turbochargers — Alpha Covers w/ Tial housings
- 10psi on 92 octane
Starting the new year out with some bang — been driving around on our boosted 500hp FR-S for a couple months now and dealing with one of the nuances of making more and more power — the need for fuel system upgrades and tweaks. Now it’s time to go over what I’ve done to the car, what options I went with, etc.
I was very close to building our own turbo kit for the FR-S — however PRL Motorsports had an option available and as I already have a very good relationship with them I decided to give their kit a try after we had some back and forth. I did make a few tweaks based on what I like to run on turbo vehicles — making it a little custom “VitTuned” off shoot, but the heart of the kit is their work and fabrication.
Before you say you can’t find it on their site — you’re right, you can’t. You can contact them directly, or myself — and I can help you build the package that’s right for your FT86.
Now getting into it– I chose to run a Comp Turbo CT4X-5862 .82 a/r turbocharger on my car. This is quite a bit larger than what most people run (it’s a large 4″ in, 2.5″ out cover — most kit options will have something like a 3″ in and 2″ out cover, which is quite a bit “smaller”). But I knew I was going to shoot for more power — and I just happened to have the turbo sitting on a shelf as well.
Basic fuel system upgrades included a DW300c in tank fuel pump and Deatschwerks 900cc port fuel injectors.
OK, enough with the build info — what’s this thing make? I have to say, I was quite impressed with the power output of this motor with this turbo. On 4psi (wastegate pressure) we made roughly 100hp more than stock on 92 octane fuel.
Cranking it up to 12psi running a conservative tune for the 92 octane fuel, we were able to hit just shy of 320whp. Quite a solid power figure considering this was more than 2 times stock power and still running on the stock clutch.
Speaking of the stock clutch — when I tried to crank it up a couple weeks later on E40 (40% ethanol content) it let go. So I put in an ACT 6 puck sprung setup and got back on the dyno with E55 (55% ethanol content). Boy did it make power… 16.5psi made an awesome 450whp, 3 times stock power!
About a week later I came back in with E75 in the tank to push it a bit more. Did it pick some more up? You bet, she put down 485 whp without even breaking a sweat at 18psi (3.2x stock power relatively).
But this is where some of our troubles begin… And no, it’s not the motor. I’m happy to report it is currently running smooth and strong.
You got it — at this power level the stock returnless fuel system is complete maxed out. As you can see from the following datalog plot — port injector run at 17ms pulse width — this is beyond even 100% “duty cycle”. Since this is a dead head system, I did not have a fuel pressure sensor mounted yet to monitor the fuel pressure — but one can assume it was about 20-25psi “differential pressure” (actual pressure over the injectors), given the PW the ECU was commanding to hit fuel targets.
So it’s time to build a fuel return.
There are a couple ways and already a couple kits out for the platform — some that I don’t particularly like (because, bluntly — the fitment is complete shit). I chose to build a return the same way I have done it on other platforms. Over the course of troubleshooting the fuel system, I also chose to run a new -6 AN feed line in addition to the -6 AN return line. Here is a rough parts break down of the fittings necessary.
The fittings and line were sourced from SummitRacing.com — I used mostly Russell items, with some odd balls dropped in depending on what was in stock (or what I had available already).
I used the stock rails and recommend you do the same (less line and less fittings, better fitment). I had a set of aftermarket rails, they fit terrible, no room to mount the DI computer (which is grounded to the head via the mounting bracket — your car won’t run otherwise), couldn’t clip in one of the injector clips… it was so bad I threw them in the trash.
To give you a verbal description of how the fuel return works — start at the tank. The feed comes out into a 3/8″ EFI fitting and goes into the matching hose end. The feed line then snakes into the engine bay and splits with the T fitting.
The T fitting then splits and one end runs into a 5/16″ EFI hose end into the DI pump. The last out on the T runs into the side of the regulator. Then you start at the next regulator side and feed that end into the port injector rail with the remaining 5/16″ EFI hose end — or in my case (as the pictures show) you run this into your flex fuel sensor, and then into the rail. Same exact concept though. Side note — you can with a fuel return relocate the flex sensor into the return line. I chose not to do this since I already ran the flex sensor with the dead head fuel system and it was “easier” to leave it in the feed line.
The final line — is the return line off the bottom regulator running back into your tank via the bulkhead fitting.
In the cage itself — you do have to replace the stock regulator with a blank to feed all the fuel into the feed line and towards our external regulator — as the following 3 pictures depict.
Now some people might debate this choice of blocking off what is known as the “Venturi” feed in the cage itself — and might be concerned about the in tank siphon from one side of the tank to the other on the FT86 platforms. I have been doing fuel returns this way for years when converting dead head fuel systems — it works, and it works great.
Here’s some facts about the Venturi feed — it is NOT designed to work the in tank siphon. The in tank siphon works like any other siphon — gravity and pressure! Simply put, the pressure in the tank will normalize the fuel level between the two sides. I’m not just saying this — I have been driving the car like this, the siphon works perfectly FINE without the Venturi. What the Venturi is designed to do is create a type of vortex/swirl feed at the inlet to the factory cage — this is designed to create a low pressure area that pulls fuel into the cage and helps keep it full during low fuel level conditions in the tank (and works so-so as anyone that’s ever done any aggressive driving on high powered vehicles can tell you — my dead head fuel systems running 100% OEM cages will starve the pump still). This is an effect you ultimately break the moment you use your return to feed the Venturi — under power the return line flows very little fuel and pressure, whereas the OEM Venturi is fed directly by pressure from the pump! With a Walbro 485 in the tank — I can hit it in boost with 1/4 of a tank and still maintain fuel pressure. Ultimately for someone running aggressively at a track event — a surge tank setup with their fuel return is the way to go (no in tank setup will ever be 100% when it comes to avoiding fuel pump starvation).
So what did all this work net us? Tons of room on the fuel system — at the normal 15psi boost pressure I run day to day, I have a ton of breathing room to push it harder if I want to in the future. Without the return, I was already at 13ms pulsewidth on the port injector, now I’m not even at 10ms.
Maybe more power in the future? We’ll see…
It’s honestly pretty funny how far internet trolls and wannabes that want to take a bite out of you will go. News flash — I’ve been doing this for years and seen all the lies and dyno fudging one ever needs to see that will last a lifetime. Yet these “people” still seem to think tuning is all about power and racing their dyno sheets (fake at that) online.
But sure, I’ll show you how to make 20hp+ on a DynoJet without ever changing a thing and getting the ability to claim “there was so much wrong”. This is really juvenile internet troll behavior — but hey it’s the weekend, let’s spend a little time burning them down.
Make it as hard to read any information as possible. For the sake of this blog post though — I want it legible so I won’t go that far. Not a single thing was tuned in the tune for these pulls. Not one. The tune is exactly the same.
How is this even possible??
Let me be clear — anyone can fake any dyno to push whatever agenda they want. They’ve been doing it for years, this isn’t anything new.
So how did I get the numbers to go up without changing the tune (I can certainly claim I did and that in the base “everything was wrong”)?
Note how the curves are very similar — it really is the same tune. Very rarely will the overall curve of a “retuned” vehicle follow the same path if it’s a different tune. So the first thing you should note — the graphs are all individual, they are not overlayed. There is no correction factor listed anywhere on those graphs. So let’s shed some light.
But that’s not all that’s changed — we also went from using 3rd gear to 4th gear, which “added” 10hp on it’s own, as seen here on the right. Note the “gear ratio” information — higher number is an earlier gear (3rd in this case vs 4th gear). This is not an unusual phenomena — dynos can and will read a bit different based on the gear you use.
But hey, for internet “facts” sharing — we just gained 18hp by fixing a tune that was “so wrong” — LOL?
It gets better. What if we mess with the dyno weather station? Wow, we have an amazing 278whp now! I can leave the correction factor, gear ratio and all other analytical data off, just share these numbers and now I’m the best “tuner” in the world — I just made 26whp over someone else’s tune, without changing anything (but I can certainly claim I did — who’s going to know, right?)!
But if we actually pull up the run conditions, note correction factor now — 1.04. With a very weird humidity… yup, the graph had someone screwing with the weather station. Here’s a news flash as well — it is relative child’s play to screw a DynoJet run, even AFTER it’s been saved. I can do this (or anyone else with basic computer skills) with any DynoJet run file someone sends me — make their graphs read “higher”.
Finally a 1:1 overlay to show you can even fudge a direct comparison if you want to.
And hey, as far as the Internet is concerned — if I post these graphs and just claim the base tune was wrong, who’s going to know, right?
In conclusion? If you want to make your tooner friend look good, there’s legitimate ways, and then there’s BS ways. Spreading lies and false hope never helped anyone — you just lose any credibility you could of had as you and whatever friend/tooner/shop you are trying to “back up” is clearly a liar with no integrity. But hey, that’s just the game, right? Sure, but I’m not playing.
What appears to be the one absolute and unsurprising fact in this industry — someone is always out to try to take a bite out of you. Or try to get their friends to come after you so they are not seen as participating. What is surprising (or maybe amusing?) is how they like to come at you at times. It would be a much more entertaining read if they had some understanding about the things they were frantically smashing into their keyboards. Unfortunately — it’s quite clear they have little to no experience in anything but slinging mud. I cannot help but just shake my head and go back to helping my customers paying for my time.
Fortunately, I had the time and opportunity to bring up a few examples — had a 2014 Civic Si on the dyno to tune and it turned out to be a chance to touch base on the results of tuning a stock 9th gen Civic as I still get asked about it a lot.
This one really gets me. When someone doesn’t understand why something in a tune was done the way it was — it’s clearly shit. Their first reaction is to grab their pitchforks and torches when it should be to pause and analyse why something was done the way it was. I see this attitude frequently and can only wonder if someone like this would ever take any advice or constructive criticism that could help? If something is truly “shit”, lay out why in intelligent fashion — and no, not just a dyno sheet or 3hp. Or switching gears on the dyno and claiming you made power with your “tune” when all you’ve done is introduce another variable that just made any use of the baseline invalid. We’ve made 6hp on some 9th gens just by letting it sit and cool off . Not even the best troll attempt I’ve seen — but cute, I guess?
A truly good example is a tuned VTC map — particularly the low cam (non-VTEC). Low cam was tuned to determine the best cam advance, and then high cam was tuned to determine the best cam advance. Finally the transition was optimized to provide the best curve possible when going into VTEC during a full throttle pull. Optimal power at 5000 rpm out of VTEC was with 5* VTC — however it was 35* VTC in VTEC at 5000 rpm for optimal power. So what happens when VTEC engages and you have such a large VTC transition? The VTC system is still at 5* and the motor makes less than optimal power until the cam has a chance to move to 35* VTC. As a result the transition was optimized by a very short and quick snap from 8* at 4900 rpm to 32* at 5000 rpm while VTEC is still off — a very smooth transition when at WOT and no loss of power is seen or felt. If you don’t see this difference on the dyno — time to replace your piece of shit roller (hah, I just went there).
What does this look like on the dyno? The graph at the left illustrates this. Solid lines are horsepower and torque with the VTEC “pre-phasing” trick done, dashed lines are without. Pretty obvious torque dip at VTEC, right?
Wait, if this is so good, why are you telling everyone about it? Because this is not a secret. When the K series first came out Hondata pioneered this trick and published it publicly for EVERYONE to use. The irony here is — very few use it, and even less understand why. It’s a better idea to just go on the Internet and make it obvious you’ve done like two Honda K series vehicles and now you’re an expert? More like wet behind the ears.
Hold on, it gets better… you know even *Honda* uses this trick now? Just look closely at the stock Honda 9th gen tune. For the two seconds it takes to see it — it’s not tough. They didn’t use this on the RSX’s or the 8th gens — and their use of it is very subtle on the 9th gen.
I’ll be Mr. Nice Guy — to the right is the stock 9th gen VTC map. Note the values at 5000 rpm at full throttle — they go back up slightly in anticipation of the VTEC crossover (which is 25 at 5000 rpm in VTEC on the stock Honda tune) — the only difference is they didn’t use a 4900 RPM break point as they really don’t care if the car loses 4hp across 500 rpm worth of power from running 7-9 degrees too much VTC (I guess they haven’t met the experts on the internet yet?). Yes, Honda intentionally mapped the motor this way — there is absolutley no reason to bring the VTC back up in an area it naturally wants to taper down to keep making power on the low cam.
Shocking, I know.
Let me repeat.
Tuning is about so much more than making power. I have customers with 150k+ miles on their turbo vehicles. I also have a dyno at the shop readily at my disposal. I know what makes power — in fact I use the dyno not only for tuning customer’s vehicles and builds at the shop, but as an R&D tool and apply what I discover into our eTunes to help deliver reliable cars for customers around the world.
Part of tuning is making decisions and judgement calls that will ultimately determine a setup is reliable long term — or not. In varying conditions year round — that the tune has to take into consideration and adjust for. This is why people go to reputable and experienced tuners, instead of a random guy offering you tunes for $50? Right? Maybe I’m wrong, what do I know?
I’ve posted several comparisons of stock 9th gens. I always answer — yes there are benefits, it will make a bit more power, and the power will be more consistent.
Here is the simple and visual example of this. The engine coolant temps (ECT) where 185-188 degrees F and intake air temps were 59-60 degress F on the stock tune baselines. Yes, I made sure these were consistent to avoid any extra variables when doing the comparison — if you’re paying someone 60 bucks for 3 baseline pulls I can tell you with absolute certainty they are not paying attention and simply don’t care. Yes it does matter — I’ve had customers datalog their baselines and in some situations there’s not even a dyno fan on the car — I’ve seen 20-30 degree difference in ECT (180 and 210…) and even a 30-40 degree swing in air temps between pulls. If someone thinks it “doesn’t matter” — they should not claim to be any kind of EFI tuning specialist.
Anyway… stock pulls. Three pulls with consistent conditions. Left graph is torque, right graph is horsepower. Quite the difference on the top end right? 10-12 HP swing in some spots. Yes, the stock Honda tune is very inconsistent in it’s power delivery — and no it is not the “knock control” as some would make you believe. The simple answer is just this: emissions.
Now four (yes four) pulls with consistent conditions — 185-188 degree ECT and 68-70 degree IAT. A negligible .5-1.2hp swing. Virtually nothing. This is on a dyno that is accurate to .1hp (no, a roller dyno does not nearly have this kind of accuracy — having a wheel/tire on the car can cause a 2-4hp swing).
And as every tuner in the world loves to do and not tell everyone… we overlay the highest “tuned” graph with the lowest “baseline” graph. This behavior should come as no surprise — as a ton of “tuners” make a living off nothing more than the way their dyno reads (big numbers sell tunes and that means you’re the best right? Hm… I’ve got some graphs I can post… no, I probably shouldn’t go there).
And now I’ve run out of thinks to rant about…
For a little change of pace I want to discuss the AEM Infinity retrofit we did on a customer’s Toyota Celica, what really prompted this change and the features of the AEM Infinity we implemented on this platform — as well as a few others that are available.
We had the customer’s car in the shop for a fair amount of time — installing the Monkey Wrench Racing turbo kit he had purchased with the engine management package. The turbo kit did require quite a bit of customization to fit, but we took care of all that and the car was ready to tune.
I spent a good 3 weeks on the stand alone ECU that was included with this package as I was never quite content with the drivability of the vehicle with the included engine management. I don’t want to name any names, but it is the “accepted go-to” EMS for this platform — with fairly hefty price tag (customer paid something like $1500 for it). Unfortunately it is not a very good EMS and is very dated — it’s comparable to running something like an AEM V1.
Did the car make power? Sure — it’s not hard to make power. However this EMS was plagued with random misfires and quite a bit of time was spent chasing the misfires (hardware, coil dwell, sync, etc) and eventually they became very rare — but not quite gone. When taking the car off the dyno and on the street no misfires were apparent — I’ll get into that more later.
After the car was off the dyno, I put in a week of street tuning for drivability — cold start, cold start driving, warm start, warm start driving, the works. The transient response that was achieved with this EMS was not remotely what I would call “great” — warm or cold. Many would consider it to be “acceptable”, but in my book if the car doesn’t put a smile on my face from the moment I fire it cold and tear off down the street without giving me weird moments of hesitation — it’s simple crap. Why? Because I work with dozens of different EMS and dozens of different platforms — I have a very intimate understanding of what “great” drives like.
And before people start questioning — yes, all the cold start, post start, throttle pump, throttle enrichment, delta tables were tuned, every single parameter that was exposed was tuned to try and improve overall drivability.
Finally we had no choice but to let the customer try the car out and see what he thinks — and we went in not at all happy with the results this EMS let us achieve.
Sure enough, he brought us the car back a week later as agreed to let us fit a much more modern and advanced EMS on the car — an AEM Infinity.
Before I go into the details of this EMS, let’s go into what we expect to be able to do with our EMS and how it compares to what the previous EMS (these are some basic features as this car was run at 10psi on wastegate — no boost control):
As can be seen, the basic features look quite similar (the AEM Infinity is capable of much more — including traction control and DBW), so why would the Infinity be so much “better”? Let’s look into a few bullet points.
This is the most annoying part of the previous EMS — once we retrofitted the AEM Infinity on the vehicle, we did not experience a single misfire. The AEM was sync’d and tuned without experiencing a single misfire during the tuning session. This leads me to believe the previous EMS was having trigger sync issues and/or problems running the coils/injectors. The coils & injectors themselves were not a problem (and neither were the cam/crank sensor) as they all functioned perfectly on the AEM Infinity.
A picture is worth a thousand words. Yes it works with the AEM Infinity — just requires proper install & configuration. Not a “big deal” to get it working. And the feedback/control is completely adjustable.
This was extremely lacking on the previous EMS — in fact I’ll go so far as to say it was completely junk. When you’re paying $1200 or more for an ECU you would expect a modern EMS to have wideband O2 control already without having to pay extra and depending on a narrowband O2 sensor — which amounts to slow feedback and inaccurate feedback (fueling swinging back and forth, etc).
As you can see, with the AEM I configured feedback well into boost (in fact the system is so fast and accurate you can run it full time on high boost applications as well).
The previous EMS? It simply didn’t have this ability — and I wouldn’t trust it using an analog input into the ECU. It’s not accurate and the EMS has no control over the O2 sensor directly (since it requires a separate controller) — so if the sensor has a fault the EMS would know. The AEM *does* control the O2 sensor and is very quick to detect a fault (from experience), and you can chose to put the ECU into protect mode if the sensor faults. Fantastic.
Two words: It works. Here’s a short clip from the street tuning session with the AEM while I was out touching up the car (IE: drivability adjustments, finding any spots it may knock with “real” load on the car on the street).
This is just a small glimpse at the control — we also have the option to enrich lambda during knock events and individual cylinder knock thresholds.
This was a real source of frustration with the previous EMS — 8 channels of datalogging (and not real time at that — you have to record and then view) is horrible. To use the term “unacceptable” is an understatement — the more data you have to view the better you can tune the EMS and you get a much better picture of what the EMS is doing if you have to do any diagnostics. Not enough data and you can be left chasing your tail — at the very least it takes up much more of your time as you’re stuck with scenarios of “oh I wish I had logged that parameter… but all 8 channels are already used..”.
The AEM lets you record everything. And the “on board” datalogging is expandable by the size of the USB drive you plug into it if you need “on board” datalogging. The fact an EMS company advertises “256K” of onboard datalogging is just laughable — that’s NOTHING and utterly useless (not enough data, or not enough data at meaningful refresh rates).
This was the biggest point of annoyance for anyone driving the car previous — the throttle response, transient response and drivability just wasn’t there with the previous EMS. Sure if you had a race car, you put your foot down and it worked fine. However with the new AEM Infinity installed and tuned (which I might add only took two days to fully tune on this vehicle — yes you read that right) the drivability of the vehicle soared to perfection.
No I’m not exaggerating — I could fire the car up cold or hot and stab the throttle immediately without any delay or hesitation. Off I went down the street to do further fine tuning.
The best part? I was able to deliver the car to my customer driving perfectly in every scenario. Awesome.
I guess what I can say about the faults of the old EMS lie in a couple of points.
Does that EMS look good on paper? Mostly, yes. But looks and practice can be completely different things.
Why is this old EMS so popular? I think this mostly comes down to what select groups have learned and push adamantly on this platform — simply put they are stuck in their ways and either don’t know anything better or don’t want to try anything that could be better. It “works for them”, as I hear a lot.
However if you are looking for something better on this platform that can improve your vehicle — we have better options available that don’t break the bank.
Now I am sure this is going to ruffle some feathers and some might even dust off their pitchforks — so be it. It has to be said — living and breathing cars, modifications and tunes day in and day out you see some outrageous things come your way. I am going to break it down into two simple categories — food for thought and enthusiast beware.
This is the guy that has a shop or works for a shop — they have a dyno and you would think they would know how to use it. In fact, some of them do know how to use it very well, and the break down begins with the fact they understand very little, if nothing at all, about EFI tuning and/or the engine management software they are using. Hell, they might even be working at a shop that has a stellar reputation!
They are able to post up amazing numbers on said dyno, but the vehicle will just run terrible either the moment it leaves, or a couple days later. When the car comes back to them — they cannot figure out the source of the problem and will at times chase mechanical gremlins that do not exist.
One prime example of this is a customer with a 2012-2015 Civic Si — said customer had some work done at his location and the vehicle posted up absurd power figures on pump gas (93 octane) — nearly 500hp. Anyone who has any experience with that platform will raise an eyebrow — maybe it’s possible? Highly unlikely with how knock prone those motors are. But hey, the customer was initially quite happy with the numbers — and then the bad news. After a couple of days the car would be completely gutless, and any attempts to have that behavior remedied kept failing.
He finally got a hold of me and we went over what was going on — turns out he was an existing customer of mine that had a tune for his car when it just had bolt ons. After reviewing his turbo datalogs, it turns out it was running 17-18psi of boost and -2 to -6 degrees of timing (yes, NEGATIVE). It doesn’t take a rocket scientist to figure out this does NOT make power. In fact — it makes about a whopping 200 horsepower with all that boost, not anywhere NEAR the claimed dyno figures.
And yet it continues to get worse — part throttle closed loop feedback was disabled, the primary O2 sensor completely disabled, VTEC point untuned, VTC mapping completely untouched. No excuse for any of these items to be the way they were — so where was the break down? Upon inspection of the actual calibration — the tune file was basically nothing more than the calibration for his vehicle with simple bolt on parts. Quite literally the injector scalar (how big the fuel injector is) was adjusted and the fuel map was roughly skewed upwards in boost — nothing else in the calibration was setup anywhere close to how a turbo calibration should be done to get a reliably running vehicle. You read that right — he was running MY N/A tune on his car with a turbo. You can imagine how well that works.
So how did it make power on the dyno? Quite simply — with every flash of the ECU its “learned” parameters were reset and under boost the motor was seeing effectively naturally aspirated ignition timing. Wait, won’t this cause detonation? Absolutely — this platform does not have active knock feedback logic, only having a very slowly learned correction (which works poorly on modified vehicles to begin with) and the motor will definitely make power.
So the sum of his “expert” tune was nothing more than a couple of hits on the dyno with a calibration that was ripped (stolen) off the FlashPro when the vehicle came in for work to be done.
But a couple days later — the ECU goes limp mode to protect the engine. In this case — the owner of the vehicle was very lucky. We dialed the boost back to 11-12psi and tuned the car properly and it hasn’t had a problem ever since.
You know who I am talking about here — they are all over social media and online forums. They will make you promises and whisper sweet nothings via private messaging systems to get your attention and make your wallet a little bit lighter. I have seen examples of them even claiming:
What skills, experience and accomplishments do they actually have? Well it’s quite simple:
Yup — that’s right: all it takes to claim to be a tuner extraordinaire on the internet is a laptop and the ability to transform drivel that would normally flow through their lips into text via their fingers rapidly clacking away at their laptop keyboard.
Even better — in many cases they’ve even purchased a tune for their own vehicle from an established tuner. This has happened so many times I have lost count — I am more than happy to share examples with the reader privately.
But you bet they will be all over the internet trying to snag their next victim — sometimes advertising their services, sometimes trying to stay under the radar and snag their victims via the amazing stories they like to tell via private message.
And at the end of the day — they have little to no experience (you will see stuff like “I tuned my car”, “I tuned all my friends”) and no accomplishments of any kind. Of course they will claim that they “have to start somewhere”. Anyone sign their vehicle up as R&D when they paid for a tuning service?
Amazingly enough — they will always have someone “vouch” for them and their “skills”.
This is a vicious cycle — I’ve seen it so much that I just shrug and let our business and our work speak for itself.
But be warned — the tooners of the world will throw timeslips, dyno sheets and vouches at you all day long and in their minds it gives them completely credibility. Sure — going fast and big numbers are fun, but it paints a very poor picture of any experience or ability to tune a car properly to do anything beyond that. When you have a vehicle you need to drive day in and day out, there is a lot more that goes into setting up a tune than dyno numbers and time slips.
The goal is to hopefully share some insight with the reader and maybe prevent another case of Tooner Attacks. Ultimately it is up to every enthusiast to do their own research — and I encourage you all to do so.
What I really wanted to title this is “ignorance is bliss — no WAY could ethanol possibly even a single HP over pump gas”. I was inspired to do this test by a very bold claim on a certain online forum — I’ll keep the names out of it, but it definitely gave me a chuckle.
There are so many faults with this bold claim — but the very first thing any reader should get out of it is that when someone posts something in absolutes like this you should take their statement with skepticism and simply do your own research.
Not a single hp.
Really? The one thing I’ve learned in the 12+ years of tuning ALL kinds of motors, is you simply can’t deal in absolutes. Something may make power that could be taken as “margin of error” for the type of dyno you’re on or just very negligible gains that will never be noticed in the “real world” (say like 2-3hp for example, but it really is relative to the platform, etc — this is a question best left for another discussion).
The irony here is the statement preemptively tries to attack the skills and credibility of anyone doing a pump gas to E85 (ethanol) tune. I can’t say if this is really just the ignorance of the author of the post or just a gigantic ego that blinds him to reality.
My response is simple — “Oh please”. Fortunately for us, there are a few of us that are quite capable at what we do and we have actual tools at our disposal to do quantifiable testing instead of listening to someone spew hearsay on the internet.
Now for the sake of transparency — our pump gas is 92 octane, not 93 octane. However — big deal — the gains from going to 92 octane to 93 octane on a motor that isn’t ignition limited on 92 octane will be minor or nonexistent (and ours wasn’t — I could roll past MBT on the ignition map and not get detonation, there was just no more power to be had — which is what the statement about 91 octane implies). Hey don’t take my word for it — try it yourself.
As for the argument of WHY would E85 have gains over pump gas? What the author of said post seems to miss is the properties of the fuel — it is naturally oxygenated and has much better cooling properties than pump gas (regardless if said pump gas is 91-93 octane). I’ve done hundreds upon hundreds of E85 tunes — N/A or F/I — and I have yet to see a motor that doesn’t gain SOMETHING over pump gas (see what I did there?).
OK enough with the bull shit — here’s the results and the actual FACTS.
The car is a 2015 Nissan 370Z, with bolt ons:
VVEL and VTC had been previous tuned for this setup on pump gas, and were left alone for the E85 portion.
After the vehicle was filled up with E85 (tank took 16.5 gallons), fueling was tuned to maintain a proper air/fuel ratio without touching the timing map — the motor continued to run the same timing curve as it did on pump gas. Simply running E85 and maintaining proper fueling netted us 8whp. Wow, this is already certain more than Not A Single HP. And well above any kind of margin of error.
Next I adjusted the timing map to see if we could extract a bit more power — and 4whp was found. The amount of timing added was very negligible (could run this timing map on pump octane without seeing detonation — it however netted no gains on pump gas), and what we can gather from this is that due to the larger volume of fuel we are injecting with E85, we needed to ignite the mixture a minute amount of time sooner to get a complete burn of the mixture in the cylinder.
What does this look like overall? We picked up 12whp and as much as 12wtq through parts of the torque curve. Wow, that certainly looks like it’s a lot more than Not A Single HP. The best part about it all? This is free power — the stock fuel system supported E85 without upgrading anything — we ran stock injectors and the stock fuel pump. And to top it all off — fuel economy during freeway cruising was not impacted at all — if anything it was 1-2mpg better than what I was used to seeing on pump gas.
Really, what should be said about E85 gains on this motor is depending on the quality of your pump gas (because yes, even quality of 93 octane can vary), you may see between 2-4% gain in power with E85. You will also see it through out the whole power curve — more than can be said about the Stillen intakes I tested on this car, which had margin gains through a couple of spots of the power curve. Quite literally the dollar per HP gains from just running E85 (which is readily available at the pump by the shop) surpasses a $560 intake system.
There you have it — quantifiable results vs hearsay (or apparently I have some kind of secret to making power with E85).
Now I want to get into this and say that I was not in any way displeased with the Skunk2 Alpha race header on our Scion FR-S — the power pick up over stock was significant, and I expected it to maintain a bit of a torque dip being it is of equal length design and we see that dip remain in some form with all the equal length headers.
I have been running this header since it came out on the platform and after helping a few customers with their Tomei unequal length race headers, it really peaked my curiosity and I wanted to do a legitimate comparison. None of this “open two dynos from two different cars” as is so common with the vast majority of “comparisons” being done. It opens a can of worms and seemingly endless debate about this and that.
Some information about the parts and vehicle as it sits now. After spending countless hours testing and tuning the car using ECUTek (which I still offer for customer cars), I have switched it to a MoTec M1 stand alone ECU for it’s advanced features and ability to rapidly tune new configurations on the vehicle (no more “Flash and Wait” — given this car is a test mule, this saves countless hours on R&D and as they say: time is money). This comparison is done with both configurations tuned on the MoTeC.
Other relevant information:
The use of E85 fuel makes the vehicle much more consistent and the comparison much more reliable — the correction factor used on a dyno gives you an “estimate” to compensate for weather differences, but it is only an estimate. The use of E85 ensures the motor is able to be run at MBT for ignition timing even if we have a temperature swing — something that cannot be said of 91-93 octane pump gas. From my tests on this vehicle I found a whopping 1hp difference from running E85 in 40*F weather vs E85 in 90*F weather while tuning on the Dynapack. This helps aid in the consistency of the test. With pump gas a pull used for comparison that was done in 40*F can and will make a fair bit more power than a pull done in 90*F, so you really have to be careful when doing parts comparisons on pump gas which can just lead to more debate. That being said, even with the E85 fuel, I went to great lengths to make sure the conditions were pretty much identical between the two tuning sessions.
With the swap to an UEL race header it goes without saying that I expected to flatten out the torque in the area where the dip remained with the EL race header. However, I was also expecting to lose out on top end as this seemed to be the “expected” results between the two styles of header. And so I was about to find out how true this was… Bearing in mind the Skunk2 Alpha is a header I’ve been running for 9 months through many tuning sessions where I’ve eeked out everything there is to be had on this setup.
So without further ado — the test was simple. The vehicle was fully tuned in great detail with the existing setup — all fuel, timing and cam timing dialed in. As soon as the Tomei UEL was delivered it installed after the Skunk2 was removed and the car went right back on the dyno for more tuning.
I was indeed partially surprised by the results — the story is that EL should have better top end? Doesn’t look like it’s even remotely an accurate statement as not only did the UEL match the EL, it carried torque much better up top and didn’t drop off as soon. The mid range also filled in significantly and the low end was no worse than the EL — unless you count the blip at 2600 rpm. Which I don’t, as that blip is literally nothing more than a blip and I can assure you from driving the car with the Tomei on it, you won’t ever miss that blip, it’s like it doesn’t exist.
For the curious, the data from the two runs (which were about 5 days apart) demonstrates the weather conditions were virtually identical (RemoteTmp, Baro & RelHum).
This was an interesting test, and I’m happy I could fit it in before the car gets torn down for our turbo kit build — yet there are many more items I would love to test. It would definitely please me to have a full “header comparison” database for this vehicle — if I could borrow every header and take a week to test them all out, I would.
As it stands now — the car makes more torque than it did horsepower in stock form, which is quite amusing.
It’s that time again — got my hands on a 2015 Nissan 370Z and the typical bolt ons we see on this platform for some tuning and parts testing! As always, I tune the car completely stock first to get a good “tuned” baseline, and then retune after every set of mods. This is a very fun platform to tune due to the very flexible VVEL system.
So what we will have on this test is:
The parts in question are the following:
So without further ado, here we go.
It was pleasant to see there was actually a fair bit of room to improve over the stock mapping on the ECU — especially with the refined ignition control available to us now. One of the nuances I was able to fix was the throttle closer on the top end and the delayed throttle opening on the low end the stock ECU exhibits — this opened up some good torque gains down low and helped smooth the power curve up top.
The VVEL system is also extremely tune-able, and I was able to net extra torque down low with adjustments to this system — however through the rest of the curve Nissan got it mostly right, not surprising since the vehicle is stock.
Not a whole lot to say about these results — clearly the exhaust modifications made power after we bolted them up to the car — but since our starting point was already a “tuned” calibration, very minor changes were necessary to extract peak gains from the parts and the fueling was still dead on since the stock intakes had been retained for this portion of the test. I expect this would not be the case if the car was still running a 100% factory tune on the ECU instead of my tuned calibration.
Depending on the intakes you chose to put on a vehicle tuned via MAF (aka AFM), you can skew the fueling dramatically — fortunately with the Stillen V3 long tube intakes I found that the mass air flow calibration was very close to the stock intakes and only required some minor adjustments to maintain proper fueling throughout the curve. However, even with perfect fueling, these intakes actually LOST power throughout the WHOLE power curve.
You might be thinking to yourself, “Whoa, what? They’re just filters on a stick….”. Indeed I was quite surprised as well.
That’s where the tuning begins — the engine required significant remapping of the VVEL system to not only return to the power the stock intakes were making, but also gain power over the stock intakes. After fully retuning the ECU, our results are some minor torque gains down low and through the mid range, and about 10-12hp on the top end.
My personal thoughts? Wow that was a lot of work for minor gains — but it does go to show how a naturally aspirated engine is a finely tuned machine with all the parts working “in harmony” with the ECU mapping to actually make power. Sometimes one small change can have drastic effects.
Well, with that out of the way, what do the gains look like over the “stock tuned” setup overall?
And what does it look like over a completely stock vehicle?
It’s always quite fun and interesting to progressively install new parts and retune a car to see what kind of relative change we see with those parts. This weekend I had the opportunity to help Jay retune his 2015 Scion FR-S after he installed his Tomei UEL race header. We had previously tuned this car when it was stock (well, stock being relative — he had a slightly modified stock airbox and the stock muffler delete, nothing major).
Well on to the race header — after Jay installed it, we baselined the car on the existing tune to see just what the changes were from the part alone. It looked like so. I was a little surprised to see the results as we gained almost no power through most of the power curve with the part — it did however fill in the torque dip area, which was pleasant to see.
Next up is the retune. I exclusively used ECUTek for tuning these vehicles — they have the best stock ECU solution on the market with superb support. I knew there was definitely more in it — the motor on the car has dual VVT and quite a bit of work can be done to squeeze out any “hidden” power the new parts (the race header in this case) can support. Sure enough, we got solid gains! The graph on the right shows the gains from tuning over the race header baseline, very pleased with the power pick up.
Let’s compare that to our stock tuned power curve — and you can see a very solid power gain all across the board over the previously “stock tuned” car. The
mid range torque pick up was quite impressive — as much as 28 torque to the wheels with the header and retune. I have seen other headers on this platform pick up more power on the top end — but they did not fatten up the torque curve (especially in that mid range dip) as much as this header does. So it’s definitely a trade off if you’re looking at this header as your next mod.
Where does the car stand overall from when it was “stock”?