Andy.F

Its just a cheap ebay cold air kit but with a decent K+N filter (from an APS kit) on the end of it.

Difficult to say Shaun, for example, just bolting an APS CAIK on to a 99 car with the maf in the original housing and original position, changes the measured airflow by 25%, even rotating the maf by 90 degrees can have a massive effect.

14N-FR

iTrader: (3)

It is in the wing or engine bay with a feed?

Andy, it's Ian F btw, do you have any other mapping dates available?

Shaun

[Images removed due to bandwidth issues - will be sorted soon]

Three Day Modification Holiday - Day 3
Well that's been a busy three days on the trot!

Now that I've had time to collate data and start the thought process of explaining all this... let's get to it.

So with the new Hybrid GT2 Intercooler and Twister Induction Kit fitted, the next step was to re-visit Tracktive Solutions for the remap.

Richard Bulmer and I have spoken in-depth about the results so far and how progression could be made. It was perhaps the final reason why I decided to go with the FMIC if I'm being honest. However like I said previously, it was a shame in many ways that we didn't try the inlet mods in their own right.... but I had made the decision to "crack on".

I rocked up to Tracktive Solutions bright and early yesterday. We had a good chat about "life" in general and slurped on a couple of cups of coffee, before Richard's technician prepped the car for Richard.

After a couple of W(ide) O(pen) T(hrottle) pulls in 4th gear to see what's what, I asked Richard how it was looking initially. It was looking very very promising with the airflow at the MAF measuring just over 380g/s and still enjoying the increased ignition that was put into the previous map.

For the information hungry out there, the airflow now measured at the MAF had increased (over the previous LM450 Billet mapping session) by over 35g/s at this stage. Whilst not to be taken in isolation, comparing a like for like MAF scale and MAF position, increases in the flow measured should indicate BHP increases (some suggest that MAF airflow - dependant on position - can also indicate WHP). The more air you are consuming, the better and bigger your burn cycle.... ergo more power (all things being equal).

If I've learnt anything over the years, data does not always equate to power or improved performance.

Once Richard was happy that everything was "happy" he started to try different ignition advance, fueling and boost curves.

After about 1.5hrs he concluded "we're about done".

We basically ended up with a more ignition in places, fuelling altered and slightly increased boost.

I did ask Richard what his preference was on the "use of ignition vs boost" scenario and he explained that whilst making the most of ignition was his preferred option, the best would be loads of boost and loads of ignition. In reality the case is how the engine responds to each and finding a happy medium, that the engine prefers.

I can certainly understand the principal of leaning towards ignition, as I know only too well that it makes for a nice sharp throttle.... and in reality at the top end, can keep temps lower and a tad more efficient. It can all depend on a number of factors obviously.

One thing Richard did say is that when trying different combinations, a dyno was the easiest environment to "delve". I can understand the logic of this. However, this was to be "dyno free" at least with regards to the mapping side of things. This is maybe an avenue we could try later on using the Tracktive Dyno, to delve into some experimentation of what works best..... might be interesting.

I did double check with Richard how he had set-up the MegaROM maps. MegaROM is pretty extensive, in that you can (if required) have two full maps available. You can however only choose to alter the boost map for your second map and utilise the same ignition and fuelling maps as used for you main map. This is what Richard had "enabled" for mine, as he saw no great benefit in moving away from the "base" ignition and fuelling for my needs..... that would be better utilised if I was going to have another map for different fuel.

Simplistically put MAP A is high boost and MAP B is low boost.

I have no intention of using anything other than MAP A for the road. I can't see the point in having this set-up and running lower boost for road use... if I want lower boost I will use the throttle to control it!!

Potentially I'm starting to "push" the OE 2ltr engine now, so the only real time I may use MAP B is when I do any public track days. No point in me smashing the **** out of my engine for a quicker "public" track day session. Others will have differing views, but that's my opinion. Competition stuff would be a different thing all together! Let's be honest though.... based on my past record, that isn't going to be a consideration for me is it?!?

I did ask Richard what his opinion was on the increased airflow since the recent changes. He said (any inconsistencies with the MAF sighting aside) it was difficult to say. It was either the induction changes allowing more air in, the intercooler changes allowing more air through or a combination of the two. Without doing back to back tests it was difficult to know for sure. This was immaterial to a degree as we were, where we were now.

At this stage things looked very positive from a mapping data perspective.

Richard did say to me before I left, "If that doesn't make anywhere near 450bhp now... DON'T TRY AND CONTACT ME!!!". Of course he was joking, but I did think "You've jinxed it now!".

I'd made a call to Iain @ Litchfield Imports and also contacted David @ API to let them know of how things had gone.

It was now time to get some logging, acceleration and dyno runs completed, to see if the results from the mapping reflected real world positives.

I wanted to get some formal dyno data, but will need to organise that at a later date as that didn't go to plan.

The Drive Home
Well it was certainly "quicker" than the drive to Tracktive for sure! lol

So what was it like with the changes?

Power "feel" aside, the Hybrid appeared to drive exactly like it did with the OEM TMIC. No "dead" zone and it "felt" just as nice and good as previously, whether it was on and off throttle or WOT.

There was nothing to negatively comment on in this area, apart from a definite increase in oil temps.

I noticed this when driving over to Tracktive that morning (off boost obviously). Temps on cruise have increased by about 2.5degs and with boost, between 2.5degs and 5degs. The only thing that makes sense is the decreased airflow to the radiator and engine itself (including sump), because the intercooler is now in the way. I'm still running with the undertray and I suspect removal of this would help, but I'm resistant to doing that. In reality it is not a problem as the temps are well within tolerance for road use. However this may lean me towards getting an oil cooler fitted (which will be required imo for better track use).

The Induction Kit..... I don't like the noise, coupled with the audible noise from the re-circulating dump valve, venting back into the turbo inlet pipe. Perhaps I'm getting old, but I loved the quietness of the OEM set-up in that respect. Apart from that, no other negative comments.

Power delivery?

It felt quicker.... but then we all "feel" that when we have just had something done, whether it has actually improved things or not!

I wanted facts!

At this stage I could certainly say that from about 6k onwards, it felt a lot more grunty. It certainly felt more of a screamer, which coupled with the potential higher BHP and the JDM rev limit, would be no great surprise. Up to this rev range it felt the same as before, but again that is no great surprise either, as on the previous PowerStation dyno run from 6k onwards, something looked like it was holding the car back as the power levelled out. Hopefully that had now been sorted.

It did feel quick.

Facts and Figures
OK... it was back to using EcuTEK Delta Dash, my favoured data logger of choice, to see what I could determine from the latest round of modifications.

Off I set last night in ambient temps that were the same as previous logged runs and road dyno graphs.

First off was to do some Road Dyno 4th gear runs.....

I completed FIVE runs on my normal stretch of tarmac, same weight (in fact I must be a tad heavier now with the FMIC - meaning a small negative impact on the figures attained), similar environment conditions.

All runs were within 1bhp/2lbft of the medium figures and the results instantly gave me the information I was looking for.

LM450 Billet, Hybrid GT2 FMIC, Twister Induction Kit, Tracktive Remap


That's a +43bhp increase in what we saw over the LM400!

The peak torque was a tad down on what I was expecting.... but more on that in a while.

All in all we were looking at a solid 445bhp at the flywheel, based on previous comparison runs (Road Dyno vs MAHA Dyno) against this current increase of 12% in power.

We were already at a stage of comparison imo, where it was only really worth comparing the LM450 Billet vs the LM400 NON Billet. This was purely down to the fact that the LM400 Billet had already been identified as being the better lowdown without doubt. But in the name of openness here is that exact comparison.

LM450 Billet vs LM400 Billet


You can see how the LM450 Billet loses lowdown.... which we had already identified and expected. However from around 4k the LM450 Billet leaves the LM400 Billet behind.... again this is to be expected.

Just look at how the LM450 Billet holds on to the power at the top end though!

The final dyno comparison was to see how the LM450 Billet compared to the LM400 NON Billet. Let's be honest, the LM400 NON Billet was never a slouch down low, so this would be an excellent comparison.... can you have you cake and eat it?!

LM450 Billet vs LM400 NON Billet


In answer to the last question..... hell yeah!!!

I don't want to over egg this comparison but that is pretty remarkable, being able to gain an extra 45bhp with no detrimental effect in lowdown grunt.... already compared against a smaller turbo that is no lowdown slouch!

It's hard for me to convey how this latest generation LM450 Billet performs.... it just quite simply rocks!

Turbo Dynamics and Litchfields really have found an awesome housing and wheel combination in this unit. It really does work!

After completing these dyno runs it was then on to recording a few 4th gear WOT runs for Delta Dash Logging.

Let's go back to the lower than expected torque figures......

According to Delta Dash the ECU was retarding ignition (on its learnt correction curve) between 3500-5123rpm. The maximum retard was -3.5degs of ignition... Mmmm, that's where some of the torque has gone then in the same area on the graph! lol

Don't get me wrong, the ECU is doing exactly what it's meant to do. I dare say that it is running too much ignition for the boost and is compensating. The overarching Advance Multiplier is still @ 16, so everything is safe and the ECU is as happy as Larry.... but I'm being robbed of power that could potentially be put back in.

The exactly the same thing was happening between 6188-7116rpm, with a maximum retard of -2.4degs of ignition. Potentially robbing me of power at the top end.

The ECU is very intelligent, but with the Advance Multiplier @ 16, is also over cautious (which isn't a bad thing). It may only need a reduction of 2degs in the first instance and 1deg in the second instance, to ensure no correction is being made.

This does not always relate exactly, but each degree of ignition can mean 6bhp. I'm not sure of the torque effect though. Still some minor/quick fettling can be done to the map to hopefully better the curve even more. This is exactly how Delta Dash and EasyECU can benefit the customer (and mapper).

So talking of boost.... where are we with that now.

1.75bar peak, 1.68bar to 5500rpm, 1.63bar at 7100rpm!

The whole set-up seems to flow quite well.

Airflow has peaked at 388g/s with the MAF voltage reading of 4.94v's. We are close to the limit of the current MAF scaling now, which on the MegaROM code is scaled to a maximum of 400g/s.

One last thing on the logs..... how much NEGATIVE effect would you think an engine bay located induction kit gives?

Let's have a look....

Please consider that all run's done last night were based on a full 4th gear run. Pulling over for 5-10mins to look at the results (engine still running). Get some airflow up (couple of mile) and perform another run..... this was done more or less back to back for an hour. Fairly hard environment on the car.

Logs appear to show a story that I wasn't really expecting, certainly when you consider the induction kit has no formal and proper ducting to it. Ambient temps for runs was 12degs. Temp measured at the MAF (right at the inlet in essence) was 17degs at the start of the logs, which immediately went down to 15degs and held that for the entire run!

Whilst I'm not an expert, I would suggest that air ducting or temperature issues on the inlet are hardly a major concern, even though it is within the engine bay.... it was a real suprise.

Oh yeah.... how could I forget. The acceleration results! lol

Let's do the full rundown:

LM450 Billet
40 - 60mph = 2.5s
50 - 70mph = 2.0s
60 - 80mph = 1.9s
70 - 90mph = 2.1s
80 -100mph = 2.4s

LM400 Billet
40 - 60mph = 2.3s
50 - 70mph = 2.0s
60 - 80mph = 2.1s
70 - 90mph = 2.4s
80 -100mph = 2.8s

LM400
40 - 60mph = 2.6s
50 - 70mph = 2.2s
60 - 80mph = 2.2s
70 - 90mph = 2.4s
80 -100mph = 2.8s

VF
40 - 60mph = 2.6s
50 - 70mph = 2.5s
60 - 80mph = 2.6s
70 - 90mph = 2.9s
80 -100mph = 3.6s

I think I can accept the 40-60mph advantage of the LM400 Billet. The rest... I'll keep the LM450 Billet performance figures thanks!

At the level the car currently stands at, the OEM ECU is sucking up the power level and range of modifications. Based on the MegaROM facilities available and the driveability..... it's just a no brainer to retain it and save myself £1k+ for a replacement, which in essence will give me no extras in my instance for me personally.

So what next?
In my opinion there is potential for adjustments in the map to "better" the curves. That will happen via Tracktive sending me an altered ROM file.

In my opinion there is potential for adjustments to the induction side.

Ideally I would of liked to have put a much bigger airfilter in, but due to the requirement now (based on current flow) this may require a rescale of the MAF. There is little headroom left based on its current location and MAF pipe diameter. This will require a fair amount of mapping to get "right" (rescale) in the driveability stakes, and for the extra potential it may provide at the top end (bhp wise), I'm not sure it is worth it. I will have to consider this.

API will certainly better the ducting to the induction kit (which shouldn't effect flow, but just try and ensure cooler air is provided) and shield it from the rest of the engine bay.

I've been talking to API today to get this sorted.

That's probably it for this current modification stage, as I'm very happy with the results..... even if there is potential for some "more". At the end of the day the turbo and recent ancillaries have delivered exactly what I was told they would!

Bang on!

I'll let you know what transpires with the "tweaks" and also (when appropriate) what we will be doing next (yes... you did read that right!).

Just to summarise who has helped, supported, provided what service / product for this current round of mods:

(In no specific order)

Tracktive Solutions
Mapping / Advice

Litchfield Imports
Turbo & Fitting / Advice

API
Induction Kit / Fabrication / Fitting / Advice

Harvey Smith
Hybrid GT2 Intercooler / Advice

Andy Forrest
Advice (and I'm not even a customer - what a gent)

Many thanks to you all.

Watch this space as time stands still for no man in the world of Subaru modifications!

Toffee

iTrader: (3)

Sorry if I have missed this in the above post, but what torque did you manage to get?

Shaun

Based on my calculations, and pulling it inline with a Dyno Dynamics rolling road (same as what you used) - as these give much higher torque figures than MAHA rolling roads - I would say in the region of 412-420 ftlb.

This is lower than expected, as mentioned above, because ignition is being pulled in this area. A different map should sort that and pick the peak torque up.

However, peak figures tell only part of the story as the torque throughout the rest of the rev range is much higher, in comparison to the LM400 runs. The acceleration figures back that up!

Don't concentrate on peak figures!

ScoobyDoo69

iTrader: (17)

Why would timing be pulled if it was mapped yesterday?!

Shaun

Whether it was mapped the other day or last year is irrelevant. What is relevant is that it must of been fairly close to optimum, and whether the difference in environment conditions or load (it was a lot cooler than when it was mapped) has just knocked it over... I don't know.

I always send the logs over to Tracktive when I test, and any changes required are made and the ECU is reprogrammed. Since I have this facility it changes the approach to how Richard maps I suspect. We try several different things remotely.

If it was Joe soap without easyecu, then I suspect the map would be slightly more conservative.

It's horses for courses.

It is 100% safe to do this and ultimately the ecu is doing exactly what it's meant to be doing.

ScoobyDoo69

iTrader: (17)

No disrespect to Richard, but when I have my car mapped, I would want it mapped perfectly there and then. Having to do logs and send them back and forth, uploading maps etc seems a bit unprofessional to me. But if you're happy doing that then fair enough.

APIDavid

iTrader: (4)

That'snot the way it is usually done. It is just Shaun being a keen type playing and having fun. Reminds me of when I was his age -





last year if I recall

Heeeeeeeeheeeeeeeeeee

rickya

iTrader: (18)

When your driving your Scooby you have no idea how many times either FLKC or IAM has pulled timing unless you are logging with e.g. EasyECU or Romraider etc. Any car that has been mapped regardless of when will go through this process of knock control, its part of how the ECU is designed. This could be due to higher temps, slightly lower octane fuel batch, noise that ecu thinks is knock, etc etc. Most of the time the FLKC may pull timimg just for a second or two as a precaution, its what the ECU is meant to do.

As far as sending logs & rom files back & forth, I think its a breath of fresh air for mapping IMO. I recently sent Duncan (Racedynamix) a few logs for him to have a look at & make sure certain things are as they should be. Within a couple of log emails back & forth, I was reassured within 30 mins & I didn't even have to leave my couch!!

Great informative thread Shaun

evonorth

The dyno queen is back to his best,,,
I hope you have a few sprints lined up for this car soon,,
The point of more power is what sir

Shaun

I think you have totally misunderstood the concept of this as a "project" (rather than a retail service) and the way everyone involved is working on this. This is not about me paying a supplier, having it fitted or mapped and job done.

The reason for this mapping approach is quite simple... I want to learn / share my experience and Richard is willing to spend more time in getting everything as optimum as possible (this was all made clear up front in this project thread, specifically in regard to the mapping side of things).

"Perfect" in a customers eyes is not always the most "Optimum" in reality of what could be achievable given more time. To try different things and see how the car performs with those changes is the bit I'm interested in. I have regulary had multiple maps (but have not shouted about it) to try different things during this whole project from Richard, because that is what we agreed we would do and I have the software to do it and Richard has the enthusiasm.

A tuner will variably continue to develop and remap their own cars over a period of weeks and months. Because they have the ability and tools to do that.... this specific project is no different with Tracktive. In many ways I have a golden oppurtunity, that otherwise would potentially cost a fortune if this was a retail item of work.

If it was a "drive in - drive out" service, the approach would of been totally different. As it would of been with Litchfield Imports and API etc.

The way the whole of this project is being done is different to how you would go about modifying your car normally. It would cost a retail customer £1k's and £1k's.

I hope this makes it a bit clearer.

Like they say.... you get nothing out, if you put nothing in.

Shaun

Trust me Mr North..... those dyno queen days are long gone, but unfortunately the reality is, people want to see those figures so I have no choice.

Car will be on track mate..... few more track orientated mods to do over the coming months (for reliability sakes) and she will be on the black stuff.

evonorth

Sound job mate
i bet its an awesome rd car for sure

Hammer man

iTrader: (25)

"Fun"... Its what these cars are all about.

MelTypeR

iTrader: (4)

Top Results Shaun!!!

Now we just need to get them both out the track!

Mel

Shaun

Just so we are clear what I'm referring to about the ignition retarding. I'm referring to the Learnt Correction and NOT the Knock Ignition Retard within Delta Dash. Knock Retard sits happily at 0, along with the Advance Multiplier at 16 (maximum value). I just wanted to make that crystal clear.

Paul,
It's a totally different animal to my old SPEC C. The grunt of the SPEC C made that a better road car imo. I'm referring to the mid range torque really, as that just went regardless of the gear you were in..... very lazy to drive. With 550lbft..... it's hardly surprising though.

This is quite a "buzzy" set-up, which suits the JDM rev limit and turbo quite well actually.

It's also easy to drive and goes well enough for me.... for now. You don't have to rev it to 8k to make progress.... but you can if you want to.

I suspect this is at the same peak power level as Rob's SPEC C was, when he ran just VPower. The difference is the spool and lowdown grunt though of this LM450 Billet, compared to the turbo Rob had. Saying all this though, Rob's car was hardly slow on tight handling circuits was it.

Mel,
I'm very happy I had this turbo put on in the end. I was slightly hesitant over my decision originally, thinking "What have I gone and done this for..... I'm going to regret it". I'm glad that couldn't be further from the truth.... it really is a great turbo for a 2ltr imo.

If I was to recommend a turbo for a 2ltr, I would recommend the LM450 Billet straight off the bat. No questions. It's just so driveable for a turbo that will give 450bhp. The LM400 vs the LM450 Billet comparison is truly outstanding.

Would be good to get on track with you Sir.

Shaun

Instead of "running" straight to Tracktive I have decided to do my own "personal" investigation today (although I have asked for another map, but Richard has just shot off to Jordan for several days to map a car - so in effect the running has started lol), with regards to delving into the Learnt Ignition Correction. I like to learn and play with my toys!

As I have said before I have the ability to retard ignition myself via Delta Dash. Unfortunately I'm only able to retard ignition across the whole of the map, not in specific area's. This isn't a big deal, as at least a reduction in ignition overall, should see an effect on the Learnt Correction.

Assuming the bad weather holds off, I should have some feedback later today. Completing some more Road Dyno runs with any changes, should also see if it's picked up any power in those specific rev ranges already being corrected.

Should be interesting in a geeky kind of way.

Barnesy

Any graphs of the LM450 billet against standard VF?

harvey

iTrader: (48)

Great effort Shaun, lots of interesting information and good results. Only a matter of time before you try to go further???
Because you are close to the limit of your MAF Sensor you may not want to go further (for now) but a bigger air filter will probably result in a further improvement although I am not sure by how much. Moving to a CAK will result in reduced air flow unless the inner wing filter has a greater surface area and there are no tight bends. The under bonnet air temperature you have at present is fairly good but a simple and easy to make cold air divider would get the filter breathing at close to ambient which would be the best solution.
I discovered a long time ago that an efficient induction system was a far greater priority than the disadvantage of partially heated air from the engine bay, providing the temperature was not excessive. With a simple and cheap maximum and minimum garden thermometer anyone interested in testing air filter temperatures or improving them can do this for around a tenner.
I made reference to using a manometer to evalute different induction systems and seek improvements. It is certainly not a flawed method and here is a link to an article published by Julian Edgar that someone kindly sent to me. I see there were five parts to the article in all but have only skim read article 2 but it looks very interesting.
http://autospeed.com/cms/A_0637/article.html

Good luck with further fine tuning.

Shaun

LM450 Billet vs VF (decat, panel filter, remap)

Shaun

Agreed about the Induction side..... it would be interesting to see what a much bigger filter would do to the set-up. However, since we are close to the current MAF limit without delving into a rescale, I'm not wanting to alter the induction at this precise minute. Litchfield's have offered up a large MAF pipe and larger cone for testing..... so that maybe worth looking at in the future to see if any benefits can be seen.

API will be fitting some inner wing ducting to the current induction kit and also farbricating a devider between the filter and "hot area". This should ensure a "best bet" environment.

However... yesterday (whilst I was out logging and adjusting the ignition as mentioned in my previous post) I saw the same "minor" increases between ambient and inlet temps. Under much warmer conditions than previously, it was 20-21degs ambient and the inlet temps on boost were 23degs logged. More on what happened with this yesterday later on today, after I manage to get some more logging done tongiht (need some comparative lower temp - as all my previous dyno runs - Road Dyno information). Logic and current data from yesterday suggests that this are looking "healthy" though.

juggers

iTrader: (1)

Shaun you need to join a race/rally team, this is one of the most detailed threads i've ever read!!

Shaun

I would put it another way.... I need to get a life!!

juggers

iTrader: (1)

You could also say that to

Shaun

Right...... just been out again tonight in slightly cooler temps. Temps were not as cool as I wanted it to be. They were 23degs inlet temps yesterday and 17degs inlet temp tonight. I really wanted them to be around 13degs as per the majority of other Road Dyno runs. Still, 4degs isn't going to make any really difference.

However, the results were not as I had expected.

When I was running with the OE TMIC, I would see quite a bit of variance in power figures recorded within Road Dyno, between varying ambient temp runs. So I was expecting to see similar now.

That has not appeared to be the case.

The power run with 23degs inlet temps (20degs ambient) vs 17degs inlet temps (14degs ambient) is minute in reality and the curves are more or less the same. I suspect this is now down to the better efficiency of the Hybrid GT2 FMIC.... which is good.

So what did the difference to retarding the ignition by 2degs (to remove any Learnt Correction) make...... diddly squat!

In fact if anything it has made the acceleration between 40-60mph worse as it has increased the time by .1 and on the original map it was holding onto the power slightly better at the top end!

So whilst I have no correction being logged, the curves are not as good and the car accelerates slightly slower.

I think I'm going to reset the map back as the ECU seems be coping quite well on the original map.

ScoobyDoo69,
There lies your answers!

Shaun

To substantiate what I was referring to above:

LM450 Billet Original Map (Learnt Correction Logged) vs Original Map with -2degs of Ignition (No Learnt Correction Logged)


Red line is the run with the -2 degs of retard.
Blue line is the original 450 map.

The eagle eyed among you will realise the vehicle weight for "Red" line has been reduced. The car had 1/4 tank less fuel than normal, so the weight was adjusted to allow for this (not that it makes any great difference, but just ensuring everything is equal for a true picture).

APIDavid

iTrader: (4)

" If I was to recommend a turbo for a 2ltr, I would recommend the LM450 Billet straight off the bat. "

Shouldn't that read 2 .0 lt TWIN SCROLL Shaun ??

I would hesitate to recommend the LM450 or SC46 to most 2.0 STi Type UK drivers.......

David

Shaun

Yes sorry... I meant for a 2ltr twinscroll set-up and the statement was in relation to the LM400 and LM450. I haven't tried any other "makes" so unable to comment on that one.

kennyc

iTrader: (1)

Shaun - only mods that have been done to your JDM 2.0l is Litchfields billet 450 turbo kit, front mount, fuel pump, injectors, exhaust and remap? and you are getting about 100bhp and 60lbft ATW and you losing about 3-400rpm spool up to 3400rpm

What is safe to push this engine to? Did you do testing with top mount with this turbo?