Welcome to the North American Subaru Impreza Owners Club Tuesday October 15, 2019
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC Technical > FA Series Factory 2.0L Turbo Powertrain

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.







* As an Amazon Associate I earn from qualifying purchases. 
* Registered users of the site do not see these ads. 
Reply
 
Thread Tools Display Modes
Old 01-06-2017, 09:18 PM   #51
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

All but one pivot tables are done. Corresponding graphs are done too. All tables and graphs update in real time.
* Registered users of the site do not see these ads.

Last edited by Boggie1688; 01-06-2017 at 09:35 PM.
Boggie1688 is offline   Reply With Quote
Sponsored Links
* Registered users of the site do not see these ads.
Old 01-06-2017, 09:38 PM   #52
Miggy-Verlander
Scooby Newbie
 
Member#: 456662
Join Date: Oct 2016
Location: Lakewood, Colorado
Vehicle:
2016 WRX

Default

Quote:
Originally Posted by kduni View Post
From a statistical stand point a sample size of 125 is sufficient to make conclusions with a high degree of confidence.
Hell no. Thousands of these cars are sold monthly, you think less than .0001% is enough to draw conclusions with a high degree of confidence? L. Oh. L.

OP, in spite of my other comments I appreciate this. While I dont think this can definitively answer reliability questions, there is still valuable information that can be found. Thank you for the time and effort!
Miggy-Verlander is offline   Reply With Quote
Old 01-06-2017, 09:48 PM   #53
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

Let's just focus on getting more data. More data equates to a better clearer picture. Where we end up is anyone's guess.

There are all those failures in the failure thread that are not in the data set.
Boggie1688 is offline   Reply With Quote
Old 01-06-2017, 09:59 PM   #54
kduni
Scooby Newbie
 
Member#: 434061
Join Date: Nov 2015
Location: West Texas
Vehicle:
2016 WRX
Metallic grey

Default

Quote:
Originally Posted by Miggy-Verlander View Post
Hell no. Thousands of these cars are sold monthly, you think less than .0001% is enough to draw conclusions with a high degree of confidence? L. Oh. L.

OP, in spite of my other comments I appreciate this. While I dont think this can definitively answer reliability questions, there is still valuable information that can be found. Thank you for the time and effort!
Why don't you educate yourself about statistical method, power, variance, sample size etc. You are showing only that you know NOTHING about statistics by your comments. Statistically speaking of sample size of 30 is enough to draw conclusions with some confidence the higher the number the more confidence one has.

I give anesthesia for a living, I have had a lot of statistical education to interpret studies for my practice. A study with 500 participants would be considered overkill/very strong and yet there are 7 billion people that could potentially be tested, which means you sampled 0.0000000714 percent of the population and yet decisions are made with people's life in the balance.
kduni is offline   Reply With Quote
Old 01-06-2017, 10:12 PM   #55
kduni
Scooby Newbie
 
Member#: 434061
Join Date: Nov 2015
Location: West Texas
Vehicle:
2016 WRX
Metallic grey

Default

OP, why not categorize the mods into stages? You would have to define where 1 stage starts and another ends, and while it would not be perfect or please everyone, as long as you are consistent in your classification of the stages it would be useful data that could then easily be graphed. Great job so far by the way!
kduni is offline   Reply With Quote
Old 01-06-2017, 10:26 PM   #56
Miggy-Verlander
Scooby Newbie
 
Member#: 456662
Join Date: Oct 2016
Location: Lakewood, Colorado
Vehicle:
2016 WRX

Default

Quote:
Originally Posted by kduni View Post
Why don't you educate yourself about statistical method, power, variance, sample size etc. You are showing only that you know NOTHING about statistics by your comments. Statistically speaking of sample size of 30 is enough to draw conclusions with some confidence the higher the number the more confidence one has.

I give anesthesia for a living, I have had a lot of statistical education to interpret studies for my practice. A study with 500 participants would be considered overkill/very strong and yet there are 7 billion people that could potentially be tested, which means you sampled 0.0000000714 percent of the population and yet decisions are made with people's life in the balance.
Yeah 30 people is enough to draw inaccurate and incomplete conclusions. Especially if those 30 people are all from the same demographic, like this. There is valuable information that can be obtained from this survey, but to call it conclusive is absolutely insane.

I worry for your patients.
Miggy-Verlander is offline   Reply With Quote
Old 01-06-2017, 10:33 PM   #57
zore
Scooby Newbie
 
Member#: 202850
Join Date: Feb 2009
Chapter/Region: Tri-State
Location: Central New Jersey
Vehicle:
2006 WRX Limited
Silver

Default

Most of these cars are going to end up in ditches or wrapped around trees anyway so I wouldn't worry too much if these are 200k motors or not.
zore is offline   Reply With Quote
Old 01-06-2017, 10:59 PM   #58
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

Quote:
Originally Posted by kduni View Post
OP, why not categorize the mods into stages? You would have to define where 1 stage starts and another ends, and while it would not be perfect or please everyone, as long as you are consistent in your classification of the stages it would be useful data that could then easily be graphed. Great job so far by the way!
You read my mind, so much so that I grinned.

For better or for worse, this is why I didn't opt not to have this as an option. Bias would be inserted into the results. Stage 1 vs Stage 2 are not very well defined and as a result people might select Stage 2 when they might have Stage 1.

However, taking the modification data and turning it into a Stage 1/Stage 2 result will take a bit more spreadsheet trickery. I started this, but didn't complete it yet. You can see it in the raw data sheet.

I decided to follow Cobb's definition. Hopefully this pleases the masses. Cobb defines Stage 1 as a tune. Stage 2 is defined as a J pipe. This is what I plan on using to dictate the Stages from the poll. I want to take the modification data, and create a formula that automatically decides if a user has Stage 1 or 2.

Unfortunately, I'm not 100% sure there is a formula that recognizes text. I'm still trying to decide how I'm going to implement this. Doing this one of by hand is not efficient. Maybe some googling later will yield a function in google sheets that can do this. Otherwise, I'm thinking of assigning a value for each mod. +1 for all none relevant mods: EGR, TGV, Catback, Intake, Filter, AOS, etc. +100 for tune or J pipe.

Stage 1 will be defined as equal to or greater than 100. Stage 2 will be defined as equal to or greater than 200. The spreadsheet will then have a column that auto populates with Stage 1 or Stage 2. That should be enough to create a table and make it graphable.

A project for tomorrow! Stay tuned!

Edit: Or maybe a pivot table that has each user with a custom formula. This might actually be better. I'm undecided!

Last edited by Boggie1688; 01-06-2017 at 11:17 PM.
Boggie1688 is offline   Reply With Quote
Old 01-06-2017, 11:57 PM   #59
2017wrx
Scooby Newbie
 
Member#: 452036
Join Date: Aug 2016
Chapter/Region: AKIC
Vehicle:
2017 WRX
WRB

Default

I just finished statistics and 30 is the number. It's in all the textbooks. Anything lower we would still be able to make a good conclusion, but use DF instead of N. This would change out confidence interval.

These calculations are usually make within 99 98 95 or 90 present confidence interval.
2017wrx is offline   Reply With Quote
Old 01-07-2017, 06:52 AM   #60
simpleJ
Scooby Specialist
 
Member#: 391748
Join Date: May 2014
Chapter/Region: NESIC
Location: The Shire
Vehicle:
2015 WRX CWP (sold)
2019 STi CWP

Default

There are several things that could hold this back from being statistically significant.


The n-value sure as hell isnt one of them
simpleJ is offline   Reply With Quote
Old 01-07-2017, 07:32 AM   #61
tramp
Scooby Specialist
 
Member#: 459629
Join Date: Dec 2016
Chapter/Region: International
Location: Knobbley Mountain
Vehicle:
2017 WRX base
WR Blue

Default

WTF? Everything around here seems to turn into some sort of **** wagging contest. If you disagree go elsewhere,if you agree input the data and wait for conclusions.

Me? I'm going to drive the car another 200 miles today. See ya.
tramp is offline   Reply With Quote
Old 01-07-2017, 12:14 PM   #62
Sillo
Scooby Specialist
 
Member#: 423270
Join Date: Jun 2015
Chapter/Region: Tri-State
Location: Long Island
Vehicle:
2015 WRX (sold)

Default

Filled it out, interested to see what comes out of this.
Sillo is offline   Reply With Quote
Old 01-07-2017, 01:33 PM   #63
GlarryHoodDIT
Scooby Specialist
 
Member#: 460907
Join Date: Jan 2017
Chapter/Region: RMIC
Location: Colorado
Vehicle:
2016 WRX

Default

Interesting idea, I signed up so i can participate! Hopefully we can mine some good data!!
GlarryHoodDIT is offline   Reply With Quote
Old 01-07-2017, 01:34 PM   #64
simpleJ
Scooby Specialist
 
Member#: 391748
Join Date: May 2014
Chapter/Region: NESIC
Location: The Shire
Vehicle:
2015 WRX CWP (sold)
2019 STi CWP

Default

Survey should also clarify if factory installed block is running or blown
simpleJ is offline   Reply With Quote
Old 01-07-2017, 01:43 PM   #65
GlarryHoodDIT
Scooby Specialist
 
Member#: 460907
Join Date: Jan 2017
Chapter/Region: RMIC
Location: Colorado
Vehicle:
2016 WRX

Default

Just did the survey, just a suggestion but shouldn't midpipe be changed to downpipe?

Someone please correct me if Im wrong, but from my own installs the J pipe and Downpipe are two different pieces, and the midpipe is part of the catback.

Should I have clicked midpipe with my catback?
GlarryHoodDIT is offline   Reply With Quote
Old 01-07-2017, 02:47 PM   #66
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

Quote:
Originally Posted by simpleJ View Post
Survey should also clarify if factory installed block is running or blown
Quote:
Originally Posted by GlarryHoodDIT View Post
Just did the survey, just a suggestion but shouldn't midpipe be changed to downpipe?

Someone please correct me if Im wrong, but from my own installs the J pipe and Downpipe are two different pieces, and the midpipe is part of the catback.

Should I have clicked midpipe with my catback?
Unfortunately its a bit to late to go back and make changes to the survey. It will have to be on of those things that we just take into account when looking at the data.

If I change it now, all of the previous 140 entries need to go back and update their information. You can already see how model year has had this affect. Nearly 33% of the data has no model year, and none of the entries have been updated.

I think the survey will be as is for now. I agree with kduni, once a trend in failure is identified, a secondary or follow up survey can be used to dig deeper into details.

The present data is already showing that this motor is indeed very reliable. If it wasn't the case, you'd see a much higher number of blown motors.

I'm going to work on the Stage1/Stage2 sheet, hopefully I'll have it done here soon.
Boggie1688 is offline   Reply With Quote
Old 01-07-2017, 03:00 PM   #67
GlarryHoodDIT
Scooby Specialist
 
Member#: 460907
Join Date: Jan 2017
Chapter/Region: RMIC
Location: Colorado
Vehicle:
2016 WRX

Default

As for the stage 1 and 2 sheet, are you going to make sections for stage 1+ and stave 2 +, and stage 2 + ebcs? Those seem to be the most common set ups.
My understanding usong Cobbs specifications is
Stage 0 - stock
Stage 1 - Stock + tune
Stage 1+ - Tune + intake
Stage 2 - J pipe/downpipe
Stage 2+ - J pipe/downpipe and intake
Stage 2+ebcs - j pipe/downpipe and intake and ebcs

Is that correct?
GlarryHoodDIT is offline   Reply With Quote
Old 01-07-2017, 03:09 PM   #68
kaeknows
Scooby Newbie
 
Member#: 357191
Join Date: May 2013
Chapter/Region: SCIC
Location: Socal
Vehicle:
2016 WRX
ISM

Default

13k miles, nameless catback exhaust only, 91oct california gas, engine warmed to temp before driving.

1k first oil change
4k 2nd oil change
8k 3rd oil change
12k 4th oi change

everything good here *knock on wood*
kaeknows is offline   Reply With Quote
Old 01-07-2017, 03:33 PM   #69
2017wrx
Scooby Newbie
 
Member#: 452036
Join Date: Aug 2016
Chapter/Region: AKIC
Vehicle:
2017 WRX
WRB

Default

Only 90 octane in alaska
2017wrx is offline   Reply With Quote
Old 01-07-2017, 04:17 PM   #70
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

Quote:
Originally Posted by 2017wrx View Post
Only 90 octane in alaska
DAMM REALLY!?!?! Wow and I thought us California owners had it BAD!

I'll throw 90 OCT in there because I doubt this affects anyone who has already put their info in.
Boggie1688 is offline   Reply With Quote
Old 01-07-2017, 04:27 PM   #71
Boggie1688
Scooby Specialist
 
Member#: 165955
Join Date: Dec 2007
Location: California
Vehicle:
2015 WRX Base
WRB

Default

Quote:
Originally Posted by GlarryHoodDIT View Post
As for the stage 1 and 2 sheet, are you going to make sections for stage 1+ and stave 2 +, and stage 2 + ebcs? Those seem to be the most common set ups.
My understanding usong Cobbs specifications is
Stage 0 - stock
Stage 1 - Stock + tune
Stage 1+ - Tune + intake
Stage 2 - J pipe/downpipe
Stage 2+ - J pipe/downpipe and intake
Stage 2+ebcs - j pipe/downpipe and intake and ebcs

Is that correct?
I'm not going to break out the stages like that. While it is nice to have, I just don't see it being helpful or valuable right now.

Also my "formula" for determining whether the car is stock, stage 1 or stage 2 is particularly dirty. I'm not a programming wizard, so I'm sure my implementation sucks. But it works.

If you'd like to make the formula for it I'll be happy to use it. The data is there, and I have left how I did it on the far right. You can't edit the sheet, so copy it to your drive or excel/openoffice.
Boggie1688 is offline   Reply With Quote
Old 01-07-2017, 05:18 PM   #72
ScoobNoob89
Scooby Newbie
 
Member#: 419506
Join Date: Apr 2015
Vehicle:
2015 WRX
Arrest Me Red

Default

I wouldn't try to force everyone into stages. Not everyone mods their car the same and it will skew the data if you try to pretend they do. For example, my car has an intake, charge pipe, TGV/EGRD, EBCS, BPV, and muffler deletes. From my understanding most people install the EBCS with a down pipe, but mine is stock. "Official" stage 2 is a new down pipe. But people like me can install stage 2 supporting mods on stock exhaust. My next engine mod will be an intercooler too. I guess at that point I'd be at "Stage-everything-but-2"...
ScoobNoob89 is offline   Reply With Quote
Old 01-08-2017, 08:10 AM   #73
2017wrx
Scooby Newbie
 
Member#: 452036
Join Date: Aug 2016
Chapter/Region: AKIC
Vehicle:
2017 WRX
WRB

Default

This made me look up the 2015 and 2016 wrx on consumer reports.

It has the worst used reliability rating.
2017wrx is offline   Reply With Quote
Old 01-08-2017, 10:51 AM   #74
GlarryHoodDIT
Scooby Specialist
 
Member#: 460907
Join Date: Jan 2017
Chapter/Region: RMIC
Location: Colorado
Vehicle:
2016 WRX

Default

I always felt it was kind of skewed trying to look at CR for tuner vehicles. The majority of folks dont mod the engine on their Corrolla or their 125hp DD they hate driving amyways. The aftermarket community around WRXs and STis is absolutely huge, and unfortunately that can lead to inexperience with people modding their cars are not understanding what theyre doing. Id be willing to be 8/10 people with tgv/egr deletes couldnt explain the function of either, and wouldnt understand any of the downsides (ie LSPI protextion) to removong them.
Not that Im trying to call out people as new, hell im very new to this scene, but yesterday I got into an argument with a "stage 3" 2015 STi driver, who told me an aftermarket downpipe didnt affect AFrs because it was post turbo!
How does the Forrester XT from '15 up do with the same engine? If that still shows high engine problems there might be something.
GlarryHoodDIT is offline   Reply With Quote
Old 01-08-2017, 11:54 AM   #75
tramp
Scooby Specialist
 
Member#: 459629
Join Date: Dec 2016
Chapter/Region: International
Location: Knobbley Mountain
Vehicle:
2017 WRX base
WR Blue

Default

Quote:
Originally Posted by GlarryHoodDIT View Post
How does the Forrester XT from '15 up do with the same engine? If that still shows high engine problems there might be something.
I'm curious about that too. Is it the same engine or a toned down version?
tramp is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

All times are GMT -4. The time now is 10:56 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2019 Axivo Inc.
Copyright ©1999 - 2019, North American Subaru Impreza Owners Club, Inc.

As an Amazon Associate I earn from qualifying purchases.