Welcome to the North American Subaru Impreza Owners Club Sunday November 23, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Water/Methanol Injection, Nitrous & Intercooler Cooling

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.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 03-22-2006, 04:16 PM   #1
Freon
Scooby Specialist
 
Member#: 88322
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Indianapolis, IN
Vehicle:
2009 BMW 135i

Default Comparing logical map routines for use with WI

A couple of the other threads really got me thinking about how you should properly control the duty cycle or total flow of a water injection system.

First I think it is important to set out some goals and purpose. I'm not going expound on these, but they are generally what I believe are the goal and purpose of a WI system:
  • Cool charge air due to adiabatic temperature increase when pressurising a gas (see: Ideal gas law, Charles' law, Boyle's law, etc)
  • Cool charge air due to inefficiency of compressor (see turbo compressor maps)
  • Maintain appropriate AFR throughout wide operating conditions

1D
I'm pretty convinced I would not bother with a system that is based purely on an on/off switch based on a boost pressure setpoint (i.e. System is either spraying or not, based purely on if manifold pressure is above or below a preset level). It just seems there is no way to really regulate proper flow. Way too much flow just past the set pressure and low RPM, then not enough at high RPM and peak boost. For instance, lets say you have a pressure set at 10psi and peak is 18 psi. You reach 10psi @ 3500rpm, system sprays just as much water as it does at 18psi @ 6500rpm. The problem should be largely obvious. With any combustable injected by your WI system, your AFR is going to be all over the place. Maybe it wouldn't be so bad with pure water, or low alky content, but thats a different discussion and I don't want to digress. Anyway, I'm going to call this 1D mapping.

2D
Variable based on boost pressure is surely an improvement. This means you can regulate decently from 10psi to 18psi at the same RPM. But you still cannot regulate properly across RPM ranges. At 18psi @ 4500rpm, you will be dumping richer than 18psi @ 6500rpm. The airflow is higher thus you need more mass water flow for the same cooling effect, plus the turbo may be starting to choke and drop off the efficiency island, requiring an extra bit on top of that. I'll call this 2D mapping. Alternately to boost pressure, one could use mass airflow or RPM. RPM makes no sense whatsoever, and I'm just going to dismiss it. Airflow makes more sense than RPM, but still you can have a situation where you need water flow at 15psi @ 3000rpm, but none at 0psi @ 6500rpm. These two points may have the same mass airflow. This method WOULD be great for drag racing or nothing but full throttle pulls since you can predict boost and "built it into the map", whereas you cannot really build in RPM into a boost based 2D map.

3D
There three ways that make sense. Boost, RPM, and airflow in all three possible combinations. I've built a few maps, and made some quick approximations of how I think each would look. I'm just trying to get the general shape and trends right. The arrows indicate approximately how spool up (yellow) and off throttle (red) would look like as far as cell hits, at full throttle. I am assuming some turbo that hits ~22psi @ ~5000rpm, then tapers back down to about 18psi @ 7000rpm. Since so many turbos run more boost in midrange than at redline, I think it is important to represent this. Particularly the simple 2D boost system screws the pooch on tuning for this.


Boost vs. RPM is the first that comes to mind. This is how the UTEC does its primarly fueling and timing maps (though remember fueling still uses airflow). The newest WRX firmware in fact lets you drive the spare solenoid using Map 5 boost map. So a big lightbulb goes off here for UTEC users. You should be able to drive, say, the Aquamist HSV directly with it, maybe need a transistor to step up current, but I think its actually in spec without. Only downside I see is the 13-14hz speed. Not sure how much of a concern this speed limit is, but I have to assume at high RPM there are cylinders that are getting or missing their water/alky mixture on some cycles. Anyway, we have boost so we can compensate for adiabatic temperature increase, and by adding RPM we can build some knowledge of airflow into the map as well. Again, higher RPM at the same boost, greater flow, need more WI. This is a simple extension of the 2D map. You can sorta build in knowledge of airflow and the turbo efficiency by inferring airflow, though not as well as actually reading mass airflow. Given the readily available UTEC, this may be great for many people.


Second is Airflow vs. RPM. This probably makes the least sense, although I will have a hard time justifying that, other than I have far more positive things to say about the two other possibilities. Adding RPM to airflow really just allows you to make some assumptions about what boost pressure probably is. For instance, if you are reading really high airflow at 3500rpm, you know you're probably running a lot of boost. 40lbs/min @ 3500 rpm means far more boost than 40lbs/min @ 6000rpm. But it really doesn't directly correlate with the temperature increase of the charge air. General trends are still at least predictable.


(please note the red line is an instant move, when the throttle plate closes, boost and airflow both fall to zero nearly instantenously)
Finally, we have airflow vs boost. The first thing that should pop into mind is this exactly how centrifugal compressor efficiency is determined. We have all the elements we need to know not only the theoretical temperature increase, but also total flow. Thus we can tune each cell for the additionl total entropy. Temperature, and mass flow. In fact, with a compressor map you could overlay and create a new airflow vs. boost entropy map which would essentially be what you would want your total WI duty cycle map to look like.

As a quick aside, another possibility would be to include a compensation map for EGT or intake air temperature. This would allow a bit more variance in ambient conditions. I think in general doing the WI DC 3D map right in the first place (airflow vs. boost) would lessen greatly wild values in any sort of EGT compensation map. Other 3D methods would yield more variance in final charge air temp and EGT to begin with. Certainly a minor bump in WI DC for high ambient temps is a good idea with all systems, as none inherently compensate for this.

I have to say I think airflow vs. boost is optimum. A slight compensation (+/- 5%) for high or low ambient temp would be icing on the cake.

I also would like to think a proper system would not always yeild the same AFR at the same RPM and boost, something people would generally expect to be consistent. I think this is actually "correct", especially when building in air temp compensation. Conditions

FWIW, I think the SMT-6 would also be capable of driving a duty cycle based device on an RPM vs. analog basis. (RPM vs. airflow, or RPM vs. boost) So those interested might want to check out their literature. I had one as a basic fuel tuner on my last car, though I sold it since it doesn't make sense for fuel and timing tuning on a WRX/STI.

Source file:
http://freon.shackspace.com/car/ecu%...%20mapping.xls
(note I will probably go back and tweak this file, it's not quite right, just for general reference)
* Registered users of the site do not see these ads.
Freon is offline   Reply With Quote
Old 03-22-2006, 04:38 PM   #2
BadTrip
Scooby Specialist
 
Member#: 62211
Join Date: May 2004
Chapter/Region: South East
Location: Burlington, NC
Vehicle:
03 WRX
Sonic Yellow

Default

Nice work. I'm not qualified to argue/debate the points one way or the other, but props for the info and data......and sharing it with the masses.
BadTrip is offline   Reply With Quote
Old 03-22-2006, 05:08 PM   #3
Richard L
Scooby Newbie
 
Member#: 85153
Join Date: Apr 2005
Location: Sussex England
Vehicle:
2004 Citroen Picasso
Silver

Default

Freon,

This is excellent work, it might put people off for life.

Richard

I will read it in more details - all good stuff.
Richard L is offline   Reply With Quote
Old 03-22-2006, 05:40 PM   #4
jmussetter
Scooby Newbie
 
Member#: 15350
Join Date: Feb 2002
Location: Bay Area, CA
Vehicle:
2002 WRX Sedan
Midnight Black

Default

If you were using a UTEC spare solinoid to control your WI, you could use MPS_UtecLogger to get a LOG_MAP of your IDC% for each cell. Then build a duty cycle map for the spare solinoid based on how much your injectors are spraying in each cell.

If your goal was to inject a constant % of mix (ie: 25% spray to gas) and have it be correct at all boost and RPM levels, then some simple math based on your fuel injector flow rate and your spray nozzle flow rate.

If you do the math right, you should be able to take the IDC of your injectors at each load point and calculate out the required IDC of the Alcohol spray for each corresponding load point in the map... and you'd have yourself a VERY linear responding system sprays your desired % mix ALL THE TIME!

-jason
jmussetter is offline   Reply With Quote
Old 03-22-2006, 05:57 PM   #5
Richard L
Scooby Newbie
 
Member#: 85153
Join Date: Apr 2005
Location: Sussex England
Vehicle:
2004 Citroen Picasso
Silver

Default

We do have a less well known 3-D system2s, RPM and boost based. The RPM scale has volumetric efficiency trim between 2000-9000 RPM.

All the failsafe features is in-built.

The map is very similar to one of yours. The steps show is linear interpolated. The step in the chart is slightly exagerated.




Do you have some thoughts on "failsafe" ? We have a 0-5V flow sensor that will allow you to do a "close-loop" fuel/ignition trim based on any spare analogue input to your ECU. This will iron out any clogged water jet problems.

Just a thought, not that expensive either.

Richard
Richard L is offline   Reply With Quote
Old 03-22-2006, 06:02 PM   #6
Richard L
Scooby Newbie
 
Member#: 85153
Join Date: Apr 2005
Location: Sussex England
Vehicle:
2004 Citroen Picasso
Silver

Default

I have given a great deal of thoughts on your choice of 3D mapping. I think RPM against air flow is more accurate, provided the sensor is reliable.

Richard
Richard L is offline   Reply With Quote
Old 03-22-2006, 07:41 PM   #7
klattery
Scooby Newbie
 
Member#: 48503
Join Date: Nov 2003
Chapter/Region: MWSOC
Location: Minneapolis, MN
Vehicle:
2004 STi ATP3071
Silver

Default

Nicely done Freon - glad you created a new thread instead of the Hmmm.

I would think the best 3d map for injection is the one which can best predict intake temperature. I'm not sure which of those combinations would do it best. Seems like Boost vs RPM or Boost vs Airflow would be great. Either would be an improvement over just boost. If I read you right though, Freon, the correlation with intake charge is why you prefer Boost vs Airflow?

Not sure why Richard likes RPM against airflow since I can't see how it would correlate with intake temperature (which I think is Freon's point too).

But I would love to hear your comments.
klattery is offline   Reply With Quote
Old 03-22-2006, 08:04 PM   #8
Freon
Scooby Specialist
 
Member#: 88322
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Indianapolis, IN
Vehicle:
2009 BMW 135i

Default

As far as failsafes, I am happy if the water injection system gives me a simple triggered output to start. From there a relay can be used to pop all sorts of things. I don't think it is reasonable to expect a good comprehensive solution

Personally, I would implement a bypass to the boost control solenoid so the wastegate sees all boost pressure all the time. This is simple with an additional BCS and a couple extra hoses. This instantly cuts 5-10psi off the top, depending on the turbo and wastegate setup. I would also have the trigger light a warning LED somewhere in the cabin. I've already swapped my stock solenoid, so I could use the stock one to bypass my 3-port. When energized, air from the compressor nipple would go straight to the wastegate actuator. On my stock turbo, its only a 7psi spring. Those runnin 20-22psi probably have 15 psi springs, etc. Only problem is it would basically make the solenoid sit energized 100% of the time when in failure mode, so it might fail quickly. I don't think the solenoids are designed to stay energized. $20 solenoid, $3000+ engine, you decide. FWIW, the factory solenoid is something like 30 ohm, to be safe, I'd say you want 1 amp of capability. If not, you'll need a relay.

It would be quite some doing, but you could technically trigger a map switch on a UTEC. The map switch is just something like a 12bit encoder. You'd need to build a small logic for it, but it is possible. Also of note the UTEC will not switch immediately when driving, regardless.

I can't think of anything else WI kit providers could do that would work on a variety of systems. Maybe intercept airflow and add a small degree of voltage? Not sure how practical that really is. I think the wastegate solution and a big blinking idoit light is probably more realistic.

I think the RPM vs. airflow would vary engine output the greatest due to air inlet temperature changes. Colder air will increase airflow, thus increase water injection on your map, even at the same boost level and lower charge air temps out of the turbo. I'm not sure you really want that under those circumstances. Sure, an airflow vs. boost would also increase water injection under that circumstance, but not as much, or if it did, it would be due to an aspect of compressor efficiency you can accurately model into your tune. Concurrently, I would think on really hot days, airflow vs rpm would not spray enough, again, the higher inlet temps would lower airflow at the same boost level. So you run less WI and higher charge air temps. Keeping boost level in the mix mitigates a portion, but not all, of that risk, which is why I like boost vs. airflow.

I guess a lot of it has to do with what you're using WI for. Intercooling? Improved combustion? Alittle of both?

Airflow vs boost is probably the easiest do as well. No need to decode the crank signal or use an inductive clamp. Every car is different. You really only need two 0-5v inputs for MAF vs boost. One tricky part is the airflow meter output is not linear. You need some precision up top since a tenth of a volt represents a lot of airflow at the top range, but not as much down low. As long as you can define the axis directly it should work well. For instance, I would expect airflow columns to be something like 0, 3.8, 4.1, 4.3, 4.5, 4.6, 4.65, 4.7, 4.75, 4.8 (volts).
Freon is offline   Reply With Quote
Old 03-22-2006, 08:28 PM   #9
Richard L
Scooby Newbie
 
Member#: 85153
Join Date: Apr 2005
Location: Sussex England
Vehicle:
2004 Citroen Picasso
Silver

Default

I may have made two mistakes assuming the MAF sensor is mounted after the turbo and intercooler and it is temperatures compensated. I don't have a Subaru so please ignore my suggestion.

I am very happy just to watch the thread as it progresses. I will ask a questions when I don't understand certain aspect of the Subrau setup. It has an interesting engine.

Richard
Richard L is offline   Reply With Quote
Old 03-22-2006, 10:26 PM   #10
Ultimate Collision Center
Vendor
 
Member#: 69072
Join Date: Aug 2004
Chapter/Region: Tri-State
Location: Peekskill
Default

The failsafe you are describing is very similar to what I sell (safeinjection by Snow Performance). It works great and I will have videos of it soon enough.

Dan
Ultimate Collision Center is offline   Reply With Quote
Old 03-23-2006, 12:02 AM   #11
bcblues
Scooby Specialist
 
Member#: 17256
Join Date: Apr 2002
Chapter/Region: MWSOC
Location: America's Outback
Vehicle:
04 STi
Blue w/ custom bug splat

Default

Thought-provoking! This is going to take a while for me to digest. Thanks, everybody, especially you, Freon!

What about running off IDC and Boost?
bcblues 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

Similar Threads
Thread Thread Starter Forum Replies Last Post
map for use with accessport. riderome AccessPort 0 04-21-2010 12:03 PM
Purchase of ECU Maps for use with Open Source Flash Utilities? f4phantomii General Community 10 03-30-2006 12:06 PM
Does Cobb have a stage 2 map made for people with a 100% BOV. phat pat Engine Management & Tuning 2 09-09-2003 10:46 AM
Spark Plug for use with nitrous?? Supapooh Normally Aspirated Powertrain 8 02-22-2002 10:19 PM
Catalytic Converter for use with a Turbo? Geoff Technical Forum Archive 2 06-27-2000 01:36 AM


All times are GMT -4. The time now is 08:21 PM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.