She's finally running right =) - Page 2 - Miata Turbo Forum - Boost cars, acquire cats.

Welcome to Miataturbo.net   Members
 


Reply
 
 
 
LinkBack Thread Tools Search this Thread
Old 07-15-2009, 12:21 PM   #21
y8s
2 Props,3 Dildos,& 1 Cat
iTrader: (8)
 
y8s's Avatar
 
Join Date: Jun 2005
Location: Fake Virginia
Posts: 19,037
Total Cats: 406
Default

Quote:
Originally Posted by Stein View Post
Closed loop, as Travis indicated, offers retarding of timing by knock sensor reading, but it is more than that. I also compensates by adjusting timing by water and air temperature. Just like open and closed loop fuel, it adjusts timing based on temps. Take a look at the two corrections tabs under the closed loop page in the "corrections" tab.
So it's named poorly.

Closed loop in control theory means you are providing an input value that generates some system output value. In the case of ignition timing, there's no separate output measurement to compare to the input. It just sets the value and goes on its merry way.

That doesn't change if you add in corrections. They're still considered inputs to the final timing value. We assume the output is the actual timing in degrees but have no way to know without a timing light.

So really they're both open loop but one has corrections based on sensors.
y8s is offline   Reply With Quote
Old 07-15-2009, 12:48 PM   #22
Elite Member
iTrader: (9)
 
Join Date: Jun 2008
Location: Houston, TX
Posts: 1,583
Total Cats: 8
Default

Quote:
Originally Posted by y8s View Post
So it's named poorly.

Closed loop in control theory means you are providing an input value that generates some system output value. In the case of ignition timing, there's no separate output measurement to compare to the input. It just sets the value and goes on its merry way.

That doesn't change if you add in corrections. They're still considered inputs to the final timing value. We assume the output is the actual timing in degrees but have no way to know without a timing light.

So really they're both open loop but one has corrections based on sensors.
Closed loop theory says that the input is modulated by a dependent variable of the input. Spark advance is the input, and the dependent variable in this case is knock index. Now this is kind of a partial differential type of control system, because there are actually thousands of inputs that would create the undesirable knock, but were summing it all up to the input spark advance.

The closed loop controller would be something like this.
Input=degrees of advance BTDC
Output=knock index
dAdv=change in input to modulate the output
38* advance,0 knock, 0dAdv
40*,50,-1
39*,10,-1
38*,0,0
An open loop controller looks something like this
38*advance,0 knock,0Adv
40*,50,0 <computer
42*,120,0 <computer
45*,Max,0 <computer
Next event motor explodes

The learning parameters actually go both ways, and have 2 dependent variables. One is torque, and the other one is knock.
If torque is up, and knock is 0, advance timing.
If torque is up, and knock is non zero, retard.

So if knock is ever non-zero the controller will retard the ignition timing. If knock is zero then it runs through a PID.

This modulates increases or decreases in timing:
If torque is up a little bit, and knock is 0, advance a little bit
If torque is down a little bit, and knock is 0, retard a little bit

Torque is found by the differential in omega, or angular velocity of the motor. If all is fixed, and angular velocity is up with timing changes then advance basically. If angular velocity goes down with more advance then retard.

Through basic physics though dW can be equated to dT. Which is how they get their advance matrix which is what runs through the PID, which is what tells the system how much to advance...
TravisR is offline   Reply With Quote
Old 07-15-2009, 02:00 PM   #23
y8s
2 Props,3 Dildos,& 1 Cat
iTrader: (8)
 
y8s's Avatar
 
Join Date: Jun 2005
Location: Fake Virginia
Posts: 19,037
Total Cats: 406
Default

I undrstand how the knock sensing strategy works, but it really should be called "closed loop knock" not ign advance. ign advance isn't the output. it's a technicality

As far as the torque sensing, the way I read it in the manual was that it measures a change in the engine period based on rotation of the engine relative to where it was.

example: a constant speed motor will rotate 360 degrees each time. when you accelerate, if torque changes, there is a relative twist in the driveline. the engine rotates *slightly*. so instead of the crank spinning 360, it now spins (360-*slightly rotated motor*) and the adaptronic can measure that when it gets the next crank pulse. Right or no?
y8s is offline   Reply With Quote
Old 07-15-2009, 02:12 PM   #24
Murderator
Thread Starter
iTrader: (76)
 
18psi's Avatar
 
Join Date: Apr 2008
Posts: 36,163
Total Cats: 2,573
Default

I love where this thread is going Your guys' discussion is helping me understand more the ignition tuning side of the Adap. Going to install my mbc today (for now, later I want to run an ebcs) and start tuning around 10psi. Right now it was hitting about 6 or so.
18psi is offline   Reply With Quote
Old 07-15-2009, 02:45 PM   #25
Elite Member
iTrader: (46)
 
Stein's Avatar
 
Join Date: Dec 2007
Location: Nebraska
Posts: 4,812
Total Cats: 138
Default

Quote:
Originally Posted by y8s View Post
example: a constant speed motor will rotate 360 degrees each time. when you accelerate, if torque changes, there is a relative twist in the driveline. the engine rotates *slightly*. so instead of the crank spinning 360, it now spins (360-*slightly rotated motor*) and the adaptronic can measure that when it gets the next crank pulse. Right or no?
I was under the impression that it measured the acceleration time of one rotation compared to the previous rotation and, based on the same RPM, kPa and TPS data that is stored, if the time gets shorter, it knows it can advance more next time. If the time to go through the revolution gets longer, it knows that it lost torque and can bump it back. Basically, it says, "the last time that I was at these settings it accelerated faster (or slower) so I need to bump timing up or back".

That said, I assume it will only adjust the timing in areas of the map where you are accelerating, so constant cruise cells won't see much action. As it pinged the cruise cells, it would think "humph, changing timing here didn't do much" as the motor didn't show a positive or negative speed change. I don't know how much it is tied to TPS, but I'm (strictly guessing) thinking it has to be a minimum TPS percentage.
Stein is offline   Reply With Quote
Old 07-15-2009, 03:07 PM   #26
y8s
2 Props,3 Dildos,& 1 Cat
iTrader: (8)
 
y8s's Avatar
 
Join Date: Jun 2005
Location: Fake Virginia
Posts: 19,037
Total Cats: 406
Default

Quote:
Originally Posted by Stein View Post
I was under the impression that it measured the acceleration time of one rotation compared to the previous rotation and, based on the same RPM, kPa and TPS data that is stored, if the time gets shorter, it knows it can advance more next time. If the time to go through the revolution gets longer, it knows that it lost torque and can bump it back. Basically, it says, "the last time that I was at these settings it accelerated faster (or slower) so I need to bump timing up or back".

That said, I assume it will only adjust the timing in areas of the map where you are accelerating, so constant cruise cells won't see much action. As it pinged the cruise cells, it would think "humph, changing timing here didn't do much" as the motor didn't show a positive or negative speed change. I don't know how much it is tied to TPS, but I'm (strictly guessing) thinking it has to be a minimum TPS percentage.
I hope it's not that way because it's way too unreliable. what if you go up a hill while accelerating? down? a bump in the road will affect engine rotation speed in some cases.
y8s is offline   Reply With Quote
Old 07-15-2009, 03:13 PM   #27
Elite Member
iTrader: (46)
 
Stein's Avatar
 
Join Date: Dec 2007
Location: Nebraska
Posts: 4,812
Total Cats: 138
Default

Quote:
Originally Posted by y8s View Post
I hope it's not that way because it's way too unreliable. what if you go up a hill while accelerating? down? a bump in the road will affect engine rotation speed in some cases.
Dunno. Just my impression of what I read. I'm guessing it must have to ping a certain number of points to weed out fliers before changing. Maybe it stores 10? 20? times at the same setting before allowing it to change. Obviously, strictly speculation here.

Don't worry, Travis will look into it and give us a full report so we don't have to guess. Right Travis? Becuase I don't know, I've never tried it.
Stein is offline   Reply With Quote
Old 07-15-2009, 03:31 PM   #28
Elite Member
iTrader: (9)
 
Join Date: Jun 2008
Location: Houston, TX
Posts: 1,583
Total Cats: 8
Default

A recitation from ye Adaptronic Bible:

5.2.10. Adaptive Ignition
The adaptive ignition should be used as a tuning aid as in some cases left to its own
devices it can retard/advance ignition to non-optimum levels. It works by dithering
the calculated ignition timing by a fixed amount “angle”, and back to the calculated
value every set of ignition events ("no of engine periods"). Because the engine drives
a compliant load (through flexible engine mounts and tyres), a change in torque will
cause a phase shift in the engine's rotation. Over a small period of time this
corresponds to a small change in RPM. The RPM is sampled at the beginning, middle
and end of every ignition dither cycle. The difference is calculated; if the RPM is
greater when the ignition is more advanced, this corresponds to an increase in torque
during the advanced ignition period. If the RPM is the same, there is no torque
increase. This RPM shift is referred to as the "Advance Metric", and will have a
positive value if the torque will be increased by advancing the ignition.
By default, the ignition dithering will only take place if the conditions for adaptive
ignition are met (that is, minimum/maximum engine speed and minimum/maximum
coolant temperature). When the ignition dithering is not in effect, the Advance Metric
will read zero. Under the "Always sense torque" tickbox is enabled in the Adaptive
Ignition parameters, the dithering will always be active and the Advance Metric will
always be calculated. This is disabled by default as it has a poor effect on idle
smoothness.
Once the advance metric gets above the higher threshold, the advance metric is reset
and the current cell in the ignition map is advanced by 0.2 degrees (slow converge) or
1.0 degrees (rapid learning). The other adaptive parameters such as the RPM and load
tolerance are set as per adaptive fuel control parameters.
- The knock threshold determines the level of knock allowed before the ignition will
be retarded.
- The minimum and maximum rpm values set the conditions for entry into the
adaptive ignition loop.
- Always sense torque will dither the ignition regardless of the conditions, when it is
off dithering will only occur when in adaptive ignition mode.
- Update other parts of map will check that all loads and rpm’s higher are equal or
more advanced, all loads and rpm’s lower are equal or more retarded, if not they will
be changed.
The adaptive ignition system is noisy, to remove some of this noise, there is a filtering
option which uses the number set to determine the percentage that comes from the
sample taken, while the remaining percentage comes from the previous value. The
lower and higher thresholds set the minimum and maximum values that the adaptive
ignition can change the map, the stabilize time sets the time that the ecu needs to be
held at a set rpm and load for the adaptive ignition to occur.
TravisR is offline   Reply With Quote
Old 07-15-2009, 04:13 PM   #29
Senior Member
 
Join Date: Nov 2007
Location: somewhere in NoVA
Posts: 550
Total Cats: -1
Default

Uhm, so in summary... ?

She's finally running right =)-adaptive-ignition-screenshot.jpg

Really, the closed loop statement by Y8S it more so how I read it... but either way if it gives me what I want, I don't care what they call it.

Open Loop would be great if you were the master of tuning and knew that you were perfect... closed loop with adjustments seems to be the best bet.

Anyhow, I am now running into other questions after reading the manual. It was not the clearest in some areas; but it is helping lead me in the right direction.

I logged a short drive and my max knock was at 125 "hits" at 5800 RPM with 150ish kpa / <13.1 AFR. I for sure need to work on my timing for higher boost levels.

As far as knock sensor calibration, how are you all doing this? I looked at my log and see that there is likely no knock in some of the areas that my log shows there is knock... I am entering values into knock offset. Tho I really can't tell what is real and what is not. Well the funky part is that I could not hear knock but the sensor says otherwise. So I quickly have done this... based on my log. When I see that I am off the chart... I figure that this is the pin on a gernade?

She's finally running right =)-knock-offset.jpg
Prospero is offline   Reply With Quote
Old 07-15-2009, 05:41 PM   #30
Junior Member
 
Join Date: May 2008
Location: Ft Walton Beach, FL
Posts: 48
Total Cats: 0
Thumbs up Awesome!!!

Now Its my turn. I just ordered my LC-1. should be here in 3 to 4 days. I hope that solves the starting problems I have. I may order a 3bar MAP replacement, just in case mine is no working properly. Ill be running normal aspiration with stock injectors. Once I get it running properly then and only then ill throw in my IC and GT2560R Turbo plus some 600cc injectors. Does anyone has a suggestion on what brand and type of injectors I should go with?

thanks
etnad is offline   Reply With Quote
Old 07-15-2009, 05:50 PM   #31
Elite Member
iTrader: (46)
 
Stein's Avatar
 
Join Date: Dec 2007
Location: Nebraska
Posts: 4,812
Total Cats: 138
Default

Quote:
Originally Posted by etnad View Post
Once I get it running properly then and only then ill throw in my IC and GT2560R Turbo plus some 600cc injectors. Does anyone has a suggestion on what brand and type of injectors I should go with?

thanks
How much do you want to spend? New Deatschwerks 600's are $349. I have them and they worked great. Plug and play. Used RX7 460's or 550's are 75-100. Cleaning is about $80. Some go without cleaning but I would not suggest it as the condition is really an unknown.
Stein is offline   Reply With Quote
Old 07-15-2009, 05:53 PM   #32
Senior Member
 
Join Date: Nov 2007
Location: somewhere in NoVA
Posts: 550
Total Cats: -1
Default

IMHO, Deatschwerks 600's are a great way to fly. Most users of the Adaptronic are running with them and it will make it easier for all of us to work together all having a common part.

The install was easy as pie, they were great quality, and I love the price!

Cheers,
Prospero
Prospero is offline   Reply With Quote
Old 07-15-2009, 05:58 PM   #33
Murderator
Thread Starter
iTrader: (76)
 
18psi's Avatar
 
Join Date: Apr 2008
Posts: 36,163
Total Cats: 2,573
Default

Quote:
Originally Posted by etnad View Post
Now Its my turn. I just ordered my LC-1. should be here in 3 to 4 days. I hope that solves the starting problems I have. I may order a 3bar MAP replacement, just in case mine is no working properly. Ill be running normal aspiration with stock injectors. Once I get it running properly then and only then ill throw in my IC and GT2560R Turbo plus some 600cc injectors. Does anyone has a suggestion on what brand and type of injectors I should go with?

thanks
600's from dw for the WINNNNN
What makes you think your Map sensor is misbehaving? Do you have a log of it acting weird? I knew mine was toast when NO MATTER WHAT it read 83 as atmospheric pressure, and only went down to like 82 or so when the car was started. That is way messed up. If yours is reading 100 at atmospheric and drops down below 45 or so at idle/vacuum it should be ok.
Quote:
Originally Posted by Prospero View Post
IMHO, Deatschwerks 600's are a great way to fly. Most users of the Adaptronic are running with them and it will make it easier for all of us to work together all having a common part.

The install was easy as pie, they were great quality, and I love the price!

Cheers,
Prospero
I really regret not getting the 600's
While I'm fairly happy with my 460's which cost me 150 after cleaning and everything, I know I'll probably want a little more than they will provide me in the future. Oh well, I guess I'll just have to find out what they max out at.
18psi is offline   Reply With Quote
Old 07-15-2009, 05:59 PM   #34
Junior Member
 
Join Date: May 2008
Location: Ft Walton Beach, FL
Posts: 48
Total Cats: 0
Default Deatschwerks 600's

Great, Ill buy those since Deatschwerks 600's. Thanks Stein! I had about the same issues with my adaptronic as 18PSI did. Hard to start, and when it finally started it was horrible, lke misfiring and very very rough. I followed all your instructions on how to isntall and wire the thing but I didnt have the wideband. So I ordered it today, hopefully that will help. Travis sent me a few hints and maps but had no success. I have to also say that this is my first ECU install. I can open a carb and modify it to do whatever I want. But Im new to this electronics stuff, so I need to go slow and some stupid questions will be online very soon....

But I cant be an expert without firts being a NOOBIE..
etnad is offline   Reply With Quote
Old 07-15-2009, 06:02 PM   #35
Senior Member
 
Join Date: Nov 2007
Location: somewhere in NoVA
Posts: 550
Total Cats: -1
Default

Glad you are going this route!

So you should add some detials to your profile, location, car, config, stuff like that. We like to know who we are picking on...

Cheers,
Prospero
Prospero is offline   Reply With Quote
Old 07-15-2009, 06:05 PM   #36
Elite Member
iTrader: (46)
 
Stein's Avatar
 
Join Date: Dec 2007
Location: Nebraska
Posts: 4,812
Total Cats: 138
Default

Quote:
Originally Posted by etnad View Post
Great, Ill buy those since Deatschwerks 600's. Thanks Stein!
Travis sells the Deatchwerks 600's. I'm guessing they are the same price from him or Deatschwerks.
Stein is offline   Reply With Quote
Old 07-24-2009, 01:40 AM   #37
Murderator
Thread Starter
iTrader: (76)
 
18psi's Avatar
 
Join Date: Apr 2008
Posts: 36,163
Total Cats: 2,573
Default

OK, question time.

-launch control
I set "cut fuel" and set the rev limit to 3k. Set "disable above" to 2kph. For "speed sensing" I have "use mvss" enabled and did the whole "read 50kph". When I drive with launch control enabled it for whatever reason still cuts at 3k. What am I doing wrong?

-idle
Also, at idle I have set it to around 14.0 afr. It idles great, but after a few minutes it will stumble for a little bit, then run good again. What gives?

-Btw
I have master trim at 15, not -15. Setting it to -15 Leans it out like a **** and it just goes bananas. With 15 it runs just fine.

-Also
Stein you mentioned my coolant temps were a bit high and to lower the engagement point of the fans. I CANNOT for the ******* life of me find that button/area. Where the **** do you adjust coolant temp of the fan engagement? I am such a
18psi is offline   Reply With Quote
Old 07-24-2009, 05:05 AM   #38
Senior Member
iTrader: (3)
 
Join Date: Aug 2005
Location: Gig Harbor, WA
Posts: 504
Total Cats: 3
Default

-launch control
dunno, haven't really messed with it. Although it sounds hideous if you do spark cut instead of fuel cut

-idle
My car does that some times. It will idle fine, then you pull up to a light and it tries to idle down to ~600rpm for bit and then it goes back to normal.

-Btw
I'm claiming injector dyslexia from swapping to many injectors in one day

-Fans
On the Aux out page look for an output labeled "Water Temp" change the low threshold.
TurboRoach is offline   Reply With Quote
Old 07-24-2009, 04:45 PM   #39
Murderator
Thread Starter
iTrader: (76)
 
18psi's Avatar
 
Join Date: Apr 2008
Posts: 36,163
Total Cats: 2,573
Default

Change it to what? I have it set at 50 for low and 120 for high the way it is now. My temps are anywhere between 108 and 117 when the car is fully warmed up. My dad says thats fine. What are your guys' temps?

Another problem I'm noticing is when I turn on the ac, the rpm start jumping around WILDLY and sometimes so much that it dies. I tried to tune the area it surges around in as per the manual, but the rpm jumps so much I can't even find what rpm I need to tune when pressing space bar to get to the cell. How do I get rid of this?

Also, I set up the idle to around 14.0 afr. Idles great. After a little driving I stop and the idle is jumpy again. Ok, so I adjust it again since now its ritch. idles nice again. Drive around. Idles nice. Drive around some more. Temp changes. Now its fucked up AGAIN....I'm guessing my coolant temp enrich is fuxorzed...HALP

Also, when I'm driving I have set everything to super rich, like 10-11's. It will pull weakly til it gets to like 3500 or so then JERK like its got a shitload of power, then after 4k or so it is SLOOOOOOOOOOOOOOOOOOOOOOOW...AFr's are set to 11 and stay there for the most part, boost stays at 8psi the whole time, but the car is slower than a ******* bicycle. What gives? Is it my timing?

Anyhow, I'd love some answers
18psi is offline   Reply With Quote
Old 07-24-2009, 04:49 PM   #40
Elite Member
iTrader: (46)
 
Stein's Avatar
 
Join Date: Dec 2007
Location: Nebraska
Posts: 4,812
Total Cats: 138
Default

Quote:
Originally Posted by 18psi View Post
Change it to what? I have it set at 50 for low and 120 for high the way it is now. My temps are anywhere between 108 and 117 when the car is fully warmed up.
That's the problem. Fans won't turn on until 120*C. Way too high. Mine is 93*C low which is just over thermostat temp and 100*C high, or 212*F.

No sense having 50* low, as once the fan turns on, it will never turn off. 120* high means your fan won't even turn ON until your water gets oto 248*F.
Stein is offline   Reply With Quote
 
 
Reply

Related Topics
Thread Thread Starter Forum Replies Last Post
Project Gemini - Turbo Civic on the Cheap Full_Tilt_Boogie Build Threads 57 07-19-2017 05:11 PM
Changing from 460cc to flow force 610cc in diypnp 90 Turbo MEGAsquirt 19 10-19-2015 04:23 PM
Time to start learning and play with tuning The Gleas MEGAsquirt 3 10-01-2015 10:30 AM
3rd Time's a Charm...hopefully. zephyrusaurai Meet and Greet 2 09-28-2015 11:59 PM
Got bad Crank Angle Sensor and Random Misfire codes, but runs fine. Greasyman General Miata Chat 2 09-28-2015 11:44 AM


Thread Tools Search this Thread
Search this Thread:

Advanced Search

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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -4. The time now is 10:36 PM.