MEGAsquirt A place to collectively sort out this megasquirt gizmo

Timing at idle - requesting some help

Thread Tools
 
Search this Thread
 
Old 04-29-2011, 01:24 PM
  #61  
I'm Miserable!
 
Techsalvager's Avatar
 
Join Date: Jun 2009
Location: albany, ga
Posts: 1,866
Total Cats: 0
Default

Originally Posted by Dem768
Given these settings, I should be able to get my car to idle under 1100.

I'm now running 16 degrees at 14.5 but sounds best at about 1100. I'm happy with is but maybe I'll try lower. Thanks for the input!
I altered how long my IAC intake piping was to the main piping.
Now it idles around 14.7 to 15 AFR's at the same settings
same timing, rpms, etc
Techsalvager is offline  
Old 04-29-2011, 02:12 PM
  #62  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Dem768
Hornet- when you said to disable the valve to set your mechanical idle. This made sense to me but my issue is:

1) What should our valve frequency be? If i use something in the 30's 8-9 DC is closed. If i use 62 18-19 DC is closed?

2) Once the valve is closed should I set it to the low, good sounding 950 idle? Even if this means that my car stalls when I rev it and let go? Or should I crank the idle until it can recover from falling RPM's?
You don't really disable the valve. What happens is that in warmup mode, the DC of the valve stays steady. That's important for tuning.

Answer to 1: I just used the MSPNP default for valve frequency. Haven't really messed with it but I suppose it makes sense that closed DC would be related to frequency.

Answer to 2: I've set mine for 900RPM at idle and I don't stall. If you are stalling after a throttle blip, then adjusted dashpot settings (done last) should be able to catch it. Note that the dashpot is only available in closed loop mode.
hornetball is offline  
Old 04-29-2011, 02:14 PM
  #63  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Techsalvager
I altered how long my IAC intake piping was to the main piping.
Now it idles around 14.7 to 15 AFR's at the same settings
same timing, rpms, etc
Makes sense. With that alteration, you should probably retune your idle starting with VE.
hornetball is offline  
Old 04-29-2011, 02:15 PM
  #64  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Braineack
No I know this and I have a background in Art.
After years of using and building MS, I think you have a bit more than only "a background in Art." LOL.
hornetball is offline  
Old 04-29-2011, 02:19 PM
  #65  
Boost Czar
iTrader: (62)
 
Braineack's Avatar
 
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,493
Total Cats: 4,080
Default

Im still no EE, but I'm learning. its really frustrating.

but i did come up with a clever idea to solve some heatsoaking issues using two AIT sensors...that gets implemented this weekend.
Braineack is offline  
Old 04-29-2011, 02:23 PM
  #66  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Braineack
Im still no EE, but I'm learning. its really frustrating.
Degrees are overrated. AFAIK, they are simply one indicator of aptitude -- and often not the best indicator.
hornetball is offline  
Old 04-29-2011, 02:31 PM
  #67  
Junior Member
 
minileprechaun's Avatar
 
Join Date: Oct 2008
Location: DC
Posts: 65
Total Cats: 0
Default

Originally Posted by hornetball
Moving on to the exciting world of "Closed Loop Settings."

To illustrate the relationship among the above, let's use the following values for an example (assuming a target idle RPM of 900):
Deadband Range = 50
Upper Limit = 250
Lower Limit = 75
Fast Recovery = 10 (0.1 seconds or 10 corrections per second)
Slow Recovery = 150 (a correction every 1.5 seconds)

Got all that? Good. Here are some recommendations for tuning closed loop:
3. Make the Slow Recovery slow. This gives the engine plenty of time to catch up to the DC change. Oscillations, such as those you are seeing, means that the DC is changing too fast for the engine to catch up.
you're really good with this tutorials! thanks, man.

funny thing, i got the dashpot part to work and now the car doesn't stall at lights or in neutral! of course it could use finer tuning.. but for now the car doesn't stall while driving (different numbers than what i posted earlier in this thread).

as for closed-loop .. that's going to need more work.
current car actions after ignition include rev to 1400rpm, then drop, drop drop and try to recover and stall. (i can blip the throttle myself to keep it alive) this rev/drop/revive is tuned via the deadband range, upper/lower limits, and fast/slow recovery... correct? tuning those numbers should get me to the same idle i had in warm-up only mode? (i followed the other directions with setting my slow idle speed at 900, fast idle at 1400, and my minimum DC at 11.. from the warm-up only mode)

do i care about the "other time constraints"? right now my startup delay (x0.1) is 255. does this mean that MS will wait 25.5 before using these closed-loop settings?

Fast Recovery = 10 (0.1 seconds or 10 corrections per second)
Slow Recovery = 150 (a correction every 1.5 seconds)
as for this ... Tunerstudio shows these two settings as
Fast Recovery (x0.1sec) _______
Slow Recover (x0.1sec) _______
i understand that 10x0.1 will come out to be 10 corrections per second
but how is 150x0.1 1 correction per 1.5 seconds?
minileprechaun is offline  
Old 04-29-2011, 02:43 PM
  #68  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by minileprechaun
as for closed-loop .. that's going to need more work.
current car actions after ignition include rev to 1400rpm, then drop, drop drop and try to recover and stall. (i can blip the throttle myself to keep it alive) this rev/drop/revive is tuned via the deadband range, upper/lower limits, and fast/slow recovery... correct? tuning those numbers should get me to the same idle i had in warm-up only mode? (i followed the other directions with setting my slow idle speed at 900, fast idle at 1400, and my minimum DC at 11.. from the warm-up only mode)

do i care about the "other time constraints"? right now my startup delay (x0.1) is 255. does this mean that MS will wait 25.5 before using these closed-loop settings?


as for this ... Tunerstudio shows these two settings as
Fast Recovery (x0.1sec) _______
Slow Recover (x0.1sec) _______
i understand that 10x0.1 will come out to be 10 corrections per second
but how is 150x0.1 1 correction per 1.5 seconds?
Yes, closed loop should achieve your target RPM. BTW, that target depends on your CLT. When you first start, it should be 1400 decreasing to 900 as you warmup. Remember that the closed loop is reactive.

In your case, a startup delay of 255 does mean that the MS will wait 25.5 seconds before starting the closed loop process of DC corrections. That seems excessive. Don't think you need more than 5 seconds for this (I'm using only 1 second).

Megatune time units are 0.01 seconds. In your case, Tunerstudio time units are 0.1 seconds, so a value of 1 = 10 times per second, a value of 10 = Once per second, a value of 150 = Once every 15 seconds, etc.
hornetball is offline  
Old 04-29-2011, 03:47 PM
  #69  
Junior Member
 
minileprechaun's Avatar
 
Join Date: Oct 2008
Location: DC
Posts: 65
Total Cats: 0
Default

Originally Posted by hornetball
Yes, closed loop should achieve your target RPM. BTW, that target depends on your CLT. When you first start, it should be 1400 decreasing to 900 as you warmup. Remember that the closed loop is reactive.

In your case, a startup delay of 255 does mean that the MS will wait 25.5 seconds before starting the closed loop process of DC corrections. That seems excessive. Don't think you need more than 5 seconds for this (I'm using only 1 second).

Megatune time units are 0.01 seconds. In your case, Tunerstudio time units are 0.1 seconds, so a value of 1 = 10 times per second, a value of 10 = Once per second, a value of 150 = Once every 15 seconds, etc.
alrighty .. must remember to decrease startup delay drastically!

looks like i confused myself with fast/slow recovery rates .. i'll have to adjust those accordingly.

thanks for helping clarify it all
minileprechaun is offline  
Old 04-29-2011, 05:05 PM
  #70  
Senior Member
Thread Starter
iTrader: (7)
 
miatauser123's Avatar
 
Join Date: Sep 2009
Location: Royal Oak Michigan
Posts: 754
Total Cats: 68
Default

My car starts and will shoot to about 3.5-4k immediately and then usually stall on the first start.. Cant seem to get to the bottom of it.. Take a look at my msq and let me know if you see anything odd...

The tune and the starting log are posted.. This is a warm start too.
Attached Files
File Type: msq
best tune.msq (34.9 KB, 258 views)
File Type: xls
sr.xls (40.8 KB, 102 views)
miatauser123 is offline  
Old 04-29-2011, 05:40 PM
  #71  
Boost Czar
iTrader: (62)
 
Braineack's Avatar
 
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,493
Total Cats: 4,080
Default

MS2 is different. but the rpm shoot up is due to the idle PWM cranking table.
Braineack is offline  
Old 04-29-2011, 09:30 PM
  #72  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Dem768
My car starts and will shoot to about 3.5-4k immediately and then usually stall on the first start.. Cant seem to get to the bottom of it.. Take a look at my msq and let me know if you see anything odd...

The tune and the starting log are posted.. This is a warm start too.
Log doesn't seem to match to msq. Are you sure they are a matched set? Startup delay, DC values, ignition advance values, etc. don't match.

Some general comments:

1. Your TPS value is always 0, yet there are lots of TPS thresholds set in the msq. Have you done a TPS calibration? Do you have a TPS?
2. Plot your fuel pressure. Looks really strange. Something wrong with your fuel pump/filter/pressure regulator?
3. In your idle control, I would expect the Cranking DC to be a bit higher than Minimum DC. In fact, it should be near the warmup mode lower temp DC (assuming you had one setup that gave a satisfactory fast idle speed). This "cracks" the throttle a bit for start.
4. Your Lower Limit is nice and tight, but your Deadband is loose. You do realize that at a warm idle speed of 1050, the MS will not change DC at all between 900RPM and 1200RPM with that loose Deadband, right?
5. The RPM spike has to come from air. There are only three paths for air on a 1.6 (excluding the throttle and vacuum leaks). They are the idle screw, the control valve and the air valve. See http://www.miata.net/garage/isc.html. Perhaps your air valve is faulty?
hornetball is offline  
Old 05-01-2011, 09:46 PM
  #73  
Senior Member
Thread Starter
iTrader: (7)
 
miatauser123's Avatar
 
Join Date: Sep 2009
Location: Royal Oak Michigan
Posts: 754
Total Cats: 68
Default

These are both current. Just now tried to cold start the car and it stalled..

One more thing. Where does MT get its Fuel Pressure reading from? Does my car even have a sensor? Should my stock fuel pressure regulator be connected to vacuum?
Attached Files
File Type: msq
best tune.msq (34.9 KB, 213 views)
File Type: xls
start up.xls (42.6 KB, 103 views)

Last edited by miatauser123; 05-02-2011 at 09:14 AM.
miatauser123 is offline  
Old 05-02-2011, 11:28 AM
  #74  
Junior Member
 
minileprechaun's Avatar
 
Join Date: Oct 2008
Location: DC
Posts: 65
Total Cats: 0
Default cold good, idle good, warm bad :(

is there a trick to warm start in closed-loop mode?

warm starting now has the same issues as i previously had with cold starting. ignition, RPM up to a good range, then RPM drops and the car can't keep alive and stalls.

in closed-loop mode, with a cold engine, i was able to adjust fast recovery to keep the car alive when the RPM drops after first ignition. it's not perfect, as it still revs and drops, revs and drops.. until the coolant warms up . .but it's not as bad as before. the rev and drop ranges from 850 to 1100. i thought i'd want it to be steady around 1400rpm until the engine warms up . .but that isn't happening.

i was able to adjust slow recovery at idle so the car sits nicely around 950rpms and only occasionally revs/drops (900-1000).

but with the car warm (coolant above my warm idle temp), the engine can't keep alive. if i adjust the fast/slow recovery to keep the engine alive during warm start, cold or idle settings are messed up and don't work properly.

sooo .. what's the trick?

if my cold idle is set to 1400rpm and warm idle is 900rpm .. should the engine hover around 1400rpm when warming up? if so .. i don't know how to do that.. ha.
minileprechaun is offline  
Old 05-02-2011, 11:39 AM
  #75  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by minileprechaun
is there a trick to warm start in closed-loop mode?

warm starting now has the same issues as i previously had with cold starting. ignition, RPM up to a good range, then RPM drops and the car can't keep alive and stalls.

in closed-loop mode, with a cold engine, i was able to adjust fast recovery to keep the car alive when the RPM drops after first ignition. it's not perfect, as it still revs and drops, revs and drops.. until the coolant warms up . .but it's not as bad as before. the rev and drop ranges from 850 to 1100. i thought i'd want it to be steady around 1400rpm until the engine warms up . .but that isn't happening.

i was able to adjust slow recovery at idle so the car sits nicely around 950rpms and only occasionally revs/drops (900-1000).

but with the car warm (coolant above my warm idle temp), the engine can't keep alive. if i adjust the fast/slow recovery to keep the engine alive during warm start, cold or idle settings are messed up and don't work properly.

sooo .. what's the trick?

if my cold idle is set to 1400rpm and warm idle is 900rpm .. should the engine hover around 1400rpm when warming up? if so .. i don't know how to do that.. ha.
Post MSQ.
hornetball is offline  
Old 05-02-2011, 11:58 AM
  #76  
Senior Member
Thread Starter
iTrader: (7)
 
miatauser123's Avatar
 
Join Date: Sep 2009
Location: Royal Oak Michigan
Posts: 754
Total Cats: 68
Default

Hornet - any ideas from looking at my last log and msq? the previous one might not have been matched.
miatauser123 is offline  
Old 05-02-2011, 12:07 PM
  #77  
Elite Member
iTrader: (4)
 
hornetball's Avatar
 
Join Date: Mar 2008
Location: Granbury, TX
Posts: 6,301
Total Cats: 696
Default

Originally Posted by Dem768
Hornet - any ideas from looking at my last log and msq? the previous one might not have been matched.
Haven't had a chance yet. Your stock fuel pressure regulator does need to be connected to manifold pressure.
hornetball is offline  
Old 05-02-2011, 12:19 PM
  #78  
Senior Member
Thread Starter
iTrader: (7)
 
miatauser123's Avatar
 
Join Date: Sep 2009
Location: Royal Oak Michigan
Posts: 754
Total Cats: 68
Default

... Mmm but what if it is? Or was? Does it conflict?
miatauser123 is offline  
Old 05-02-2011, 12:24 PM
  #79  
Boost Czar
iTrader: (62)
 
Braineack's Avatar
 
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,493
Total Cats: 4,080
Default

Originally Posted by Dem768
Where does MT get its Fuel Pressure reading from? Does my car even have a sensor?
It doesn't, and no.

Should my stock fuel pressure regulator be connected to vacuum?
Of course. The FPR is very important in fueling...it dials back the pressure at idle so it's easier to idle larger injectors. and it increases pressure in a 1:1 ratio in boost so you have more fueling capacity.

Without the FPR attached to vacuum/boost, you're running around 50psi of rail pressure at all times. This means you need to really dial back your VE table at idle, so far that you might be asking the injectors to cycle faster than the deadtimes allow. And in boost the fuel pressure will not increase, so as the boost increases, the fueling capacity of your injectors decrease in a 1:1 ratio.
Braineack is offline  
Old 05-02-2011, 12:27 PM
  #80  
Senior Member
Thread Starter
iTrader: (7)
 
miatauser123's Avatar
 
Join Date: Sep 2009
Location: Royal Oak Michigan
Posts: 754
Total Cats: 68
Default

Originally Posted by Braineack
It doesn't, and no.



Of course. The FPR is very important in fueling...it dials back the pressure at idle so it's easier to idle larger injectors. and it increases pressure in a 1:1 ratio in boost so you have more fueling capacity.

Without the FPR attached to vacuum/boost, you're running around 50psi of rail pressure at all times. This means you need to really dial back your VE table at idle, so far that you might be asking the injectors to cycle faster than the deadtimes allow. And in boost the fuel pressure will not increase, so as the boost increases, the fueling capacity of your injectors decrease in a 1:1 ratio.
I understand how a fuel pressure regulator functions. I have the Jackson FPR and then realized the stock regulator was still connected. Could this cause problems?
miatauser123 is offline  


Quick Reply: Timing at idle - requesting some help



All times are GMT -4. The time now is 01:45 PM.