Megasquirt Forum Area is for discussing Megasquirt EMS

Megasquirt typical cranking pulse width

Thread Tools
 
Search this Thread
 
Old 04-20-14, 03:10 PM
  #1  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
typical cranking pulse width

I'm starting to now experience really long cranking times when the engine is cold. Seems weird that this is getting worse as the outside temp is going up now too. I have been fooling around with the priming pulse and cranking percent, but I don't really feel like that's getting me anywhere.

My current typical cranking pulse width is arpund 11ms. Seems kind of large to me, so maybe that's one issue.

Also, it recently started this thing where now it wants to default to a locked-on value for the idle valve %PW too. I had it set pretty good before with closed loop idle control, and the idle valve would go closed over 2000 rpm. Now it choses a ramdom time to pick an arbitrarily high idle pwm and lock on to that. So I'll be cruising along, then once I come to a stop it will idle at around 1500 rpm. I'll check the idle pwm and it says it's in the 70s, when it should be around the single digits, if that. No matter what i do, it will stay at that value and rpm. So even once I start driving and get the revs up, it will leave the valve open. I've had no problem for 2 years now, and this starts.

MS2, V3
S4 turbo
Stock ports
T04B hybrid
550/720 injectors
Old 04-23-14, 11:25 AM
  #2  
Full Member

 
camman's Avatar
 
Join Date: Dec 2005
Location: utah
Posts: 189
Likes: 0
Received 0 Likes on 0 Posts
^what he said. having similar issues with idle
Old 04-23-14, 06:08 PM
  #3  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
You wouldnt happen to be using the bluetooth adapter from efi analytics would you?
Old 04-24-14, 06:00 PM
  #4  
Full Member

 
camman's Avatar
 
Join Date: Dec 2005
Location: utah
Posts: 189
Likes: 0
Received 0 Likes on 0 Posts
nope, no blue tooth
Old 04-26-14, 07:46 PM
  #5  
Engine, Not Motor

iTrader: (1)
 
Aaron Cake's Avatar
 
Join Date: Feb 2001
Location: London, Ontario, Canada
Posts: 29,789
Likes: 0
Received 108 Likes on 91 Posts
Can you post an MSQ so we can see your idle PID settings?

As for your cranking PW issue, 11MS sounds about right to me actually. That's typically around what I see. Takes a butt load of fuel to get these engines started reliably.

But are you saying that you have head to steadily increase cranking fuel amount to get the car started?

Changed your plugs recently.
Old 04-26-14, 09:40 PM
  #6  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
Ehh. About 2 years ago they were brand new. Guess I should at least check them now

And I don't really know if I can pull and msq any more. I'll have to get another usb-serial adapter because I use a tablet with a Bluetooth adapter to tune these days.
Old 04-27-14, 05:50 PM
  #7  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
Yep. They're pretty well fouled and worn down. I guess the boost really eats plugs :/ Not a whole lot of miles on them, but I'll replace them and see how things work out.

Now, about that cranking pulse width... your car is a monstrously ported high compression turbo, right? If I recall project tina, it's bridge ported, isn't it? So I would suspect yours would require a lot more fuel to get it started when compared to my stock ports.

Now off to find that adapter cable to pull an msq.
Old 04-27-14, 06:18 PM
  #8  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
Found the cable AND the computer's charging cable!!!

Edit: I just remembered I replaced the idle valve transistor when I built the ECU because I installed the diode backwards, thus shorting out the transistor. I didn't use the recommended transistor though. I had instant access to a TIP 120. Muythaibxr said it should be ok to use, but I wonder if now that transistor is causing unstable operation?

.msq file attached
Attached Files
File Type: zip
2014-04-27_18.56.00.zip (11.2 KB, 33 views)
Old 05-03-14, 09:53 AM
  #9  
Engine, Not Motor

iTrader: (1)
 
Aaron Cake's Avatar
 
Join Date: Feb 2001
Location: London, Ontario, Canada
Posts: 29,789
Likes: 0
Received 108 Likes on 91 Posts
A TIP120 should be fine. Really, the exact transistor is not important as long as the base current gain is about the same and that the C - E can handle the required current.

Now I've looked at your MSQ and, a lot of settings are a little off. Not unworkable, just odd.

Your Req_Fuel is about half what it should be. Not that it will cause problems, but will cause very high VE table values and you may in fact hit max on the VE table. I see you're already pretty close. Of course changing it now will require you to rescale the VE table. Should be somewhere between 8 and 10, depending on injector size.

The AFR table is very rich below atmospheric. Most of those can be 14.7s except during cruise and once you get past about 5000 RPM. Taper down to about 13.8 in the "power" areas and then make sure you are around 12.5 just as you enter boost. If that is the table you are auto tuning too, the car will end up excessively rich.

You also have seemed to disable EGO correction by setting the minimum RPM to 9000 RPM. The PID settings are very weird. Typically you'd have a larger I, a few G and no D. Though I have found a nicely adjusting bit of settings to be about 30 G and 5 I with no D. TPS setting is odd as well...the TPS should not be reading 120% unless it's not calibrated.

I suspect that if your TPS is reading above 100%, that can be causing your idle PID lockout.

Cranking RPM seems low to be. Better settings are around 300RPM. That may be why you are needing large cranking PW. The 'Squirt is considering the engine started way too early. A bit more ASE would help the situation as well.

Getting to the closed loop idle settings, the idle activation RPM adder should be lower. 300 RPM or so. At 500 RPM it may never activate PID, causing stuck idle valve syndrome.

The TPS threshold seems weird as well. Should be in the 1 - 3% range for a properly calibrated TPS. The dashpot adder is VERY high. I mean, very, very high. Should be in the 2 - 5 range depending on how lazy the BAC is. All the dashpot does is give the idle a little bit of a bump as the throttle snaps closed. 20% would bounce it up about 1000 RPM or more.

Decrease the crank to run taper...5 seconds is way too long and will cause the engine to be unstable for 5 seconds after starting before PID kicks in. You won't get much of a cold high idle. 1 second is typically what I use. The control interval also should run around 200 mS. 50 will cause it to be rather spastic. Your PID controller gains seem fine for the most part but without some G it may never hit your target.

PID lockout RPM dot seems low. Try 600.

At first glance, that's what I've noticed.

You may want to check out the MS3 (yep, MS3) idle documentation:
Idle Control

Also my MS2 base map settings are here, updated for the latest firmware:
How To Megasquirt Your 2nd Gen RX-7: Programming The ECU (MS2)
Old 05-03-14, 12:14 PM
  #10  
Trunk Ornament

Thread Starter
iTrader: (11)
 
AGreen's Avatar
 
Join Date: Sep 2006
Location: Goose Creek, SC
Posts: 3,054
Likes: 0
Received 2 Likes on 2 Posts
I've read all that before. Most of what you said I agree with. I recently dropped the req fuel down to 4 I think and got it to run a whole hell of a lot better since then. I guess the definition of having a much lower req fuel helped me a lot. I did do some auto tuning, but I have since gone to using a tablet, so there is no more auto tune feature as far as I know on msdroid. So I have not touched the Afr table in some time.

I dabbled a bit with ego correction, and may have left it in a weird setting before completely abandoning it. It was causing a pretty bad surge while cruising, so I disabled it entirely. I just haven't gotten back to playing with it yet.

I made sure the tps does not ever go above 100%. I scaled the full open value down to where it reads about 99%. I am using an S4 tps, modified for full range of throttle travel. Works great.

I lowered cranking pulse width from 500 to where it is now in about 50 rpm increments and this one seems to work best. Kind of a lesser evil because it still sucks.

Before you got back to me, I re-read the msextra manual and already turned the idle activation adder down to 250. It's idling better, and I'm getting closer to a more consistent PID code activation, but sometimes it still won't pick up the code and the engine will die. Say I'm coasting in gear, then push in the clutch. It goes in to an idle (super high by the way, about 1500 rpm... which I suspect is from the 20% dashpot adder I had in there before). So it eventually gets down to a stable-ish 750 rpm, then if I let the clutch out for a second and rpm goes above the idle activation adder, and I immediately push the clutch back in, it won't pick up the idle pid and will die with idle pwm% at zero.

I'll try changing the tps threshold and dashpot adder, as well as the crank to run taper. I also have already changed the pic control interval to 200. I've been tinkering with the pic settings a lot and that was one that really didn't seem to work. I have added a little more I and P. And I'll try 600 for the PID lockout.

I really appreciate your help on this. I recently got the AC working again, now idle control is knocking on my door. Other than the idle issue and stupid long start, it runs, drives, and idles fantastically, so long as I don't kick on the AC.
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
Turblown
Vendor Classifieds
12
10-17-20 03:25 PM
dona1326cosprings7
1st Generation Specific (1979-1985)
4
10-29-15 06:47 AM
Aramir
New Member RX-7 Technical
24
10-18-15 02:39 AM
Robrods64
New Member RX-7 Technical
6
10-02-15 07:56 PM



Quick Reply: Megasquirt typical cranking pulse width



All times are GMT -5. The time now is 06:09 PM.