Stumbling on light throttle

General support questions and announcements for MS3. See also MS3 manuals.

Moderators: jsmcortina, muythaibxr

Post Reply
michaeldch
MS/Extra Newbie
Posts: 20
Joined: Sat Jan 01, 2005 1:31 pm

Stumbling on light throttle

Post by michaeldch »

Hi:

I have a 1985 Alfa Romeo GTV6 running on Megasquirt PCB 3.57, MegaSquirt-III daughterboard kit and the MegaSquirt-III MS3X Expansion Card. Timing input is via a crank position sensor (VR input, stock 60-2 crank position sensor) and a stock cam position sensor. The car is now running full sequential spark and injectors. I'm using the MS3X to drive the injectors and fire the plugs via a Toyota sequential igniter to drive my stock coil packs.
I used a 1JZ/2JZ 6-channel igniter, Toyota part number 89621-22030 as documented here: http://forums.hybridz.org/index.php/top ... -this-pic/
The engine is a transplant from a 1995 Alfa 164LS and is a 24 valve (4 cam) 3 liter, normally aspirated.

The engine runs fairy well - feels very responsive and powerful but I have this one persistent problem which I have not been able to solve. I thought I'd solved the issue years ago...
http://www.msextra.com/forums/viewtopic ... ch#p278392
...but I realized that the changes I had made improved things a bit but never truly solved the issue. The problem is that under light throttle, the engine feels like its misfiring. As example, I'm driving with the engine not under any real load, coasting at 2500 rpms I feel some stumbling and misfiring. When I'm at WOT it goes extremely well, idles well etc.
I actually installed the MS3X in order to switch to sequential injection and spark to solve the stumbling issue but it seems not to have made a difference. I have attached an msq and logs - all and any help would be greatly appreciated!

Regards,
Michael
2015-04-12_11.31.36.msq
2015-04-12_11.14.06_Trimmed.msl
Michael
1985 Alfa Romeo GTV6.
T3Bunny
Super MS/Extra'er
Posts: 1391
Joined: Sun Nov 28, 2004 8:44 pm
Location: Miami, Florida

Re: Stumbling on light throttle

Post by T3Bunny »

On my tablet so not able to download your files ATM, but curious and likely will soon as I get a chance. Let's take a good stab or two at this though.

My first inclination is as others stated in your prior thread. It is lean surging. But you stated enriching the mixture already with no luck. And your leaning towards timing. So here is how I would approach either fuel issues or timing ones.

First start by looking closely at your maps. Are they nice and even transitions and very smooth in 3d mode? Fuel, timing? If so, look at accel enrichments. If your currently using tps rate of change/voltage, switch to map based and see if that helps. Approach this with just one plan of attack at a time.

Second bleeding over from first. Get a driver and a fast enough laptop. Preferably a good sized screen. Show said driver how to cause the issue. Most likely you need to be out on the highway. After driver can recreate the issue, pull up timing in 2d first. Have him/her drive in and out of the problem to create it. Watch where the little ball goes. Watch for more than a few seconds. Look at the bins it is cruising in. Now it sounds like you in the "tip in" zone. Where your coming out of cruise zone with very slight pedal pressure. This is a hard area to tune. My old Fiat was running stupid high compression (14 to 1 on street gas) and this was my problem area.

Third, if you suspect "rough maps" in this area, grab and select a bunch of appropriate bins and set timing to one thing. If issue changes for better or worse, your on the right track. If timing doesn't do it, then go look at the VE maps in the same way. 2d maps are usually easier to follow for this, but switch to 3d too and see if it shows you anything else.

Most importantly is that your transitions need to be smooth. If your AE blinks on and off in this area, it will cause issues. If decell does, same. If big jumps on the map, VE or Timing, same. Approach each piece one at a time, make changes on the fly and test. I can't stress enough though that to really see a problem like this, the car needs to be being driven while you watch and feel.
Forget the porn, my bathroom is full of car manuals...
1977 Rabbit (retired)
1991 Cabriolet (retired)
michaeldch
MS/Extra Newbie
Posts: 20
Joined: Sat Jan 01, 2005 1:31 pm

Re: Stumbling on light throttle

Post by michaeldch »

Thanks T3Bunny - I'm working through your recommendations :-)
Michael
1985 Alfa Romeo GTV6.
T3Bunny
Super MS/Extra'er
Posts: 1391
Joined: Sun Nov 28, 2004 8:44 pm
Location: Miami, Florida

Re: Stumbling on light throttle

Post by T3Bunny »

Mmmmm... Okay not digging very closely through your settings, but just looking at your VE, SPK, and AFR target tables tells me a LOT. First off turn off EGO correction, even if just to test. What I see in the logs, is an engine hunting and correcting the fuel up and down.

What I see in your maps... This is an engine that has been tuned and running on these maps for a while? If so your throwing out gobs of power, and fuel. Your maps are tunable in three dimensions, your stuck in 2d mode. With a N/A motor (I would hope this one is!) you can be much more aggressive with timing. Make a mountain, and it really doesn't have to slope down to far up at the top. Very few N/A motors need to see 12.5 up at the top there either.

Urgh... Okay so trying to explain this all would take a couple hours of typing. But wait, I am sure I have done this lesson before so lets go here: http://www.diyautoftw.com/showthread.ph ... By-T3Bunny Scroll down to January 01, 2010. This is a build thread (mostly) on my rabbit with some other interesting stories thrown in for good measure. A few days before that and a few post after are also relevant. But that date has some detailed information on map building. This is grabbed from my original thread, so sometimes I am answering questions that were not grabbed as well...

I REALLY want to recommend that you buy and read this book! Chapter 12 is the one to pay especially close attention to. Read it through at least 2-3 times! (BTW this book is by Jerry Hoffman And Matt Cramer, both somewhat of legends in this community.)

But I "suspect" that the online manuals should have some really good information on map building and what to look for.

Also, I would look around and try to collect some more maps for the GTV6. There has to be a couple out there with at least an almost decent spark map. VE tables don't matter too much as I explain in the above post. AFR target tables tend to look almost the same for every engine. With them tailored for where the engine likes to cruise, rpm range, and pa range. BUT I CAN"T STRESS ENOUGH that most maps you find posted up will be pretty rough at best. Read the above suggestions so you can get an idea of what you SHOULD be seeing. Also look up my recent posts here, I posted up my .msq a couple days ago. While it has many many complicated things going on, ignore most of them and just look at the maps and AFR targets.

BTW, I personally never run EGO correction. Silly cause I should I guess, but I tend to dial my personal maps in right where I want them. At the very least I almost always turn EGO correction off during extensive VE tuning. And VE Tuning without properly setup target AFR tables is a bad idea. In some other ways I am also a bit old school. I rarely use "Automatic" tuning. ESPECIALLY on a rough car, I would rather drive it then process datalogs watching what they change. Let me throw out a few more thoughts.

Process:
1) Get a safe ignition map built, or found. Safe is fine, power can come later.
2) Build the rest of your maps, focus on bin spacing. Get it clean. Pay extra extra attention to the AFR target MAP.
3) Get car running "okay" and verify base timing. You want to make sure if the MS thinks it is at 10 degrees, it is at 10 degrees.

The last step can just as easily be done first. It most likely will need to be checked again once the very basics are setup.

5) Get car to idle nicely. This is a dance between VE and timing and idle bypass air and a bunch of other voodoo. It won't be perfect for most until much later on.
6) After hot idle is "okay, then make use of warmups to dial those in better. Again these will have to be fiddled more once VE tables are tuned. At first just focus on it warming up well enough
6) Go tune VE table. If using multiple tables, LOCK THEM OUT and only tune one at a time. Later the tuned base table can be used for a start on another one. (same for spark tables)
5) Fix any issues that rear up in the test and tune drives
6) Once VE tables are decent, you can start the search for more power and work on spark tables. Kinda a black art here... Remember to retune VE tables as needed if large changes are made on spark. And you will want to dial the VE back in when done with spark anyways.
7) Rinse and repeat most of the above steps, multiple times. As you go through and dial in some things, others will fall out of place. Save your progress, make use of the note section in Tuner Studio.

All of the above has helped me tune more cars than I can count. And easily 3/4 of them were boosted too. There is a LOT to digest and study. But once the pieces start falling into place, the sense of accomplishment is like nothing else! I tried to make a living doing this at one time, but I enjoy that challenge and accomplishment so much I gave away way too much for way too little!
Forget the porn, my bathroom is full of car manuals...
1977 Rabbit (retired)
1991 Cabriolet (retired)
Post Reply