The Top Mitsubishi Galant VR-4 Resource

Join the best E39A 1991-1992 Mitsubishi Galant VR-4 community and document your GVR4 journey.

  • Software Upgraded - Reset Your Password to Login
    In order to log in after the forum software change, you need to reset your password. If you don't have access to the email address you used to register your GVR4.org account, you won't be able to reset your password. In that case, follow the instructions here to regain access to the forum.

Hesitation at high RPM's - all gears. w/logs. (RESOLVED)

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
I've been chasing what I initially thought might be blowout at WOT. I was seeing a little bit of knock on some of my pulls, but it never seemed like something that would cause the sort of shuddering I was feeling at-speed. Well I was doing some long logs for SD VE table calibration and noticed something really odd.

My AFRatioEst will spike lean intermittently pretty much throughout my logs. Seems to happen mostly over 3000 RPM's. I got to wondering if it's not related to WOT at all, so I just log myself driving nice and easy up to high RPM's. Sure enough, I started getting the same hesitation with low load at high RPM's. It seems perfectly replicable, in fact.

For example, here's a screenshot from one of the logs I'll link to at the end. This is at a relatively low RPM, but it illustrates pretty clearly what I'm seeing.



I got to wondering what else I could find that might line up with these lean AFRatioEst spikes I was seeing. O2 sensor output looked fine (although maybe someone might see something in my logs?). WB AFR's looked ok (which tells me I wasn't getting blowout or misfire since I would expect a rich condition due to unburned fuel). Then I looked at the fuel flow and injector metrics and bingo!

Here's what it looks like with "fuel flow", "InjDuty", and "InjOn" added. You can see the little dips.



My guess is that at lower RPM's, these slight variations don't really translate into and noticeable hesitation. Now get the car up to higher RPM's...



These spikes I could absolutely feel. For comparison, here's what the fuel inputs look like just a 10ms earlier (off-spike):



I got to looking back at other pulls. In some cases, they look clean:



On others, I see it and feel it:



Now I'm scratching my head... Whatever these lean spikes in AFRatioEst are, they cause a noticeable hesitation. And I'm not sure if the correlating drops in injector flow are a cause, or a symptom.

Logs are here. I'm posting this to the ECMLink forums as well, but I'll take help from anywhere I can find it. Thanks, all!

click
 
Last edited:

swe_gvr4_1991

Well-known member
Joined
Jul 27, 2006
Messages
177
Location
Sweden, Alingsås
At first glance it seems like your CAS is giving out a bogus signal.
But before we go there, can you overlay the VSS data too?

Do these only:
fuel
rpm
vss
afr
 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
I think this is what you are asking for...

 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
Late last night, I ran across an old (2014) thread on Tuners with the following reference:



Thread is here: click

This took me to an even older thread in this forum, in which an admin suggests doing exactly what you've suggested (although he suggested 30%, not 27%, but that seems like a tuning thing once you identify this as a proper solution)

Link forum thread: click

So this morning I gave it a shot. The initial results seem VERY positive. I'm going to be checking on my MAP sensor (a GM 3-bar which is a little over a year old) to make sure the connections and boost/vac sources are good, but it may just be that these sensors are a bit "bouncy". I mean the variance is very small, but at least mine tends to oscillate a bit. That it has such a large, measurable impact is pretty impressive.

At the root of this problem is my understanding of the values I'm seeing in link when I'm having a problem. Which value(s) is showing the actual issue, and which value(s) are just symptoms? Since I have SD and no MAF, I have to assume any airflow values are calculated, not actual (sensor-based). My biggest question mark was the fuel flow, but I assumed that value was also calculated since link really only knows what I tell it about the injectors, then assumes the rest (right?).

Frankly, I don't think I would have ever guessed the oscillation of the MAP value to cause such a persistent, yet intermittent issue. Of course, this is all preliminary. I'll do some logging later on. I actually did log this morning while I was driving, and things looked good, but until I can go a farther distance, I consider my results anecdotal.

Fingers crossed and more to come...
 

swe_gvr4_1991

Well-known member
Joined
Jul 27, 2006
Messages
177
Location
Sweden, Alingsås
Your VSS seems to be ok. But you can see the impact on the speed item in the logs when the hesitation occurs. This also indicates a properly working VSS.

Those threads seem to hold good info.

The MAP sensor itself is not bouncy. It enclosed in a decopuled unit that is designed to withstand impact. There is only one way to get a bouncy signal and that is from the air pressure.
How is you sensor mounted? The nipple should be facing downwards. This is to prevent moisture to build up. If you have moisture in the nipple and the chamber behind it you would get a bouncy signal since the air is compressible and the fluid is not.
 
Last edited:

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
Actually, no. My MAP input isn't facing down. Interesting thing to check. As for where it connects to the manifold, I'll have to go look.

Is there anything else to be cautious of? I'm guessing clean power isn't a bad thing, but in this case, I believe the only thing I can really check is the ground.
 
Last edited:

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
I took a looksie today and my MAP is T'ed off a manifold barb that also feeds the FPR and boost gauge. There's a closer barb that goes straight to the BOV. Would you think that I would benefit from moving the MAP input to the closer barb? It would certainly reduce the overall amount of vac line that sits between and connects-to the MAP.

Quoting swe_gvr4_1991:
Your VSS seems to be ok. But you can see the impact on the speed item in the logs when the hesitation occurs. This also indicates a properly working VSS.

Those threads seem to hold good info.

The MAP sensor itself is not bouncy. It enclosed in a decopuled unit that is designed to withstand impact. There is only one way to get a bouncy signal and that is from the air pressure.
How is you sensor mounted? The nipple should be facing downwards. This is to prevent moisture to build up. If you have moisture in the nipple and the chamber behind it you would get a bouncy signal since the air is compressible and the fluid is not.

 

Mdlimy

Well-known member
Joined
Apr 30, 2015
Messages
111
Location
Gresham, oregon
I always hook up my map sensors with one dedicated line directly into the intake manifold. Ill tee all day long with the other parts but never the map sensor. If your line coming off the manifold before the tee is small enough you might have an issue.
 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
Easy enough to execute that little change. I'll just swap the MAP line with the BOV line and it will have its own connection.

I've done a ton of logging over the last few days. Results are very positive and I'll be putting together a large post this evening about my process and results.
 

Mdlimy

Well-known member
Joined
Apr 30, 2015
Messages
111
Location
Gresham, oregon
You sure are digging in deep with the logs, its good to see, alot of people don't understand how valuable logging is! I look forward to meeting you and the other few local pdx guys! I am planning on taking my car to hms in sherwood sometime within the next month to get on the dyno and do some tuning with the new setup, maybe we could get a few of us together. I think Evans car is damn close to driving again...
 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
That would be awesome. hell, I could use some dyno time as well. And yeah, I think his car is days from running around.
 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
And I think I've got it. For reference, I started this with an AFRatioEst that could be described as:

bumpy
erratic
spikey
(come find this thread, Google spiders...)

At cruise, there would be noticeable erratic spikes in the log. I didn't feel them much, but they had an impact. More on that later...



More troubling, at-or-above about 5k RPM's, I would feel a stumbling hesitation. In these cases, the spikes were far more pronounced and more frequent...



As you can see, in both cases, the spikes coincide with drops in the FuelFlow value. In the case of high RPM's, the drop is fairly dramatic, and I could certainly feel it in the car.

A found two very useful links during my search for answers. The first was the link on Tuners, which led me to the second link on the ECMLink forums. Both links can be found above.

Dave at ECMLink suggested that the issue might be with a fluctuating MAP reading. And true enough, while difficult to see in the log above, a detail of just the MAP and the raw voltage (displayed with a voltage range of 1-1.5v for clarity) shows a fairly significant fluctuation.



The recommended fix is to enable airflow smoothing w/SD operation, something that is typically disabled for SD setups. In this case, you uncheck the box in the LIVE configuration.



Next, you go into your direct access configuration and set your smoothing percentages to around 30%.



You can play with this, but Dave recommended 30, and after trying lower and higher values, 30 seems to be the lowest value that still offers appropriate smoothing. For example:

20% smoothing had hardly any spikes, but the AFRatioEst variance was still fairly significant.



30% smoothing had no spikes and the AFRatioEst variance was lower.



Now drivability is MUCH better, with the car responding much better to throttle inputs. There are two things that I'm not working out that seem to be a result of the smoothing I did.

1. My idle leaned out. I'm going to try setting the idle smoothing in the DA configuration to its lowest value since I never had any issues at idle before making changes. We'll see if this makes a difference.

2. I'm about 1.0:1 richer at WOT than I was before making the change.

Before (with ugly AFRatioEst spikes and dips in FuelFlow):



After (with nice, smooth Fuelflow, but much richer AFR's)



At this point, I'm working on the VE table to get WOT looking right. I'm also going to connect my MAP directly to the manifold with it's own barb (it shares a line with the BoV at the moment) and point it (the MAP) downward as it's been suggested that if the barb is horizontal or upright, crap can come through the line and muck up the sensor.

I'll post any updates that are relevant. For now, this issue would seem to be resolved!
 
Last edited:

GSTwithPSI

Well-known member
Joined
Jan 1, 2012
Messages
3,460
Location
SoCal
Glad you figured it out, and thanks for updating this with a detailed explanation of how you addressed the problem.

You're WB and AFR's look almost perfect in that last screenshot. Keep in mind, it will be pretty hard to get them to line up absolutely perfect through the entire power band. I typically leave mine on the rich side for safety. I think I'd pull a tiny bit of fuel and call it good.

I've said it before, but this is the kind of sh*t you'll only figure out tuning your own car. Unless your tuner lives with you, there's no way these kind of intermittent problems will get worked out in a few hours on some shop's dyno. Learning how to tune your own stuff is invaluable.

Now, let's figure out a way to bump your timing up /ubbthreads/images/graemlins/wink.gif
 

swe_gvr4_1991

Well-known member
Joined
Jul 27, 2006
Messages
177
Location
Sweden, Alingsås
Good to hear that it got resolved.
Keep up the good work and thanks for posting your efforts.
 

EfiniX

Well-known member
Joined
Oct 18, 2012
Messages
647
Location
portland, or
This community and people like you have been material to this process. I said as much in another thread, where I talked about the tuning I had done at English Racing. Some people took my meaning as a negative, but really, unless I leave my car with someone for a few weeks, the best I'll get on a dyno is a good starting point. Base fuel and timing maps that won't blow my motor up. There's been a lot of "had I only known" moments.

Certainly the best advice was also the first advice: unless you are staying stock, get engine management and a wideband. The idea that I could have accomplished anything with the keydiver chip and narrowband... I mean the people at keydiver do great work, but once the chip goes in, your tuning is over.

Quoting GSTwithPSI:
Glad you figured it out, and thanks for updating this with a detailed explanation of how you addressed the problem.

You're WB and AFR's look almost perfect in that last screenshot. Keep in mind, it will be pretty hard to get them to line up absolutely perfect through the entire power band. I typically leave mine on the rich side for safety. I think I'd pull a tiny bit of fuel and call it good.

I've said it before, but this is the kind of sh*t you'll only figure out tuning your own car. Unless your tuner lives with you, there's no way these kind of intermittent problems will get worked out in a few hours on some shop's dyno. Learning how to tune your own stuff is invaluable.

Now, let's figure out a way to bump your timing up /ubbthreads/images/graemlins/wink.gif

 
Support Vendors who Support the GVR-4 Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications RixRacing RockAuto RTM Racing STM Tuned

Recent Forum Posts

Top