When I turn on my A/C my car starts surging lightly going down the road and didn't do so before the EBL install. Anyone have this problem? Something I can do to fix it? San Antonio is crazy hot and would really really like my A/C working for the rest of the summer.

 

Is the surging in a steady rythmic pattern? Or maybe during lean cruise mode? If during lean cruise it may be that the AFR is too lean.

If it is during closed loop and a steady pattern, then 99% sure it is too much proportional gain. Which is actually easy to fix. The table to change is the:

PRP - Gain Multiplier vs Airflow

Only need to get the airflow value from the data log. Start by doing a 'View Log' from the Analysis screen. Then find that area where it surges. If you need obtain a log, just mark the area with a couple of space bar taps. Then do the 'Find Mark.'


The data log column named 'Gms Sec' is the airflow value used for the table lookup. Lower those areas in and around the data logged airflow value.


Further proportional gains tips:

A sign of not enough proportional gain is the INT rising and falling in a regular pattern. What is happening is that the ECM is using the integrator to create crosscounts.

In this case will need to increase the above table in those gms/sec areas.

The trick with proportional gain is to have enough to keep a steady INT. But not so much that the engine is surging.

The purpose of proportional gain is to create crosscounts.

Proportional gain is affected by injector flow. Larger injectors will increase the gain.

RBob.

 

I still haven't got into lean cruise mode, I'm not going to turn it on until I've got the WB in and the tuning down with that.

Your second guess is exactly it. I'll get a log tonight with the A/C running and hopefully I can figure it out.

 

Tier.........mine is doing the exact opposite. When the a/c is on it is smooth. Turn it off when the engine is hot and there comes the surging. Turn the a/c back on and the surging stops. I have turned the prop gains down as suggested before by 50% but it still has issues when at top operating temperature and a/c off. It does not surge when cold and warming up.

I'll have to try Rbob's above procedure.

DM

 

Not real sure how much proportional gain to take out and where exactly to do it, I don't know what indicates surging in my log file.

 

In this case set up to data log and cruise to duplicate the problem.

Once it starts surging hit the space bar twice, cruise with the issue for a minute or so, then hit the space bar again (maybe 3 times). Can make them quick taps so they are close together.

Now do a Display->Analysis, View File (button), and use the Find Mark button. The area of surging will be between the two groups of green highlighted records.

RBob.

 

If possible check out the attachment and see if I am understanding correctly? I am still running open loop with a WB. Spark is sitting still so I'm hoping reducing the 16 to 32 areas will help.

Thanks.......... DM

 

While in open loop the proportional gain is inactive. So this surge has nothing to do with the proportional gains.

The shot of the Analysis screen shows that the idle is going lean and surging. The lean AFR will cause the surge, so that is what I would work on.

RBob.

 

Thanks RBob. After reviewing it seems as the VEL always leans it out there. ?? Anyway I added some fuel back in those cells and will try it.

 

I figured out how to mark the logs but I still didn't know what to do with it so I made a 30 second log of the problem where it surges a few times and zipped it with a copy of my most recent bin. Maybe if I knew what part showed a surge I could fix it but the only thing that looks like it changes is the O2 voltage.

 

I'll take a look later this afternoon and post back. If you have a WB connected, which one is it and which channel is it on?

RBob.

 

No WB connected yet, hopefully I can get it in sometime this week but I've been super busy lately.

 

The log file shows two items of interest. The one I'd like to mention first is the A/C cycling. I don't believe it is the cause of the surge, but need to mention it. The A/C is on for 5-6 seconds, then off for 1.5 seconds. If that time frame matches the surge let me know.

The second item is seen within the O2 value and the INTegrator. It wasn't my first guess above, but is mentioned. The INTegrator is making up for a lack of proportional gain. This is shown by the INT rising and falling in a steady pattern. Along with the O2 overshooting. The overshoot, which is very high and very low O2 values is causing the engine to react. This is felt as a surge.

The attached graph shows the O2 & INT. Note the relationship between the INT and the O2 value. For anyone following along w/o an EBL, this discussion applies to all closed loop GM ECM operation.

TierAngst, I hope you don't mind me showing the graphs and such from your data log. This is just a information post to help out.

What needs to be done is to increase the proportional gains a little. Don't want to go too far. Otherwise the prop gains will cause overshoot, and a more rapid oscillation in the surge (can sound like a bad U-joint). The steady state cruising shows an airflow of 61 - 62 gms/sec. I recommend that the prop gains from the 64 gms/sec row up be increased by 10%.

The prop_1 image are the current gains. The prop_2 image are the new suggested gain values.

Try these out with no other changes. Let us know how it works out.

Thanks,

RBob.

 

I do feel the surge the same time as the A/C kicking on and off or at least in very close proximity based on me glancing over at the laptop as I obviously can't stare at it while driving. I will try your recomendation as well though and report back after.

Oh and feel free to use my stuff as an example whenever you feel like, you're a great help and I hope to be of some help too even if it is just showing my mistakes.

 

Just wanted to say that I do not consider this any sort of mistake. Getting into tuning proportional gains is an art. Takes some to understand. But once there, it is a level of closed loop tuning that brings rewards.

RBob.

 

here is another short clip with the surging right at the end of the file when I'm doing constant speed around 45-50mph this was after updating the prp. But I believe it did help the INT stabilize like you were saying. It surges just slightly every time the A/C kicks on or off.

 

I don't know if you will be able to entirely eliminate the slight surge. There are two things happening when the A/C turns on and off. The first is that the A/C compressor adds a load, then unloads when it turns off. That in itself will cause a change in the engine.

The other item is that the IAC moves to compensate for the A/C. The idle routine learns 2 positions: one for when the A/C is off, another for when the A/C is on. The A/C on IAC steps will be higher then when the A/C is off.

When cruising down the road the IAC will open when the A/C comes on. How much it opens is the difference in idle steps between A/C on and A/C off.

It may be the IAC opening and closing that you are feeling. A test that you can do to narrow it down: with the engine idling and the A/C on (actually engaged), disconnect the IAC. Watch for stalling and a high idle when the A/C turns off.

Now cruise and check for the slight surge. Is it better or worse?

If better, then trying to reduce the A/C on IAC steps at idle is the best bet. This can be accomplished by reducing the idle RPM increase for A/C on. And/or playing with the SA timing. Maybe a little more timing will help. Or, a little less timing.

Reducing the IAC steps for A/C on at idle will reduce the change in IAC steps while cruising.

RBob.