When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Since installing a Power FC, I've noticed some starter kickback on cold (<20C) starts. I do not experience the issue during hot starts. The car was also recently tuned by Steve Kan, but the kickback persists.
My first thought was that cranking timing might be too advanced, although this seems unlikely given that hundreds of folks have run the basemap and complaints of kickback are relatively few and far between (judging by search results). I have posted a few shots of the rows and columns in question (Datalogit soon). I believe rows P09-P10 under columns N01-N02 to set cranking timing, referencing this post. Arghx notes that a sub-60C ignition retard value of 15 degrees may also apply while cranking.
So to my specific questions:
1) Is the cranking timing shown in the screenshots appropriate, or should I play with retarding it a bit?
2) Is there any other Power FC setting I should be looking at (perhaps temperature related)?
3) Failing any software issue, are there potential hardware problems (sensors, etc.) I should be looking at?
Bump, I also had similar issues but with the weather warming up it seems less rough during engagement and backing off. Might be settings might be starter in my personal case. Ill sub to this thread! Good luck
I have never logged the cranking IGL/IGT. You should do it and post it here.
But I can tell you this with 100% accuracy: If all three idle speeds are set to 0 rpm, the PFC will actually run the timing in the timing area instead of the hidden idle maps.
This allows better control over idle as it now becomes 100% manual; fuel, timing, and air all by you.
I was experimenting lately with winter idle and starting. I also experienced what seemed like kickback. But I do not think it was that as I had actually reduced timing between P10,N1 up to P6,N1-N2. When I put the advance back in, the kickback feeling stopped?
Mine starts better with negative split starting and idling timing. See attachments.
My N1 starts at 800rpm not the useless 400rpm. The second attachment is IGT split but the header is cut off.
I've attached a screenshot of RPM, IGL, and IGT during a cold start (cold being 18C) in which I experienced "kickback" (if it is in fact kickback). This is visible in the RPM plot as a dip in RPM during startup. My cranking IGL and IGT are 10 and 4, respectively.
I'm fairly sure that my cranking timing is coming from P08xN01-N02, though I could do a map watch to be sure.
Any thoughts? Maybe retard IGL a few degrees in those cells and see if it clears up?
Do you have porting? What injectors do you have? Have you removed the fast idle/thermowax system or the idle speed control valve? Is this a 3rd gen or a 2nd gen Rx-7? 2nd gen does not have a fixed crank angle sensor, so it needs to be checked with a timing light.
The little blip during initial rpm run-up, about 6 seconds in, isn't cranking in the literal sense of the word. Combustion has already occurred, rpm is over 600 here. Those 4 cells of the spark table probably aren't going to impact it.
However, if you really want to play around with it, you need to do a bunch of starts with different spark settings. For example set both IGL and IGT in that area to 15, then try -5, 0, 5, 10, 20. But that's a lot of work to go through.
This is a 3rd gen with stock ports, stock injectors, fast idle, thermowax, ISC, and airpump. A map watch pretty clearly reveals the cells in question (more like P06-P07xN02-N03). I'll play around with values in that general area to see what happens.
Fortunately, the weather has warmed up and I rarely get any kicking while starting anymore (and when I do it is minor, as in the log above). Hopefully we will still get a few colder nights so I can test a few timing changes. Thanks for the help, will post back if I make any conclusions after some trial and error.
I played with retarding the timing some (nothing systematic yet, but retarding timing ~5 degrees did not seem to do anything). I'll play with it some more in a more systematic manner.
I wanted to report back that I do seem to be getting the "kick" on hot starts as well. I've attached a few logs showing the phenomenon. I noticed that the "kick" seems to correspond to the ISC duty drop after initial startup--not sure if the correlation is necessarily causation, though.
I've attached my map below. The car is a '93 with stock ports, approx. 10k miles on a rebuild completed in 2015. All emissions (except as noted below) are present. ISC is functional and plugged in. Relevant engine modifications are as follows:
Its worth noting that the car did not have this issue on the stock ECU. I also just re-did the idle learn procedure and adjusted the idle screw, but I can't tell if it has had any effect yet.
im pretty sure this problem is caused by pickup noise. your CAS signals have the lowest voltage at low rpm. the starter also generates a lot of noise at the same time. so the computer is having a tough time figuring out engine angle, and timing gets off causing kickback. i dont use a powerfc, but what worked for me on my haltech is increasing the trigger voltage threshold. other than that, not sure what to suggest, other than make sure your trigger wires are shielded as far as possible. you may want to try a different starter. and make sure the high current leads to the starter is short as possible
Attached map with a few changes but will not help your problem but is better.
Do comparison.
But I notice this:
(1) Your idle fuel is about stock. But in Settings 3 INJ ADJUST; you are reducing fuel by 7.8% more for 1000 rpm which results to 3.9% at 1500rpm.
(2) You also reduce fuel the same amount for map sensor voltage 1.0V and 1.5V. Both of these is extreme vacuum which only can occur at heavy off throttle high rpms lifting.
Here is something strange but has helped a few in the past.
Check and verify that your map is good.
Save it to your laptop.
Verify saved map matches that on the PFC.
Initialize the map on the PFC.
Reload saved map into the PFC.
I re-initialized the Power FC and reloaded my map, but the issue persists. I did return all INJ ADJUST values to 100%, which did seem to smooth out the idle somewhat (used to get occasional stalling/surging). I noticed that idle A/F on the wideband is around 16-18 now, vs. 19-20 previously (seems reasonable given that the airpump is injecting air into the exhaust ports). I'm not sure what Steve's reasoning for reducing those values to 92.2% was, but I feel like it idles better now. However, the changes to INJ ADJUST did not solve the original problem.
I appreciate you taking a look at the map. I'd like to understand a few of the changes you made a little bit better (apologies if this is covered in your tuning notes, I'll PM you for a copy).
1) Do you set the rev limit to 9999 to get rid of fuel cut?
2) Idle O2 and timing feedback are turned off. In this case does it use the values in the appropriate fuel and timing map cells, making the idle manually set? Does this disable the idle learn procedure?
3) Can you explain your logic for changing secondary transition to .1ms in settings 5 (and rear secondary to .1)? Do you find these settings are better for the stock injectors?
Thanks for your help, I appreciate it.
As to the suggestion that this may be related to CAS trigger noise. I do think that is a real possibility, but to my knowledge you cannot change trigger voltage threshold in the PFC (though I look forward to being wrong as it is something I would like to tinker with). I will think about other potential causes. The only thing that comes to mind for now is a bad ground.
I
1) Do you set the rev limit to 9999 to get rid of fuel cut?
2) Idle O2 and timing feedback are turned off. In this case does it use the values in the appropriate fuel and timing map cells, making the idle manually set? Does this disable the idle learn procedure?
3) Can you explain your logic for changing secondary transition to .1ms in settings 5 (and rear secondary to .1)? Do you find these settings are better for the stock injectors?
(1) Fuel cut is the worst way to limit anything, so 9999 is one way to prevent it for over revving. YES.
(2)Idle O2 FB requires the INJ correction map for the idle area to contain values larger than the value in Settings O2 Feedback Setting. Your INJ map has all 1.00 and the FB value is .672. How well it works with these non stock values is anyone's guess. Especially with those other leaning out values. Having the O2 FB switch turned off negates this other values and no FB control. FB control with a working air pump and cats is used to pass emissions.
Nothing to do with idle learning.
(3) The stock injector transition values are bad. I logged them with stock and non stock injectors and came up with this BEST setup for smoothest transition.
All of this is in my notes on/in a yahoo private tuning group that Raymond - arghx - now controls.
I wanted to update this thread with my resolution to the kickback I was getting immediately after cranking ended. A map watch indicated the timing was generally falling between P07-P09xN01-N02 during cranking and initial combustion. I took arghx's suggestion that I play with values of 0, 5, 10 etc. for leading and trailing timing.
I find that setting both leading and trailing timing to 5 degrees advance eliminates the hiccup I was experiencing. Unfortunately, I am not sure yet if it is the 0 split or the retarded timing in general that has solved the issue. I am also not sure why my car in particular experienced kickback on the basemap cranking timing.
02-07-17, 01:54 AM
