Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

unstable dwell and AFR on WOT

Link G4 Plus Software Tutorial

Forum Posts



Tech Articles

Discussion and questions related to the course Link G4 Software Tutorial

= Resolved threads


hello everyone,

The attaching Log pics is from a bmw M42b18 with ITB, I have some unstable dwell and AFR on full throttle. I have check the polarity of the trigger and it is correct. Do you find normaly? What else can I check. There is possibility to be interference on Trigger signal? I use VW red coils.

Attached Files

I think the scaling of your graph makes that look like a lot. The dwell is only variing by 0.00015 seconds!

Yeah, I agree with David, dwell looks worse than it is due to the auto scaling. Although the 2nd pic it looks a little more erattic than I would expect if the trigger is good. Can you do a trigger scope at WOT, 5000RPM and attach here.

Your screenshot does not give enough info to validate why the lambda is erratic. The first thing you need to check is if the injector PW and fuel pressure are stable? - if you have stable fuel flow then you should have stable lambda... In most cases where I have seen erratic lambda but with stable inj PW it is due to the car having a short dump pipe off the turbo, a leaking wastegate sucking air, or misfire etc.

Since dwell is usually depending on battery voltage, can you add that to the graph -- perhaps the battery voltage is changing and there is more interpolation on the dwell. What does your dwell table look like?

First of all, I want to thanks you all for your answers. I have add pics from my dwell table and the voltage also from the same log. I don't have the time to make a scope today at 5000rpm (I have monsoon and I need to use picoscope), but I will do it and I will attach here.

Also I can see that the injector PW it is a little also erratic.

I have make a adapter PNP for my wiring harness around 40cm without shielding cable. it could be there the problem?

ADAM, what other info, it will help you to have a better image about the situation?

Attached Files

In the pic "unstable mixture4.png" you can see injector PW is quite stable but you still have 3 or 4% variation in measured lambda. Most likely this is one of the external factors that I mentioned earlier.

Your dwell doesn't look too bad to me. Usually I am happy with up to about 10% variation. If I see more than that with a relatively heavy engine like this and a decent tooth count trigger wheel then it often suggests the ecu is not getting accurately timed edges from the trigger. That could be many things - electrical noise, wheel runout, wheel inaccuracies, sensor mounting etc. But as I say, yours is not extreme, maybe a little more than normal so it would be worth looking at the scope but that's about all.

To understand how dwell works and why there is variation, consider at say 6000RPM, 3ms dwell is over 100deg crank rotation. The most critical thing that the ecu has to get right is the spark timing - dwell is less important. So in a perfect world, if the crank was turning at a dead constant speed, the ecu can accurately predict when the next spark event needs to occur and start charging the coil exactly 3ms (or 108degs) before then. However, what happens if the crankshaft speed changes after the ecu has already started charging the coil? It cant just shift the spark timing to keep the dwell correct - it must instead stretch or shrink the dwell to ensure that the spark always occurs at the correct time. Some ECU brands only report the dwell that was commanded. Link always reports the actual dwell that was used so it can look more erratic than some brands. Some ecu's don't even separate spark and dwell timing - they just fire the spark at the wrong time when crank speed changes...

Adam, thanks you very much for your detailing explanation for dwell. I will add shield cable to the PNP adapter to be sure for electrical noise. If I have any news I will post it here.

I suggest you do the triggerscope at 5000RPM first as requested, I'm not convinced you have a problem at all.

OK, Adam, I will do it and I will post it here.

So, I take two mesurement with pico scope one in idle and another to 5000-5500rpm. I try to upload like picoscope file, but not allowed it. I attached like photo.

Attached Files

Adam did you have the time to see the trigger scope picture?

Yeah I dont see anything there that the ecu would be unhappy about, it looks fine to me. I would not change anything.

Thanks you very much.


Is the signal from the crank the right way round? if you look at the idle pattern it drops first but the cam rises? I always thought the signals were always meant to rise first

We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?