Is it expected to have RPM this choppy? I can see almost 200 RPM drops in certain areas. See my screen shot below.
Are these RPM readings likely true? Misfires maybe? Or are these OBD1 ecu just not that great as far as accuracy?
Setup is B16 with stock dizzy.

 

Could be grounding, ignition, valve adjustment issues. Although with valve adjustment out of spec, i believe youll have the check mark at the end of the powerband. Id clean all your grounds and make sure all you ignition parts are good. And plugs gapped accordingly to start with.

 

So RPM this choppy is not expected? That is my biggest question. I dont want to try tracking down a problem that is normal with this setup.
I image I shouldnt get reading like this, but just want to make sure so I dont waste a bunch of time troubleshooting expected behavior.
Im planning to setup an oscilloscope on the CKP (I think that is the correct signal. The one with the highest tooth count. That one is used for RPM right?) line and datalogging that signal to see if that signal is showing the same RMP choppiness as the Neptune Datalogs.
Then I can isolate the issue to either the signal to the ECU or something in the ECU.

Also Im curious about the valve adjustment out of spec you mention, what is your thinking there?
That it would cause a misfire and in fact my RPM choppiness is real?
Or perhaps the adjustment is somehow causing the cam to do weird stuff affecting the RPM signal to the ECU?

 

If anyone has any datalogs from some pulls to share that would be helpful as well. That way I have another entire setup to compare the datalogs too.

 

It is not uncommon to see some variation in your datalog traces... whether what you are experiencing in this case is normal or abnormal... I cannot say because I HATE using Neptune ****. If the car doesn't "feel" like it is hesitating throughout the RPM range... it is likely dampening variation in the data. I have seen perfectly good running engines using Hondata have similar variations in the datalogs... maybe not as severe, but not tracking a perfectly straight line as you might be expecting.

You have spent way too much time worrying about this already... move on to something productive. This ain't one of your problems.

 

A few things can impact resolution.

Computer CPU speed. If you have a laptop, make sure its set for max performance not battery saver. Too slow a computer = slow log of data.

Neptune datalog settings. Try adjusting the slider to bias your logging between realtime display and more datapoints logged to file (what you are concerned with)

And finally the hardware in use doesn't have nearly the amount of crank position data as new ECUs woold have. Consider an aftermarket ECU.

If the motor seems fine, then it's probably fine. Use a 3rd party knock sensor to verify you're safe and move on to the next project

 

Hi buddy, did you get to the bottom of this? I have similar issue but bigger fluctuations. In your data log, log the injector duty cycle. There's a chance that while the rpm signal is bouncing that it's messing with the duty cycle of tue injectors too. In my case, its halving the duty when the rpm signal jumps.

 

Hey, I havent figured this out yet. I have moved my ECU and conversion harness over to my 99 cx hatch (bone stock) so see what I find. Im lucky enough I have another car with exact same OBD version. I did about 3 full throttle pulls and I see much less choppiness in the RPM signal on that car. Under 100 RPM was the largest step size I could could find in those datalogs.
However on my EM1 I did randomly see HUGE spikes in the RPM like you mentioned. See my screen shot below.
The duty cycle did change drastically, but the "Fuel Value" did not change by much on the large dips. I cant graph "Fuel Value" but I can see the value in the data window, and I dont see it change much during the big RPM dip.
I figure the duty cycle is being mathematically determined by injector PW and RPM. So the if the ECU/Program thinks the ECU has dropped RPM drastically the math will make the duty cycle change drastically. Make sense?
At this point my best guess is that the signal from the distributor on my EM1 is more noisy. All of this stuff is very iterative and takes alot of motivation to go out there and try different thing to figure it out. So its slow going for me.
I do think I feel hesitations though on my EM1 when doing some full throttle pulls.
I didnt seem to feel those hesitations on my 99 cx hatch when running the Neptune RTP setup on it.
I put the chip (st00pid tune from almost 20 years ago) back into my ECU and did a few pulls and dont seem to feel the weird hesitation stutters.
The stutter/hesitations are very very subtle.
So I dont know what to think of it.
I have lots of things I want to do to narrow this down, but like I said the process is very iterative. So just takes time.
One of the things I want to check next is my grounds like ls joke mentioned. I was looking at the FSM and if Im reading it correctly the shielding for the CKP and CYP and TDC are connected to the thermostat GND.
I've also been real lazy about hooking an oscilloscope up to those signals as well. Maybe one of these days Ill get to it.

 

On my 1000cc injectors, Duty cycle = RPM x 0.00835 x Fuel Value.

Im still trying to figure out why im getting spikes too. Positive Spikes in the rpm reading wont hurt anything as they will increase the fuel. However in my case i get dips, and a dip at 7200rpm down to 3600rpm causing the duty cycle to also half, is going to cause knock.

Did you get my pm?