STILL getting knock. How???

[AlphaAssault]

You need to know two things, the injector pulse width and the RPM.
available time (ms) = (1/((RPM/2)/60))*1000
IDC (%) = (injector pulse width/available time)*100

IDC (%) = injector pulse width / (1/(RPM/1200))

You need the EPROM ECU to be able to change the software. DSMLink or a premade chip will have it disabled. If you want to do it yourself the location is listed on the DSM-ECU yahoo group archives.

At ~5200 RPMs, IDCs are 89.7. i was watching the boost gauge when it hit fuel cut, and the gauge was right at 15psi. Obviously, Ive turned the boost down a psi or two, but how do those cycles look for that run?

 

[steve]

Considering that it's recommended not to go over 80-85% IDC not great.
Have you finished that boost leak test yet (pressure test the intake)?

 

[AlphaAssault]

Considering that it's recommended not to go over 80-85% IDC not great.
Have you finished that boost leak test yet (pressure test the intake)?

No. I went through the intake with a fine toothed comb checking for loose clamps or holes (pipes are mostly solid, however), but didnt find any. I dont have a pressure tester, so this was the best I could do. All I can say on that front is that I dont have any MAJOR leaks, because I have experienced those before. does 15psi sound too early to be hitting fuel cut, or are you simply asking about leaks for safety/efficiency sake?

 

[steve]

I don't remember any more if 15 psi on a S16g should or should trigger it but I do know that any leaks, and it doesn't take much, will cause you to hit it much earlier than you would. It also causes you to work your injectors harder and poor mileage.

So your faced with turning the boost down as your only fix right now. You should think about making yourself a cap for pressure testing and supporting upgrades to go with your turbo.

 

[AlphaAssault]

I made a BLT over the weekend, and tested for leaks. I only had a bike pump, but it held ~15 psi for over 30 seconds. i hooked everything back up, kept the boost turned down, and went to log a run. Fuel cut AGAIN, still around 15psi (need to turn it down MORE, apparently) and just above 4000 RPMs. im looking at my log, however, and air flow (in Hz) is only (at the time of fuel cut) about 1400. im not even maxing out the MAF sensor (1600Hz), so im not sure why its cutting the fuel.

 

[steve]

Do you still see it cutting off the injectors when this happens?

 

[AlphaAssault]

Do you still see it cutting off the injectors when this happens?

yes. on the graph, the drop isnt quite as sharp, but its still very prominent. when driving, the hard lurching felt the same as it did last time i hit fuel cut.

 

[DSMopar]

If you're at 1400 it's not fuel cut... However....

A. Put the damn honeycomb back in the MAF, bad frickin idea with no tuning mods.

B. If you're hitting fuel cut and you're on stock injectors you're running lean.

 

[AlphaAssault]

If you're at 1400 it's not fuel cut... However....

A. Put the damn honeycomb back in the MAF, bad frickin idea with no tuning mods.

B. If you're hitting fuel cut and you're on stock injectors you're running lean.

I dont have the lower honeycomb; i wasnt the one who took it out. its on stock injectors, but a 150lph pump, if that makes a difference. and if its not fuel cut, then why does the injector pulse width drop to almost zero on my logger?

 

[DSMopar]

Copy paste us a log, not in some useful format we can't read. Like I said 1400Hz = no fuel cut. Could be it's hitting fuel cut and dropping to 1400 for a second.

Give us a log so we can see.

 

[AlphaAssault]

heres a screenshot of the specific part of the log i am referring to. However, first you need to understand that those funny little dips and spikes are just something the logger does. for example, that drop in air temperature, according to the logger, registered at -74*. yeah, dont think thats right. Airflow, and knock are both fine, the other fields have a spike in at least one place. if you have questions about the log, i have the source file, and it allows me to mouse over the graph and get exact values. and dont tell me i need a new ECU. my ECU is fine, the logger just doesnt cooperate perfectly with a '90 computer. it logs a friends 92 flawlessly. so yeah.

 

[steve]

Strange how the airflow is going up when the RPMS are dropping and then later there's no airflow but the RPMs are climbing.

Your logger is either really broken or you have some serious problems.

 

[AlphaAssault]

Strange how the airflow is going up when the RPMS are dropping and then later there's no airflow but the RPMs are climbing.

Your logger is either really broken or you have some serious problems.

youre looking at one of those spikes i was talking about. the RPMs for that run peak there where the fuel cut is, right around 2:06.

 

[steve]

But the section from 2:04 to 2:06 the RPM and IPW are dropping while the MAF HZ is climbing. That's wrong and then from 2:08 to 2:10 there isn't a MAF signal, it's flatlined but the RPM and IPW are climbing.

Your logger is either really broken or you have some serious problems.

Next time you log I'd like to see TPS as well.

 

[AlphaAssault]

But the section from 2:04 to 2:06 the RPM and IPW are dropping while the MAF HZ is climbing. That's wrong and then from 2:08 to 2:10 there isn't a MAF signal, it's flatlined but the RPM and IPW are climbing.

Your logger is either really broken or you have some serious problems.

Next time you log I'd like to see TPS as well.

from 2:04 to 2:06, both RPMs and IPW are recovering from one of those spikes. You cant look at the graph and know what the values during those times were. when i mouse over those areas, the only numerical value that is skewed at the peak of the spike. the graphical representation for that spike is the several-second-long peak you see on the graph. according to the data when i mouse over everything, the RPMs rise steadily, so imagine a gradually increasing line connecting those values at about 2:02.5 and 2:05. for example, the numerical value for the RPMs right at 2:04 is 3381RPMs. Same thing for IPW. As for the air flow signal, the numerical value for the segment you mentioned is 101Hz. the picture i posted is shrunken quite a bit, but the air flow line is much further away from the bottom of the graph on the original. this logger works much better with the 91+ ECUs, i admit, but it was free, and despite the spikes, reads things pretty accurately. its not as bad as it looks at first.

 

[steve]

I can understand a spike being a loss of data from the ECU to the logger but it's safe to assume all the commands during that time get hammered not just one or two. It's also not the TMO software at fault. I've logged plenty of 90 ECUs with it. Now 90 cars seem to have problems with "auto on" passive datalogger cables all the time and TMO only shipped active cables. You can try shorting pins 10 and 12 so the ECU stays in logging mode or look/build a different cable.

I can't really add anything more based on that log since it's all questionable and stick with my earlier comment.

 

[AlphaAssault]

I can understand a spike being a loss of data from the ECU to the logger but it's safe to assume all the commands during that time get hammered not just one or two. It's also not the TMO software at fault. I've logged plenty of 90 ECUs with it. Now 90 cars seem to have problems with "auto on" passive datalogger cables all the time and TMO only shipped active cables. You can try shorting pins 10 and 12 so the ECU stays in logging mode or look/build a different cable.

I can't really add anything more based on that log since it's all questionable and stick with my earlier comment.

alright. thanks for your help though. can you mark this as resolved? heh, i think we solved the ORIGINAL reason for this post.

 

[steve]

can you mark this as resolved?

Only the original poster or a moderator can do that. I'm neither.

 

[AlphaAssault]

Figured out the lurching -> due to bad spark plugs. Would hit fuel cut at higher boost levels, and would lurch because of plugs at lower levels. Injector pulse width would drop in both cases, but new plugs (gaped properly) solved the lurching. Just an FYI.

 

[tallywacker]

Figured out the lurching -> due to bad spark plugs. Would hit fuel cut at higher boost levels, and would lurch because of plugs at lower levels. Injector pulse width would drop in both cases, but new plugs (gaped properly) solved the lurching. Just an FYI.

Remember this lesson for next time. The obvious maintenance things must be in good shape be you troubleshoot anything else.