Tbi 350 has no power on the road

JDGMC said:

It's not uncommon to see knock cnts here and there be it communication artifacts or real knocks. What's important is if you are consistently seeing knock cnts and/or they are increasing - i.e. enough to change spark advance. You need to monitor multiple PID's under various conditions to make sense of it. See attached example

Click to expand...

Hello JDGMC,

Not to derail the thread, but I'm on the cusp of investing in a live data capable setup, so that
I can both up my troubleshooting game as well as getting to sit at the GMT400 cool kids table. :0)

Seriously, what setup are you running that created this chart? Lots of virtual gauges are OK, but
I'm much more interested in the relationships between the signals of interest? I've been raised on
graphical data, and they simply work best for me.


(credit: JDGMC's graph in reply #30 with my question added in orange?

By the way, there's so much to observe here. To me it looks like you can have Knock Count activity
before we go closed loop (as evidenced by the O2 sensors starting to switch) ...but it doesn't
seem to affect the timing? On the other hand, right around the '228 second' mark there is some
Knock Count activity, and above I see a sudden decrease in spark advance?

Also, you can watch the O2 switching frequencies change (slow down) at the warm idle, but speed
right up once you start driving? Anyway, this data graphing display looks to be a winner, so I'm
interested in what you are running.

Thanks in advance!

Yes, there is a great deal of information in the data stream when you apply graphical/statistical discovery. Unfortunately, there is no single platform to do this with automotive data acquisition software. I acquire the data using DataCat and export the data (comma delimited file). Import the file into JMP then hunt for complex correlations and use a statistical approach to validate a model. For simple scenarios I use Excel to get a better visual if I already know what you’re looking for. In the plot I posted, I used MS Excel to plot the data. I was on the 3rd iteration of tuning and that was one of the 8 plots I used to assess knock and spark advance/retard. The takeaway message on this plot is even with cumulative knock cnts at 244, 90, and 40 the command control system compensated. The vehicle idled, accelerated and decelerated just fine. I then reviewed other parameters and made adjustments. Three iterations later I found the sweet spot. It’s a fun process.

Mitchell93 said:

I would like to know though what a good knock data looks like...

Click to expand...

Stock 1990 5.7 here. I see 3-5 knock counts on startup. Assuming that’s the starter engaging and disengaging. Through normal operation I see zero KC. Under full throttle acceleration I’ll see up to 8 KC.

With that Modis you should be able to back probe at the ECM Pinout for the knock sensor and graph exactly what the computer is receiving.

Road Trip said:

I'm much more interested in the relationships between the signals of interest? I've been raised on
graphical data, and they simply work best for me.

Click to expand...

Here's a picture (for some reason I couldn't get "Print Screen" to work) in TunerProRT. If you right click on the graph, you can choose what you want to monitor (including Knock Counts and Knock Retard) Then on the upper left of the chart you can choose the "Time Range". You can also "Play Back" your data log at whatever speed you want - fun stuff!

PlayingWithTBI said:

Here's a picture (for some reason I couldn't get "Print Screen" to work) in TunerProRT. If you right click on the graph, you can choose what you want to monitor (including Knock Counts and Knock Retard) Then on the upper left of the chart you can choose the "Time Range". You can also "Play Back" your data log at whatever speed you want - fun stuff!

You must be registered for see images attach

Click to expand...

Getting back to the no power issue with the Mitchell93 truck. Hopefully he can get a scan when he gets the chance and provide a plot or data dump file for several PIDs so we can help him out.

Here's some data from a warmed up engine that was off for a few minutes then restarted, that knock deal I can not get to go away, I currently have the wire for it removed from the pcm pigtail, also the replacement pcm I got from the salvage yard wasn't giving my MAP 5 volts, so I can't trust it, I currently have the stock one back in it...

Can only attach 5... so here's 5 more...

Here are all the same pids wile I drive it on the highway, got it to 44 mph but I had a little downhill mixed in

Here are the other 5

I know how to fix it !!!!!! LS swap