Battery Voltage Profile Of An Engine Starting (1998 v70)

[MrAl]

Hello there,

I thought this information might be useful to other members who work on their own car or are just interested in this kind of thing.
This is part of an analysis technique for troubleshooting the battery and starter and even some points about cylinder compression.

The part i am showing here is just about when the engine is being started, but if you disconnect the ignition coil wire you can check if you have even compression on all the cylinders. I havent done that yet however.

This starts by simply turning on the key, then looking at the battery voltage with an oscilloscope. The waveform that appears shows what is happening with the starter and battery, and also you can see some cylinder information. There is more explanation in the image.

This is just the start of this kind of analysis. I am just learning about this so i only got this far for now. It is possible to troubleshoot the ignition system itself and even the fuel injectors but i did not get that far yet. What i did notice however in the scope waveforms for my car is that there is some variation in the plug firing which i think indicates that there is still a problem with the ignition system like the wires or distributor cap, etc., which i expected since i did not change out all those parts yet. It's getting cold out around here now so it gets more difficult to work on the cars outside.

Just to note, i numbered the different parts of the waveform and provided a little information about those parts.
One thing i left out was when the voltage regulator kicks in completely and starts charging the battery. I'll try to get that next time. It should just show the voltage ramping up to around 14 volts on this car 1998 v70 non turbo.
You will have to zoom in to see the numbers outlined in red, but the numbers just run from left to right starting at #1.

Oh BTW the oscilloscope being used is just a low cost OWON which runs around $125 USD. For what it does it is well worth that price, and it even has a 20000 count multimeter built in. That means instead of getting a battery reading of 12.45 for example you get 12.456, one more digit It also runs on batteries so i can carry it out to the car without much trouble. The laptop is more of a pain to drag out there along with the USB scope.

[volvolugnut]

Where did you make connections for voltage? I have a mini O scope I may try this with.
Volvolugnut

[MrAl]

Where did you make connections for voltage? I have a mini O scope I may try this with.
Volvolugnut

Hello there,

Directly across the battery terminals for best results, i went through the line +12v and signal ground connections on the OBDII connector. Better on the battery though.

If you perform this test in order to see the startup profile like that you have to be on a horizontal scale such as 500ms per division, but you can experiment with that. If you can freeze the waveform on the scope you can then take a picture with your phone or something.
Once the car starts, you can go to a smaller time scale to see what is happening as the engine keeps running.

If you want to do the compression test you have to remove the ignition coil wire so that the car does not start, then turn the engine over several times and freeze the waveform so you can analyze it. By keeping the engine from starting up you get a continuous reading of what is happening with the compression in all the cylinders.
I wish i did this test a long time ago i would probably have found out that one of my spark plugs was loose as the compression in that cylinder would probably be lower, by how much i dont know though. If they are all the same i would think everything is ok with that.

[850oldschool]

What were you using for probes? Did they come with the scope?

[MrAl]

What were you using for probes? Did they come with the scope?

Hi,

I used one of the probes that came with the scope, and they are 60MHz probes.
I used the 10x setting on the probe and 10x setting on the scope, but i think 1x and 1x might be better too for less noise, and i will try that sometime this week hopefully today. I'll also be trying to get a longer snapshot image that shows the regulator pulling the voltage up to the full 14 volts for charging the battery which didnt show up last time because i forgot you have to leave it longer. For some reason this car takes a few seconds before the regulator lets the battery start to charge again, dont know why. Every other car i had it started to charge right away.

Just to note, this scope came with two 60MHz scope probes and one alligator chips probe, but from what i had read some people only get one scope probe with their scope so if you get any scope of any kind check the ad writeup to make sure you get two or at least know what you are getting.
If your scope allows voltages of at least say 20 volts max i would say it would be ok to use a 1x probe which includes the alligator type probes because they are 1x also. Some scopes only allow 10 volts max though so you have to watch out. The smaller scope i have somewhere (i think it is a DS138) only takes 10v max i think so i would have to use a 10x probe only. The 10x probe knocks the voltage to the scope down by 10 times so 100 volts on the probe tip means only 10 volts getting to the scope, as long as the scope input is 1 Megohm which a lot of scopes are.

I also also thinking of soldering up a little filter circuit to use at the tip of the probe to reduce noise pickup if i dont get better results right at the battery terminals. My scope displays have all been a little noisy so far and since i dont need super high bandwidth to monitor the engine starting sequence that should allow me to get much clearer pictures.
Also during the day the glare from the glass face of the scope makes it harder to get snapshots with the camera too so maybe better at night.

[MrAl]

Hello again,

A little update. I took some more scope pics today only this time I used some low pass filters to filter the noise out of the signal so the important parts of the waveforms could be analyzed more succinctly. A description of the pics and some interpretation follows.

Pic 01: Attempt to start the engine, then another attempt and it started. The voltage levels off
then finally after 4 seconds it gradually increases to around 14v which is the normal charging voltage.
The vertical scale is 5v per division.
Pic 02: Another attempt to start (far left side) then another attempt and it started. A little more detail
because the vertical scale is 2v per division.
Pic 03: The engine is running, shot taken using 154Hz low pass filter.
Pic 04: Engine running, shot taken using 154Hz low pass filter.
Pic 05: Engine running, shot taken using 330Hz low pass filter.
Pic 06: Engine running, shot taken using 330Hz low pass filter, zoomed out.
Pic 07: Another engine starting sequence.
Pic 08: Another failed attempt to start followed by another successful start.

Interpretation:
Pic 01 basically just shows the startup dip and the voltage regulator kicking in
Pic 02 shows the same thing with more detail
Pic 03 I need more info to properly decode that signal, belt/pully ratios and related.
Pic 04 Same with this one need more mechanical info.
Pic 05 Irregardless if this is due to the alternator voltage or cylinder force during compression and intake, the idle is uneven.
Pic 06 The unevenness of this waveform is most likely due to some slight misfiring. The repetition period is clearly seen. One cylinder still needs attention. I'd have to use channel 2 to try to detect cylinder #1 to tell which cylinder is still acting up.
The dips most likely show the weak cylinder. This i assume because the engine makes a very slight misfire noise at regular intervals so i don't think it's a blown diode. To be sure i'd have to get the mechanical data relating crankshaft to alternator rotation ratio.
Pic 07 is another starting sequence.
Pic 08 is another failed attempt to start and then a successful start-up. The failed attempt shows the individual cylinder cycles.
Also, a little bit interesting from this pic is that the second group of hums during the failed attempt to start show the voltage gradually decreasing. That could be another test point for testing the battery health.