Starting-charging testing with a scope
Several years ago I posted up a bunch of testing of charging & starting systems. Over the last few months here on DN, I have read several posts about this sort of testing, and I thought I would start some discussion about this subject.
20 or so years ago, I sat in an electrical class taught by Vince Fichelli and he mentioned using the built in vehicle loads to test the electrical components on the vehicle. This got the gears to turning, and I wondered if I could test batteries, charging and starting systems, all with one test, all done in a few minutes. I decided to use my scope, and current probe & a voltage probe to do the testing. This project took me about 2 years to complete, and over that time, I tested a few hundred charging systems As vehicles came through the shop, I would hook up my scope and grab information. I also used two different conductance testers & a carbon pile to back up what I found with my scope.
When using this test, which is a dynamic test, you are able to get an overview of the total charging and starting system, which includes the complete starting and charging circuit, and by using the pico scope, (other scopes will do the same thing, but are a little harder to get the needed information), you can analyze the circuits, right down to watching the contacts on the starter solenoid close.
Keep in mind that a vehicle is engineered with a complete electrical system, that was designed for the electrical loads that were engineered for the vehicle. If perchance a battery has been installed that does not meet the electrical needs for proper starting, this test will find the problem. If you were to use a conductance tester, which is not a dynamic test, and is testing only the battery, it would find the battery “good”, but would NOT find the problem.
I will start off with the scope preset that I use. diag.net/file/f4a74o0xc… This is a capture of a known good charging system, with a new fully charged battery. I have put some notations on the known good, to show what is good and what isn't. diag.net/file/f5lz2parb… The nice thing about using a scope & current probe to test a charging system is, the amount of information that you can get with just one quick test.
In the case where a vehicle comes in with a complaint of a dead battery, you need to test the total charging & starting system to be complete in your testing, and this one test will fill that bill. One thing to do though, is to always make sure the battery is fully charged, or your data will not be correct. This even goes for putting in a new battery and testing, Is the battery charged to 100%? Most batteries are NOT.
Here is a case study on a 1005 Toyota T100 with a complaint of a dead battery. My first capture I found a lot of information. diag.net/file/f27qsaif2…
When the vehicle came in, it started good. I hooked up a charger and topped off the battery, then tested. The first thing that caught my eye was the charging voltage and amperage pulsating. When the engine was started, it would pulsate 3 to 4 times, then settle down and charge normally. When the engine was shut off for a few minutes, this cycle would be repeated. I now have one red flag, it needs a new generator.
Examining the charging system test, (the battery is a large part of the charging system), I see the OCV was at 12.55V, which is a pass. The inrush voltage is above 8.5V, (9.6 actual) which is a pass. but if I look at the charge rate after 2 minutes, which is the current flowing into the battery, (the current probe is hooked around the negative battery cable) I can see that this battery is inhaling 21.59 amps, which is enough to flunk the battery. If a new generator is installed with this battery, it will not be long until will be burned up.
There are several more things that can be seen in the waveform, by zooming in on the starter turn-on, diag.net/file/f76keeqz3… you can see the starter solenoid contacts close. As starters wear, the contacts get dirty and start to arc. Without a waveform, the only way you will ever notice this is when the starter solenoid starts to click, and there is no starter operation. The other place is to check the time between the starter turn on, and when the battery voltage rises above 10.5V. If this voltage remains below 10.5V too long, strange electrical problems start to happen. Things like power seats moving with no command, radio presets going away and a host of other things can happen. I have found that any time longer than 1 MS is too long.
I hope this might help out a little when looking for battery testers, and might help out when testing electrical systems. This test is not only for starters & generators, but will also work on many electrical components on a vehicle.
Thank you for tbe write, me as a young guy with little experience would love reading this kind of stuff. Thank you sir
What do you usually look for if you have a failing starter? Like vehicle towed in for no start, then once you check the car it starts everytime. Thanks in advance
Hey Marko. The best method I have found is to use an inductive amp clamp on the starter cable and graph the current ramp on a graphing multimeter or scope. Identify a distinguishable point and it's corresponding one next in line. Counting the ramps including the first and to the last one before it's next twin will show you the number of bars on the commutator. Ordinarily you'll see one or two…
Can i do that test like a relative compression testing? Let say i cant access the starter cable, can i just use the negative battery cable? Thank you for your response by the way, im gonna check that tomorrow for sure.
No, you have to isolate that particular load, but it is otherwise just like a relative compression test in most regards
I will put up some stuff on failed starters later on
Marko, Here is an example of a failing starter on a 2006 Honda CR-V. The starter worked just fine, it started the engine without trouble, even on a cold 35* F morning, it just sounded a little different than normal. The engine cranked over slightly slower than normal. You wouldn't notice it unless you had a good practiced ear and were familiar with this type of vehicle…
Thank you sir, i capture all components failure and after fix when theres a problem vehicle that comes, i checked my saved starter waveforms earlier and i noticed some difference on a new starter and old ones( defective). Thanks again
Marko, This is another example of a bad starter. A bad solenoid specifically. This is from a 2014 Honda 2.4L Here is a Click Click Crank event. On the no crank events there was a split second rise of current to almost 400 amps but it then dropped down to about 40 amps and didn't crank Here is a zoom of a Click no crank event. You can visually see the burned and worn solenoid contact disk. It…
Thank you sir. I actually reviewed some of my captures and found some pattern from the bad ones and god ones like the examlles you just posted. Thank you again.
These are great examples! I love testing where it lives, especially for batteries and alternators. I know there is more information they can tell us as well, but haven't been able to practice it much lately. Thanks again for posting.
Albin, thank you for taking the time to contribute this! I am always concerned about “surface charge” on batteries that come in undercharged or dead and get quickly charged up while in the shop to get through the inspection. How do you recommend preparing the battery for this test? Do you let it sit for a while after charging? Thanks
KOEO, I use 30 seconds of high beams to burn off the surface charge. Then I let settle for 5 min. Seems to work well. Usually pretty easy to tell with the before and after voltages.
William do you find after charging up the battery and removing the surface charge with your method that you will always have 12.4 volts or above on a good battery? I find we do a lot of back and forth about if the battery is fully charged or not here in especially in the winter.
Hi Bill, you can call me Will btw. I find they mostly have 12.5 to 12.6volts. The battery temp is important as well. I attached a chart as well for min voltage under load by temp. I will not test if the battery is below 65 degrees or below 12.4 volts and call it for a battery due to not holding a charge.
Interesting! Here I was always using the stalwart 9.6 V and I could be using 9.7 or 9.8 cuz it's always so nice and warm. Thanks Will
Great information! I'm new to using a scope so please excuse my ignorance. Would you explain the note about positive in coolant bottle.
Usually a test done to check the amount of electrical current passing through the coolant. Remember, that most OEM coolant is mixed with distilled/de-ionized water, which means that it is electrically not able to pass alot of current. As parts wear, and the coolant degrades, it can start acting as an alternative path for the current, on the ground side usually. Can cause all kinds of issues with…
Great information as always Albin - thanks for the time and effort to share this - have a Merry Christmas!