Testing Battery Drain 3

Hmm. It looks like battery drain is going to be a common subject on this blog. The most recent was an '03 Dodge Ram 1500 with a complaint of battery going dead after sitting. The customer had spent several dollars trying to get a fix at other shops but still had the problem. Now the truck came in on a tow because the battery was dead and wouldn't start with a jump. I used the jump box and got the engine cranked to get the truck into a work bay. I didn't need to use the memory saver because the battery was already dead. I just disconnected the battery cables for a quick battery test. The battery tested bad so I installed a known good battery to use while checking the electrical system. I put the original battery on a slow charge. I would re-test it after it was fully charged.

With the good battery on the truck and after giving it time to "timeout" the modules, I checked the actual battery draw which was well within specs. So there was no excessive drain on the battery at this time. 

I had to halt testing at this point after I was rudely interrupted by quitting time and my stomach growling...

After an overnight on the charger, the battery was 100% charged and I installed it on the truck after removing my known good test battery. I gave the truck several starts while running accessories to get rid of surface charge. This battery still tested bad. I removed the bad battery and re-installed my known good battery because before I called the customer with a report I still needed to test the charging system. The idle rpms were very erratic on the truck so my voltmeter was dancing quite a bit. I could see I had alternator output but I needed to fix the idle problem. 

I should explain something. The engine computer has a learned memory that can be cleared with a loss of battery power. One of the typical things the computer has adapted to is throttle plate coking or buildup of carbon at the throttle. As the dirt/carbon/crud built up at the throttle the engine computer made changes to its idle strategy and those changes became learned programming. Loss of the battery power here cleared the learned memory so that when I started the truck it fell back to the original idle strategy and not an adapted one. In other words, it idled poorly now because it wasn't compensating for the dirty throttle bore. I could either let the truck run and operate long enough to relearn idle or, since the problem was a dirty throttle, clean the throttle. I usually choose to clean the throttle. I was right about the throttle I see. 

After cleaning the bore and plate the idle was steady and the voltmeter showed a good alternator output. 

In the meantime the customer had picked up his original (bad) battery and taken it to the store he purchased it from a year ago for an exchange. After doing the cleaning and idle repair with my battery installed I didn't want to chance losing what I had gained with another memory dump so I connected a memory saver while making the swap back to his new battery. 

I think I've covered it all. Good fix!

That's the magic. 

Kenneth Hayes

G&G Auto Repair

TPS code 22

Today is a '92 Chevy K2500 6.5 diesel with a complaint of a TPS fault code that won't go away. The TPS (throttle position sensor) has been replaced but the code remains.
Ideally what I would like to see is the code number and scan tool data that would show me what the engine computer is reading as TPS voltage. I do have scan tools that read these older GM systems but when I tried the OTC Genisys with the Pathfinder attachment there was no 6.5 diesel under engine options. Ok, I have an AutoXray tool. No communication. OTC Monitor 4000 Enhanced, remember those? Uh.. no. I have attachments for the Solus Pro for GM but maybe I should just read the codes as this scan attempt is turning into a time consumer. Now if you want to read the codes on one of these using primitive tools, you can. You need a working SES (service engine soon) light so turn the key and see that it does come on.

Turn the key off. Find your diagnostic connector under the dash. (Not a good shot.)

I'm gonna share a tip here that you might need to know some day, or maybe never, but it's a good tip anyway. If you look closely at this diagnostic connector you can see there are small "ears" sticking out from each corner at the bottom making it wider at the bottom than at the top. Those ears are the bottom of the connector no matter how it is positioned in the car. If you are working on a car and find that connector turned upside down from this picture, or turned sideways in either direction it doesn't matter. That widest part is still the bottom of the connector. The reason that is important is that to trigger the SES light to flash trouble codes you have to jumper those two terminals on the upper right of the connector. You can't tell where the upper right is if you don't know where the bottom is. You can use a wire or I'm going to use a paper clip. Jumper the two connectors at the upper right.

Once you get the jumper in place, get in a position to turn on the key and watch the SES light. The light should begin flashing when you turn the key to the on position. Don't start the truck. Just turn on the key and watch the flashes. You may need to make a note of the codes if there are several. On this truck the light should flash a 12 initially. That is one long flash, pause, two short flashes. The code will be repeated three times. 12, 12, 12 long pause and then next code if any. In this case the only code output other than 12 was 22. Turn your key back to off and remove your jumper.
Now the short description for a 22 is probably going to be something like throttle position sensor voltage low. There is a school of thought that you'll come across that believes a sensor code means you need a sensor. It's wrong. What the code 22 is saying here is that the engine computer isn't seeing a voltage signal from the TPS. Sure that could be a bad sensor but it could also be many other things and no diagnostics have been done by reading a code. It's the first step, not the journey.
The next step is to inspect. The sensor is obviously new and the connector looks ok.
This sensor has three wires. It needs a ground on the black wire, a reference voltage on the gray wire, and the blue wire carries the voltage signal to the engine computer. Since the code is saying there is no voltage on the blue wire, I'll start with that. Connect a meter to measure voltage on the blue wire and turn on the key. The code isn't misleading at all. There is low voltage on the blue wire.

What I need to know next is if I have a reference voltage and ground going to the sensor. If the ground and reference circuits are good then I can connect the meter between the two wires and see reference voltage when I turn on the key. Disconnect the TPS and connect the meter leads to the black and gray wires on the engine side of the harness, turn the key on.

The five volt reference voltage is there. Since I'm measuring between the sensor ground and reference wires at the connector then I know the sensor is getting what it needs from the engine harness. Let's go back to the blue wire. The blue is the actual signal from the sensor to the engine computer. I measured before and saw no voltage on the signal wire. This time I'll measure the voltage on the wire while accelerating by just manually pushing the throttle shaft open.

As I open the throttle, the sensor signal voltage rises from 0 to 4 volts. The sensor does work. So why do I have a code 22? A code 22 will set when the TPS signal voltage drops below .06 volts. It does that each time the throttle is closed so that sets the code. At closed throttle the sensor reading should be .5 volts. The sensor is slotted and is adjustable. This could be a matter of the sensor not being adjusted when installed. I'm going to monitor the voltage signal, loosen the mounting screws and see if I can get .5 volts at closed throttle by adjusting the sensor.

Success! What I'm going to do now is disconnect the battery to allow the code to clear from memory. Don't do this on most cars and only for a reason on those you do. On this model that is how the codes are cleared and not with a scan tool. I'll give it a test drive and then another check of codes to see if the 22, or anything else, returns.  

All good! That's the magic!

Kenneth Hayes

G&G Auto Repair

Testing Battery Drain 2

This '99 E350 Ford sits a lot but the customer says he has to boost the battery too often. The van starts fine with a jump start so he does that and brings it in to my shop. The battery is dated November '14.
The van starts fine with no boost needed in the parking lot so the battery is not dead at this time. I want to verify we have a good battery so a battery test is needed. I've mentioned before I like to use the tester that measures internal plate resistance so I need to disconnect the battery cables at the posts. Since this battery is not dead the KAM Keep Alive Memory or the engine computer's learned memory should be saved. If I disconnect the cables the KAM will be cleared so I'll connect a memory saver to the diagnostic connector. These devices keep the engine computer powered while the battery is disconnected.

This battery tests good but is only at 70% charge. I know the battery was discharged and jumped so it hasn't been fully charged. I'll need to put the battery on the charger.

Getting a full charge can take several hours so I'll install a known good battery from my stock and continue testing while the battery comes up. After the new battery is connected I can disconnect the memory saver.

After allowing for the modules to time out I check battery draw, which is within specs.

I need to test the alternator. I put the voltmeter on the battery and measure battery voltage before starting the engine. I'll compare that reading to the one I get with the engine running.

I like to use a simple voltmeter initially and only go further when I see a reason for more testing. Here I see a need for more testing. I notice the meter is showing a charge and is in the "ok" zone but not by much. There is no load on the system and the engine is at idle. That's not looking like a good alternator output. I turn the lights on and the alternator isn't keeping up. I'd like to rev the engine and watch the meter but I can't do that from under the hood on this van.

I need deeper testing and I need to be in the driver's seat where I can rev the engine as needed. I have a tool perfect for the situation. A wireless tester that connects to the battery yet lets me use a wire free remote to see what's going on while I'm in the driver seat. This alternator is bad.

I replaced the alternator and did another test with the voltmeter and with the tester.

After a few hours the battery is fully charged.

I again use the memory saver re-install the original battery. I turn the headlights on and crank the engine a few times to get rid of the surface charge, then its time to disconnect the battery for testing.
The battery tests good. The alternator tests good. The magic is done.

Kenneth Hayes
G&G Auto Repair