They Die In Threes part 1

I started the day with three automobiles to take a look at. The 08 Mazda CX-9 had a misfire complaint. The 07 Chevy Impala had a no-crank complaint. The 02 Buick Rendevoux had to be battery boosted to start. No problem. I'll just grab the machine that tells me what is wrong with a car and get started. Oh wait, there is no such thing. It would be kind of cool to grab a tricorder and run diagnostics quickly just to see what the problem is. That isn't in the real world, though I do know a lot of people believe that to be how it works. 

I'll explain how I do diagnostics and repair. At least, I'll give a few examples. First, I guess it is similar to triage. I have to take a look at each job on the schedule and see how to best get the work done. The 07 Impala had been towed in as a no-crank. I should explain no-crank. The starter is supposed to turn the engine, or crank it over. If that doesn't happen it's a no-crank. If the starter does turn the engine fine but the engine doesn't start, that's a no-start. The Impala, for me, did crank but very slowly. It did start as well and I drove it into the shop. Because it cranked slowly I suspected I might have a low battery condition. Low battery could be caused by an alternator not working. Since the car was runnning and might not start next try it seemed like a good time to see if the alternator was working. I put a voltmeter on the battery and could see just over 14 volts so the alternator should be ok. 

You want to see the battery voltage reading with engine running to be more than battery voltage with engine stopped. The meter was already connected so I just had to notice the reading after shutting the engine off. Just under 13 volts and steady even with lights and key on. Basic testing, but looked like the alternator and battery were probably ok. 

I still had two cars outside that needed checked so this was a good stopping place. I'd probably be looking for a bad connection or a starter problem on this one. I went to the Buick that was having to be jump started. I tried the start first with no booster and it cranked fine. Not even a slow crank, a nice normal fast spin and started right up. I drove the Buick into the shop. I tried restarting several times and found no problem. A basic rule of diagnostics is to "verify the complaint". It's nearly impossible to find a problem that you can't re-create. My plan with the Buick was already formed though. Open the hood, pull the key, shut all the doors and let it sit right there at least 3 hours. After that I would test for battery current draw. Maybe the battery was being drained while sitting.
I now had two of the three vehicles situated and had to take a look at the 08 Mazda. The Mazda was towed in but only because the owner did not want to drive it with a misfire. The complaint was a misfire. The car started and ran fine with no miss. I tried to load the engine a bit by power braking but still no miss. I decided to let the Mazda sit inside as well and rest. When the engine was cool I would use a scan tool to see if a particular cylinder had triggered a misfire code and begin checks with that cylinder. So.... back to the Impala since it was the only one with the complaint verified and present.
Now the Impala was the no-crank complaint that had cranked slowly for me. I tried to start and now it was no-crank. I didn't hear a sound from the starter solenoid which is usually a click. I watched the lights and they didn't seem to be losing power. The headlights go out normally on crank with this model so there was a bit of room for error in judgement. If the lights go out on a crank you probably have a bad connection. Here they seemed ok but since I wasn't sure I wanted to test the battery connections. You can make things complicated or you can make them easy. I prefer to work with easy and let any complications come from the car, not from me. I could take the connections loose, clean them and try again that's one approach. I could even do voltage drop tests with a meter while trying to crank. What I did was put a booster/jump pack on the battery connectors. If it then cranked I had a bad battery connection. It made no difference. I put the booster negative cable directly to the engine block and no difference so the negative cable was probably ok.

Next up would be to check at the starter for battery power to the solenoid. I ran a jumper wire down to the starter and to a test light placed where I could see it from the driver seat while cranking. I had power to the starter battery terminal and I had power to the solenoid wire from the starter relay. I could have enough voltage to light the test lamp but not to engage the starter and I could do voltage drop testing at the starter but I chose the tap test or whack it with a hammer method next. After tapping the starter with a hammer it did engage on a crank and did crank slowly. Bad starter. 

Diagnostic tools needed:

Tools needed to repair:

Thanks!

G&G Auto Repair

Ford Misfire

This F150 had a misfire I could feel with the engine running. It felt like a secondary ignition problem. It's hard to describe. I've heard it called "fishbite". Sort of miss miss hit miss. Very random and intermittent at low rpms. Anyway.....

The stored codes showed #2 cylinder had been misfiring.

It's important to note that a misfire code only means there was a misfire. The code doesn't tell you why it was missing. Ignition misfires are the most common on a Ford and it did feel like an ignition misfire. The most helpful item here is that we can start focusing on #2 cylinder without having to determine on our own which cylinder has the misfire. Another useful check is called mode 6. If we pull up mode 6 on the scan tool we can look at the other cylinders for any misfires that the engine computer noticed but weren't bad enough to set codes. Only #2 had misfire counts stored.

Time to take a look at #2 spark plug. It's hard to see in the picture but there was a lot of moisture (water) on the plug boot and even more on the spark plug itself. 

The spark would go wild in that moisture. There is our misfire. A new plug and a new coil is needed for this cylinder. The other cylinders had no moisture but got new plugs as well. The plug boot on #7 was in bad shape and was replaced.

Thanks for reading!

Kenneth Hayes

G&G Auto Repair

Dodge Truck Stalling At Idle

09 Dodge Ram with 5.9 engine had been sitting unused for awhile so the owner had to jumpstart the battery. He then had to keep one foot on the accelerator giving just a slight throttle or the engine would die. 

If you've been following my posts you know that when you lose battery power you lose adaptive memory. Lose adaptive memory and some things won't work properly until the adaptives are relearned. We'll be looking closely at idle control adaptives on this one but only after making sure we have a good battery. 

The truck idled with no stalling when I drove it into the shop. This means the computer had probably relearned idle on the way here. I wanted to save the adaptive memory but I needed to test the battery. A memory saver will keep the computer powered while you disconnect the battery and you won't lose adaptives. 

The old battery was bad, so out with the old and in with the new.

Once the replacement battery was installed I removed the memory saver. Since the idle adaptives were relearned and since I saved learned idle I could look at idle air control counts for a problem. There are some things you can only see with a scan tool. IAC counts are one of those things. The ideal range here is 10-24. If you see higher numbers the computer is having to command more idle air to maintain idle speed. Usually there is a buildup of carbon on the throttle causing this. 107 here meant the throttle plate might be dirty. 

After cleaning the throttle I checked the counts again. 

They were much better but still higher than our target range. I knew the throttle was clean. There is a minimum air adjustment that can be checked. Very easy to check when your scan tool will command it. Minimum air takes the idle control motor out of the picture. Only the air allowed past the throttle is maintaining idle speed. Just enough rpm to not stall when placed in gear is my rule of thumb. The minimum air was fine. Using the scan tool I did a minimum throttle relearn twice. I've ran into that before on Dodges, sometimes more than one relearn. But it worked.

Thanks for reading!

Kenneth Hayes

G&G Auto Repair

Another Ford Lean Code

This Mustang w/4.0 engine was setting a lean code for bank 2. 

At idle and low rpm both O2 sensors were switching normally so I ruled out a vacuum leak being present. A fuel delivery problem would most likely affect both banks and I only had a code set for bank 2. I took a look at the freeze frame for the lean code. Freeze frame is a sort of snapshot of sensor data during the event that set the code. 

Short fuel trim for bank 2 shows a LOT of fuel being added. It isn't likely I had a fuel delivery problem because bank1 looks to be in fine shape. I could also see this event was on a warm engine at highway speeds. I set up to monitor both O2 and went for a test drive. When I went from a steady cruise to a hard acceleration the bank 2 sensor went to full lean but bank 1 showed the increased fuel. 

It was at this point I felt I was dealing with a bad sensor, bad wiring, or bad ecm. Easiest test from here would be replace the sensor and re-test. I did a wiring harness wiggle test while monitoring the signal and found nothing. Replacing the sensor and recording again under hard accel I could see I found the fix. 

Not bad!

Thanks for reading.

Kenneth Hayes

G&G Auto Repair

Lean On Me, Or Not

I had this '97 Ford F150 in the shop with a P0174 code stored. The customer had the problem "diagnosed" by one of the parts stores as a bad pcv valve and hose. I imagine that was an employee armed with a code reader pulling a lean code and printing a list of suggested possible causes. A pcv vacuum hose leaking is a very common problem on a Ford but just a wild guess at this point in time as to the cause of the P0174. There is a lot of difference between diagnosing a problem and guessing what it might be..... but I digress. This engine has a fuel monitor O2 sensor on each engine bank. Passenger side is bank 1 (with cylinder 1,2,3,4) and driver side is bank 2 (cyls 5,6,7,8). P0174 means the trucks computer has determined bank 2 to be running a lean fuel mixture. Too much air or too little fuel, in other words. If I had a lean code for both banks, P0171 and P0174 then I might be thinking vacuum leak too. A pcv hose pulling extra air will usually put both banks lean at idle and it will be stored in freeze frame as a problem at idle or low rpm. I decided to not immediately look at freeze frame but to first look at the live data from both upstream O2 sensors. Now, most of the time a lean, or even a rich fuel condition code sets because the oxygen sensor is working fine. I mean the job of the sensor is to report the amount of oxygen in the fuel mixture. Don't kill the messenger, right? I even saw more than one tech site that said a lean fuel code won't be a bad sensor. Well..... never say never.

I captured the data from both upstream sensors. On the top half of the first picture is the sensor that did not set a lean code. The lower half of the pic is the O2 that did set a lean code. This is a live datastream with engine running at about 2000 rpm. An oxygen sensor range is from 1 volt to 0 volt. 1000mv to 0mv. Anything below 500mv is lean, anything above is rich. A good O2 sensor with a good fuel control should switch well between the two conditions. Now since I didn't find an obvious problem with the truck, such as a bad pcv hose and since I see no particular problem with the way it is running or with anything that would affect only bank 2 I think it would be worth replacing the bank 2 sensor and running another data capture. 

After replacing the bank 2 sensor the next capture looks like this:

Notice how the sweeps are in a much wider range now even on the "good" bank 1 sensor. The bank 2 difference is very substantial. I haven't done anything to change fuel mixture other than replacing the bank 2 sensor. Now we have full fuel control and fixed a lean code without finding a lean condition. 

Hope the pics are viewable. 

Thanks for reading,

Kenneth Hayes

G&G Auto Repair

Jeep Liberty stalls at idle.

This Jeep would not stay running if you released the accelerator and tried to let it idle on its own. Just a slight throttle opening and it would idle at normal speed. So, the idle rpms were too slow. I'm going to make this one quick and simple, because it turned out that way. I like to pull up iac counts and engine rpms for a quick way to see if the throttle plate may need cleaning. Ideally the counts should be between 10 and 24. Higher and it probably needs a throttle cleaning.

This one does need cleaning.

After cleaning you can see the counts have dropped dramatically.

I said 10-24 is ideal. A idle relearn and all is good. 

No stalling. Good numbers.

Thanks for viewing!

Kenneth Hayes

G&G Auto Repair

G&G Pops the Hood Part 2

In part 1 of this series we got the hood open and secured. 

Depending on your car, there are a lot of things to take in. Look around a bit. You can't spot something wrong until you've learned what right looks like. Hopefully by the end of this series you'll be so familiar with your under hood inspection that any problem will be obvious. 

Now we have to start somewhere. Checking oil level is a good place to start. A LOT of cars in the shop these days are low of oil. I know from talking to customers that some don't realize a car can need oil added even when it isn't time for an oil change. Usually the oil level stick isn't too difficult to find. Some are marked, some aren't. 

Ideally your car should have been sitting long enough the engine is cool and the oil has drained back into the oil pan. When the engine is running oil is being pumped to the various areas it is needed. After the engine is shut off it takes a minute or so to drain back to the crankcase. The oil level stick reaches down into the oil and has markings to indicate how much oil is in the pan. When you have identified your oil stick, grasp the handle and pull straight out. Plastic handles have been known to break so don't be over aggressive. Once you have the stick out and in your hand you should see oil on the lower end. Note the condition of the oil. If the oil is terribly dirty you probably need an oil change. Discolored and thick like milkshake could mean coolant is getting into the oil. If the oil doesn't look normal to you then you should let someone more knowledgeable take a look. For our purposes here though we'll continue with how to check the oil level. Wipe the oil off of the stick. There will be a full mark and a one quart low. Insert the cleaned stick fully into place then remove to see where the oil level is on the indicator. The truck below had an oil change a few weeks ago. The clean oil is hard to see in the picture but shows a slight overfill just at the last L in full. About a half quart over. 

Let's try another. The oil condition looks good in this one and it is about half a quart low. Don't add a half quart. Just remember to check later on. Some cars can burn a half quart or so quickly but take a long time to need a full quart. If you try keeping it topped off you would be constantly adding oil to a car that would be fine if left alone.

If you do need to add oil use the manufacturers recommended oil as outlined in your owners manual. Add oil as needed by removing the oil filler cap. A funnel can help prevent spilling to the exterior.

Stay tuned for part 3!

Kenneth Hayes

G&G Auto Repair