Motor Age Garage: Switchbacks

Jan. 1, 2020
  Problems like what I had with some spark plugs are the easy kind. But then there are the bait-and-switch problems that make us look like we’re incompetent, and those are the ones that bother some of us the most.

Some problems and their solutions are kind of run-of-the-mill and take very little diagnosis. For example, my 2007 Taurus only has 50,000 miles on the odometer right now, but a few weeks ago I began noticing the nasty bite of a misfire under load when I was crowding the throttle at 45 mph on a certain hill. And then it started happening every time I drove up that particular hill.

Like this article? Sign up for our enews blasts here.

The spark plugs on this platform are slated to be replaced at 100,000, and I had only half of that on the clock. Mileage notwithstanding, it obviously was time for a tune-up, so I picked up a set of Motorcraft platinums, gapped them slightly wider than spec (my personal preference) and screwed them in on a Friday afternoon. While I was there, I plugged in a nifty little palm sized $70 scan tool I had purchased from AE Tools and to my surprise, I found that the code I got wasn’t a misfire code; rather it was a code for the No. 1 coil primary.

I didn’t know if I could trust that (DTCs can mislead), but I dropped by the local parts store and grabbed a Motorcraft coil pack just for grins. These coil packs usually just develop a dead tower rather than causing a misfire under load, but that didn’t mean the coil pack wasn’t bad. With the spark plugs replaced, the misfire was about 85 percent better, but it was still there. So the 3.0L got its new coil pack and everything was peachy.

Run-of-the-mill diagnosis No. 2 was a bit tougher. It was a 2001 Jeep Cherokee that would skip to beat the band for about 60 seconds after a 30-minute drive and a 20-minute hot soak. It consistently tossed a P0303 code, and from time to time, coolant had to be added, which was a clue that couldn’t be ignored – these Cherokee 4.0L engines like to crack heads and blow gaskets, particularly in the No. 3 cylinder area. 

I usually troubleshoot the skips-on-startup problems by starting the engine and then shutting it down while it’s still skipping. At that point, I pull the spark plugs (quickly) and see if one of them is steaming or sooty. If it’s steaming, coolant is migrating into that cylinder, and that plug won’t fire until it dries out, thus the temporary misfire, and the head has to come off. If a spark plug is sooty, an injector probably is dripping, and if pulling that injector reveals a clean tip, the injector has been mechanically compromised somehow at its pintle.

Injector tips aren’t clean unless they’re washed regularly, and a nozzle leak does that.  If all the plugs are sooty, the fuel leak is even more serious and usually a hard start concern is the case; for just a couple of possibilities, multiple leaking injectors can cause that, fuel pressure regulator diaphragms can fail and send fuel into the intake.

In this case, it turned out that the No. 3 cylinder was receiving a small dose of unwelcome coolant when the engine was shut down hot, and while the leak wasn’t a bad one, the owner wanted it taken care of.  The head gasket was seeping coolant past the fire ring into No. 3, and after having the head checked for cracks, valve jobbed and reinstalled with new gaskets and juices, the Jeep is smooth.

Those are the easy kind – but then there are the bait-and-switch problems that make us look like we’re incompetent, and those are the ones that bother some of us the most.

Windows Down
Terry, a colleague of mine, called one day to tell me his 2007 CRV had lost its cabin cooling and he was driving with the windows down, which, on a rainy summer day that is also hot, tends to be annoying. He dropped the Honda off a couple of days later, and we got right to work. 

To begin with, the compressor wasn’t spinning, which is usually the case in situations like this, but was it low on refrigerant? After identifying the charge with our Neutronics sniffer and connecting the big MAC refrigerant machine, we noted about 100 psi on both gauges. We concluded that the compressor’s inactivity wasn’t related to charge pressure. Now it was time to poke amongst the wires and stuff.

The underhood relay center cover legend has no words identifying any relay, only creatively drawn pictures in boxes corresponding to relay positions. Many of the relays are only identified with an open book and a big “I” symbol, but there are two of the relays identified with snowflakes. One, however, had the graphic of a fan and a snowflake, so our focus shifted to the lone snowflake relay.

While the fuse box icons leave something to be desired, the wiring schematics are beautiful and quite informative. This Honda has an HVAC Control Unit, but that particular box has no direct control over compressor operation – the PCM takes care of that.  The HVAC Unit does get an input telling it when the compressor is too hot, and it controls the blower motor and the doors under the dash. The compressor relay, like most relays nowadays, receives power to one side of the coil and to the common terminal, so with the engine running, the A/C switched on, and the relay removed, two of its five terminals should show battery voltage and two should show ground. 

One of the B+ power feeds would be muscle power (input to the relay) load and the other power should be a feed to one side of the relay coil. The first of the two grounds would be one from the PCM to energize the coil and the other would be coming back to the relay through ground connection on the opposite side of the compressor clutch coil. That ground turned out to be nonexistent, and when we checked the resistance of the compressor clutch coil (which is fairly difficult to access without removing the unit) we found it wide open. In these cases we typically replace the whole compressor – it doesn’t cost that much more. If the coil winding had been good, we’d have been looking at a wire harness or ground concern.

After placing a call to Ranshu, I worked up some numbers and gave Terry a call. He was more than happy that we had found his trouble and green-lighted the repair, but said he needed to drive the vehicle until the compressor arrived.

Interestingly, the next morning I got a call from Terry and he reported that his starter wouldn’t operate, but that everything else was normal. Because he was 15 miles away and I had no idea why he was having that problem, he decided to have a shop right down the street check it out. He called back and said the shop claimed we hadn’t plugged the starter relay back in properly. That was kind bum-fuddling, since the starter relay isn’t even in the underhood panel and we had never touched it. That relay is located in another panel inside the passenger compartment, but OK.           

The Piggy Bank
About the time I ordered the Honda’s compressor, I got a call about a 2008 Nissan Maxima with lots of codes and an inoperative dash panel, including A/C functions – that one had suffered the “piggy bank” syndrome, which is such a common problem nowadays that just about very busy Nissan-servicing shop has seen one. 

There is a little shelf in front of the information center screen where people like to throw change, and invariably some of those coins find their way through the narrow crack between that shelf-like trim panel and the display screen. Well, right below that shelf is the Unified Meter and A/C Amplifier with a coin-sized slot in it. This is a small fairly expensive box with an important circuit board in it. FYI: Metal money ruins electronics and shorts out cigarette lighter sockets. 

On this Maxima, we found two pennies in that expensive little piggy bank, and a circuit board that had been hot enough to burn a hole in the plastic housing. The replacement Amplifier comes from Nissan with a piece of thin cloth tape over that hole, presumably to keep coins out but still allow air to pass in and out of the unit, presumably for temperature venting. It was a pricey repair that only cost pennies to generate – so to speak.

The Sudden Rough Idle
Concern No. 3 happened to be a 2003 Mazda B2300 that ordinarily ran like a dream. It belongs to one of our maintenance men, and he pulled around to automotive with a perplexingly rough idle that had happened all of a sudden. 

Engine vacuum was very low (two to three inches) as measured at the manifold, but because the truck ran just fine driving down the road, I imagined chunks of carbon trapped under the EGR Pintle. Well, that didn’t pan out, because starting the engine cold with my hand on the EGR supply pipe, I felt no sudden wash of heat. In my mind, that eliminated EGR flow as a source of the issue. We checked for hits on Identifix for this one and saw that some shops had replaced the intake manifold for this problem but that was all the info we had. 

So we fetched the smoke machine, yanked the air inlet tube, capped the throttle body and shoved some of that pleasant lemon-scented smoke into the cavity to find a massive intake leak right behind the power steering pump. Removing the PS pump, we found a curious little plastic plug about 12mm in diameter had popped out of the place where it was supposed to live forever – and it was still lying at an angle against the power steering pump bracket. 

With an intake leak of that magnitude it was no wonder the Mazda ran bad at idle. Cleaning the plastic plug and its hole with brake parts cleaner and compressed air, we “plastic welded” it back in place in short order and now that B2300 purrs like an RX7.

Compressor Comes In         
So the compressor arrived, we identified the refrigerant once again and then pumped it out, first heating the engine compartment with the engine running so we’d be more certain we had it all.

Then the old compressor was removed, and once again we checked the clutch coil, which was totally open. Just for grins we applied current to the coil, which responded not at all, even if we tapped on the hub.

Satisfied that our diagnosis was solid and the compressor coil was compromised, we unboxed the new compressor and set it next to the old one for physical comparison of line attachments and mounting hole configuration (always a good idea). We went so far as to test the compressor clutch coil on the new one and found it solid in the ohms area, then we applied battery current and watched it click.  This unit came with oil in it, and we went ahead and replaced the desiccant bag, pulled the system down for 15 minutes, checked for vacuum loss, checked the specs, and juiced it up with just over a pound of the cold stuff from our light blue tank.

Now we would verify the repair, but when we turned on the A/C we found that the compressor still wouldn’t engage. With our mouths and eyes wide and a hearty What the…?, we went to work trying to figure out why, and discovered that it would work if we bypassed the relay. Furthermore, the relay had everything it needed to operate. Two powers and two grounds showing. That was comforting, to be sure. 

All the circuits were good. Could this relay have been bad previously?  Possibly, but it seemed beyond the pale that we’d have a bad relay and an open compressor clutch coil. Had somebody reached under the dash, found a bad starter relay, and randomly swapped it with the A/C relay? What were the odds that they would pick that particular relay?

What’s so interesting is that these odd little potted relays are made in the good old USA. One way or another, a $4 relay from Advance Auto Parts closed the deal. 

Have articles like this sent to you weekly by signing up for our enews blasts here.