My new Denso O2 sensor arrived the other day for my Eclipse.  Decided to wait until Sunday, when I have all day free, before installing it.  It's a simple procedure, but those are the ones that turn into disasters sometimes, so I figured I'd leave PLENTY of free time!

After getting the car up on jackstands, I figured I'd first go over ALL the coolant hose clamps again, just to make sure...  Main coolant hose clamps were still nice and tight.  Upper oil cooler hose clamps were all fairly tight (just one was a smidge loose).  Lower oil cooler hose clamps were VERY loose again!  I *just* tightened those yesterday, and only drove 40 miles since doing it!  One of them tightened up and stayed that way.  The other would tighten, then I'd let go and let it sit for a moment, try again, and it's loose AGAIN!  Took about 5 rounds of that before it held.  Jumping ahead a bit here, but after the rest of the day's operation, I checked them all one more time, and only that one last pesky clamp needed tightening, and only a tiny bit.  Guess I'll keep a close eye on that one, but perhaps this problem is past me now.

The O2 sensor swap went pretty easy.  Hardest part is finding the right angle to crawl under the car.  Once I figured that out (and how to position a flashlight so I could see what I was doing), it was a no-brainer to pop out the old Mitsu sensor and put in the new Denso sensor (although, it WAS a tight fit given the oversized alternator that my car now has).  My previous experiences were only with Mitsu and Bosch sensors, with the Mitsu being the much better choice for reliability under heavy abuse (high EGTs, excessively rich/lean fuel mixture problems, etc.), but costing $65 instead of the $45 for the Bosch.  This time I'm trying the Denso, which I'm told is even better yet, but once again, costs an extra $20+ (so we're up to around $90 for a sensor)!  PERFECT fit, every bit as good as the factory Mitsu, in every way.  On my test drive I noted that it seemed to initially heat up a bit slower than the old Mitsu unit I had.  Idle and low RPM cruising EGTs were comparable, but the higher RPM EGTs were about 40 degrees (Celsius) higher.  Either way, it's within specification, but this would seem to indicate that the ECU is either giving me more timing, or the O2 sensor is making my closed loop operation a little leaner (maybe I'll get back some of that gas mileage I lost when the HKS VPC came out and the stock MAF went back in).

The old sensor was replaced because of my periodic episodes of having -.2 volts from the sensor during decel (and occasionally cruise).  That one would typically have -.04 during decel.  With this new sensor, I (so far on the 40 mile test route) always had -.1 on decel.  Much more consistent, and a number the old one never generated.  Should actually result in a richer fuel mixture than the old sensor, if the ECU sees the exact same thing my gauge does.  However, as stated above, my fuel mixture actually looks like it may be leaner!  So, I'm still thinking that something's just weird with the gauge...

The test route also gave me a chance to get some good oil temp while cruising at 70mph on US23 in MI.  When the new alternator went on the car recently, the oil filter had to come off, and therefore the car got an oil change at the same time.  PURE-Tuning no longer stocks Amsoil, so I had them use Royal Purple (their low end oil...  They also offer Klotz as mid-range, and some other brand I'd never heard of as the high-end stuff).  Oil temps with Royal Purple pretty much match up with what I used to see when I used Mobil1 in the car.  That means about 5-10 degrees (Celsius again, my car is all metric!) hotter at expressway cruise than with the Amsoil.  Not sure if that's good, bad, or doesn't matter, but found it interesting...

All in all, the day's operation went well, I learned some things, and I think the car's all set for awhile.  Might have to tighten a hose clamp here and there before a long trip, but other than that, all is well!  :-)