Q: The owner's manual of my 2004 Honda Pilot recommends regular changing of the "VTM-4" rear differential fluid, which costs more than $100 each time. Is frequent replacement really necessary since I seldom use the four-wheel-drive feature?

A: I checked the service schedules for your vehicle in my ALLDATA automotive database and was somewhat surprised to find the suggested rear differential fluid change interval under normal driving conditions is 15,000 miles. Under severe driving conditions, the interval is reduced to 7,500 miles.

In today's world of nearly maintenance-free drivetrains, the frequent change interval is unusual. But there is a reason: the limited slip clutch pack in the differential. Honda Technical Service Bulletin #04-040, dated August 2008, identifies a potential "noise or judder" from the rear differential when turning due to clutch debris contamination in the fluid.

This isn't an issue of four-wheel-drive use. The rear differential will always have a percentage of clutch slippage when turning, thus generating normal wear debris. The noise/judder can occur when the level of clutch debris in the fluid reaches the point of preventing the clutch from slipping smoothly. If this does occur, the Honda bulletin outlines a repeated drain/fill/drive procedure to remove the debris and eliminate the noise/judder.

Although I can't recommend this, I suppose you could increase the fluid change interval to 30,000 miles. But, of course, if you begin to experience the judder while turning, it will cost at least twice as much to eliminate.

Q: Ever since I had my 2006 Jeep Liberty in for a diagnostic test I am unable to fill up my gas tank without having to trickle in the gas. The pump clicks off right away like the tank is full. There's a lot of pressure released when I first take the cap off. I'm getting a "gas cap" message on my dash. What did they do to my car when the test was done? Is any of this related to the gas tank recall I just heard about?

A: No. The potential recall, which Chrysler is challenging, involves a concern over the plastic fuel tank's vulnerability in a collision and is not related to an emission system issue.

Chrysler Technical Service Bulletin #14-001-09 REV. A, dated September 2009, outlines a fairly simple series of tests to pinpoint the cause for slow refueling. The potential causes include a blockage or restriction in the fuel filler tube, vapor recirculation tube, evaporative canister or control valve, evaporative system integrity monitor (ESIM), ESIM vent hose or filter, or a problem with the fuel tank itself.

I don't see how a diagnostic test could cause this problem unless an evap system hose were left plugged or disconnected.

Q: I have a 2004 Chrysler Sebring with the 2.4-liter engine. It runs great above 2,000 rpm and in any gear other than at idle. At idle, cylinder No. 3 does not fire. It runs really rough and stalls the motor at times. When I step on the accelerator, it runs fine. It is great for highway use and runs really smooth at speed. Why just cylinder No. 3? The coil pack, plug wires and spark plugs are all new. A compression check showed it low at idle. I'm curious why it picks up when you step on the throttle.

A: Compression loss is a higher percentage of total cylinder pressure at idle speed. Slower piston speeds at idle can allow more compression loss than at higher engine speeds. Thus the cylinder builds compression until it reaches a level that supports combustion. You feel this as the engine "picking up" that cylinder and running smoothly at speed.

Try disabling the spark and fuel to the No. 3 cylinder, installing a compression gauge in the spark plug hole and start the engine. I think you'll find that "running" compression at 2,000 rpm is measurably higher than compression at idle.

Is there a fix? A valve job might correct a leaky intake or exhaust valve but won't solve a cylinder/piston/ring issue.