40k water pump replacement recommendation (?)

Excellence's 2013-14 Buyer's Guide (p.105) says Boxster and Cayman water pumps should be changed every 40,000 miles or so because of the possibility of broken plastic impeller blades; the pieces getting into the heads.

You got 40,000 miles on your new 981 already? That's impressive!

I've seen 15k mile old water pumps with bad bearings and 10 year old ones that spin smoothly. If it spins smoothly, I'd leave it alone. Easy enough to test when you do the serp belt.

Grant

Grant

gee-lenahan-at-gee-mail-dot-com

If it spins freely and quietly, leave it alone. Mine had a tiny bit of wobble in the shaft when I did it at about 65K but was nowhere near failure. I was doing other stuff in the area so decided to change it but would not have otherwise.

The bearings should not have any play.

I want to reinforce that it is my opinion these water pump impeller failures arise from owners leaving a water pump with play in its bearings and probably a noisy pump too in service too long.

The water pump impeller blades want to run very close to the casting to maximize the efficiency of the pump. There is not a lot of clearance as a result so any play is cause for concern that the impeller blades can make contact and fracture as a result.

I'm not so much concerned with the fragments of the impeller but with the loss of water circulation due to the pump losing the blades that of course pump the water. Coolant does not flow as much if at all if this happens. While the engine is still running and generating heat it is not being picked up and carried away by the now non-circulating coolant. Also, the coolant temp sensor is useless since coolant is not flowing past this sensor it will not reflect the bomb with the lit fuse that is the engine and its ever increasing heat load. If one spots a plume of coolant in the mirror it will take a minor miracle for it to be something not expensive possibly even engine trashing.

As I have mentioned before when I id'd the water pump was making noise and found through feel it had some play while the other accessory drives had none I didn't even risk a 20 mile drive to the dealer. Instead I had the car flat bedded to avoid a possibly catastrophic water pump failure with the risk of substantial collateral damage to the engine.

Engines even these in the Boxster can deliver a long and trouble free life but they need help. Taking care of a water pump is an example of the help..

But I too am in the camp of it ain't broke don't fix it. God only knows what my Boxster's service/running costs would be had I replaced everything that some believe need to be changed on time or miles vs. when they need to be changed because they clearly signal they need to be changed.

It may be a fluke but the original water pump lasted over 170K miles. That's 4x the number of miles that a 40K mile replacement schedule calls for. Also, the original water pump's replacement has 100K miles on it.

As bearings wear, their wear rate accelerates generally. They wobble, or they are wearing because the surfaces are no longer mated perfectly or...

Now, if a bearing truly had no play, it would not turn, because the ID of the race would equal the OD of the impeller.

So there is some play.

A good water pump spins smoothly and silently with ALMOST no play. But there is some.

I only bring this up because we can get into trouble with "i detected play and therefore i must replace it"- well, maybe not.

My experience over time is that many replacement parts are not as good as the originals. I have also foudn that quality varies, so if you have a good one. keep it.

The (replacement) water pump in Bruce's car after 1 1/2 years was crap.Ran like it had pebbles in it.

The original water pump in the 14 year old replacement motor was smooth as a baby's bottom. I left it alone.

Keep in mind its not hard to spin them once you get the access panel off its a 15 minute job.

Grant

Grant

gee-lenahan-at-gee-mail-dot-com

... the OEM pump is very good and sturdy.
The main reason I see for them failing is that the bearing wears and the seal fails therefore leaking and eventually seizing.
The main reason why this happens (IMNSHO) is because of cavitation caused by air in the system.
Cavitation is the single worst thing that can happen to an impeller (or a propeller) because it actually destroys the part.
Once you loose a bit of the impeller, there's an unbalanced impeller that starts to wobble and damages the bearing and the seal.
What many people don't understand is that the cooling system is a CLOSED system.
Because of temperature variations the level in the reservoir will rise and fall a bit.
Just because it's on the low mark it doesn't mean that it needs to be topped off.
Once you open the coolant cap, especially if the coolant is still hot or war you will introduce air into the system and that's what creates havoc, not only for the water pump but also for the plastic reservoir.
The factory and most professional shops create a vacuum in the system prior to filling.
That way there's absolutely no air and the coolant fill all of the tubes, channels and capillaries in the system.
Leave it alone!
Happy Porscheing,
Pedxro

Pedro Bonilla
1998 Boxster 986 - 311,000+ miles: [www.PedrosGarage.com]
PCA National Club Racing Scrutineer - PCA National HPDE Instructor - PCA Technical Committee (Boxster/Cayman)

---

Racecar spelled backwards is Racecar

"Racing is life. Anything before or after is just waiting" ... Steve McQueen as Michael Delaney in "LeMans"

"If you wait, all that happens is that you get older"... Mario Andretti

"Being second is to be the first of the ones who lose" ... Ayrton Senna

it was present while the other accessory drives had no play.

It was the absence of any detectable play at the other accessory drives and the presence of play, admittedly not much play, that helped convince me the water pump was bad.

That I guess for better or worse is my definition of any play. Any play detectable by hand testing and any play that is not detectable in the other drives but is detectable in the water pump is "any play".

As I mentioned before, the play was slight enough that I had to first try the other accessory drives then the water pump and did this more than once but the water pump was the only drive that had any play.

However, I was not condemning the water pump on play alone.

I would not have even been at the water pump had it not been for the noise - a low volume grumbling noise that got worse as the engine got warmer -- that prompted me to investigate for the source of the noise.

The first step was to remove the belt and confirm the noise was not coming from inside the engine. It was not. This then prompted me to look for an accessory drive with play. Sorry. But they just do not should not have play. I can't explain this other than to suggest while you have the theory that there should be play the theory has a flaw in it somewhere.

The condition of the new belt (just a few months old) added weight to the decision the play was not right. The inner belt edge was sharp from rubbing against the water pump pulley's side. The outer edge was dull and felt just like the old belt's two edges so I was comfortable with the feeling the sharp edge was another sign of a problem.

Last a close inspection of the water pump found anti-freeze residue on the casting below the shaft. Noise. Water pump play. Belt rubbing. Coolant residue. Water pump was replaced.

(As an aside I found out much later some dry residue is ok. But the techs told me there should never be any wet coolant present and even the size of the stain is important but would not tell me the acceptable or unacceptable dimensions.)

Now this was not the first time I was up close and personal with the belt and the water pump. While I can't recall it I am pretty sure -- it is just second nature to me -- when I replaced the belt just not more than a month or two before this I checked the water pump and other drives for any signs of play. Also, I checked the idler rollers and tensioner roller for any thing amiss.

And the time before that (the first time) when I replaced the belt I admit I can't recall what the water pump felt like then either. However, my training is that water pump plus play equals bad water pump so had I felt any play at the water pump I would have probably consulted with a tech about this and taken his advice. As it was there was no play so I just replaced the belt.

Let me speak about play too. I didn't use any measuring tool, dial indicator, to measure the actual amount of play. That play was present was determined based on feel. But as I mentioned it was not a large amount of play and I had the other accessory drives to use for comparison. But I was working on the assumption that no play was the norm, which I have to say again was the case with all the drives other than the water pump.

As another aside as a machinist I could spot run out of just a few thousandths of an inch on say a part chucked in a lathe. I had more than a little experience from having to chuck uncountable number of parts in the chuck over time. My employer considered it unfitting for a machinist to have to use a measuring tool to get a part chucked to within a few thousandths of an inch if the part didn't have to be chucked more precisely. But I couldn't resist at least double checking my work -- at first -- and sure enough by eye I managed to get the part runout down to just a few thousandths of an inch. My point is that by eye or by hand there is a limit to how much one can see or feel in this case feel regarding play.

But maybe had I more precisely measured the play of a new water pump and the accessory drives with a sensitive dial indicator might have found play below which my hands could not register. This might be the play you refer to, the play that you believe has to be there.

A deteriorating bearing often feels, "rough" and may make some noise (or if its bad enough, a lot).

You can feel smooth, and you can feel rough. Smooth is better :-)

Grant

Grant

gee-lenahan-at-gee-mail-dot-com

because it started to leak.