Re: Replace high mileage coolant pump prior failure ? ... and some other 3-letter stuff (IMS etc)

So .. my soon to be 18 year old 1999 986 has 85K miles now. At 67K miles in 2010, I replaced the IMS bearing with the dual row LN bearing as well as the AOS, motor mount and clutch. In 2015, I had the coolant replaced thinking this might help the life of the coolant pump but now I'm reconsidering a preventative pump replacement. The recent Pano 20 year anniversary report on the Boxster suggested a 40K mile life for coolant pumps to prevent impeller fragments from blocking coolant passages. Apparently, the pump is also prone to loosing its drive pulley. So, what is the consensus on regularly replacing the coolant pump?

My LN dual row retrofit IMS bearing has 18K miles and five summers' worth of driving. Filter inspections at the annual pre-hibernation oil change shows very little in the way of metal and small amounts of what may be vario cam chain tensioner pad wear products(small orange reddish plastic bits flakes). I know LN have said that the retrofit bearing should be changed periodically but so far, I think the dual row bearing without a cover giving easy access for splash lubrication will see it survive at least as long again. Is the quantity of non-metallic wear products in my filter a concern?

Thanks for your comments ... here is a photo of the filter contents at the last oil change in Nov. 2016 ... 3 metal fragments (slender chip and flakes) at left and chain pad wear non-metallic bits at right. This photo is typical of what I've found in the filter for several years now.

Cheers,
John

PS: Still contemplating a new or recent Boxster especially after I came within a whisker of buying a 2016 Spyder a year ago. Not sure if it's just me but is there a general increase in interest in the early Boxsters? Maybe the advent of turbo 4's has something to do with that. Must say that I still really enjoy driving my forest green '99. The production date for my Uusikaupunki built car is 7 July 1999 and my understanding of model production years is that July was the last month of production for 1999's. If this is the case ... I'd be curious to know how many more first generation cars were built after mine. Any way of finding this out?

Unless you are terrible at recognizing a pending pump failure -- noise/odor of anti-freeze/leak -- my advice is wait until the thing sends signs it is time to replace. The original water pump in my 2002 Boxster lasted to 172K miles. The replacement water pump is at 136K miles. (The original water pump in the Turbo lasted around 120K miles.) The Boxster pump made a noise. The Turbo pump was quiet but I caught a strong odor of anti-freeze and took the car in for a check and the tech lifted the car up and spotted leak sign at the water pump.

Anyhow, I hate to think how much money I would have wasted had I replaced the water pump every 40K miles. (Around $1100 for the Boxster, a few hundred more for the Turbo. At 308K miles I would have done 7. Ouch.)

When you replace the water pump be sure you replace the T-Stat. These don't often fail before the water pump but they have a finite life. Trying to nurse a T-stat over the life of a replacement water pump is false economy.

Those plastic bits don't look scary. The light color suggests they are from the VarioCam actuator rails. The amount of debris is not enough in my layman's opinion to warrant a pretty expensive round of VarioCam actuator rail replacement.

You are of course free to show the pics to your trusted Porsche tech and get his feedback.

Thanks very much for that info. I think the dealer has the front engine cover off at each annual inspection and they haven't reported any signs of weepage at the pump nor wobbly pulleys. I'll try to remember the T-stat when the day comes.

You can actually see the water pump from underneath the car. The pump on my 2006 Boxster started to show some pink at the bottom of the pump at around 60K miles. My mechanic spotted it while doing an oil change.

PS: Not sure if an impeller vane break up would necessarily result in external leakage ... noise perhaps. Don't think there can be much forewarning of local overheating of the engine due to coolant passage blockage by vane fragments. The other aspect of car reliability that concerns me is age related (18 years) as opposed to the relatively low 85K miles. I was pleasantly surprised when we managed to drive to the PCA parade in Indiana in 2015 from Quebec without a hiccup. Then, last autumn, I drove to Virginia and back without any issues. I must be tempting fate by writing this!

Quote
Indybox
PS: Not sure if an impeller vane break up would necessarily result in external leakage ... noise perhaps. Don't think there can be much forewarning of local overheating of the engine due to coolant passage blockage by vane fragments. The other aspect of car reliability that concerns me is age related (18 years) as opposed to the relatively low 85K miles. I was pleasantly surprised when we managed to drive to the PCA parade in Indiana in 2015 from Quebec without a hiccup. Then, last autumn, I drove to Virginia and back without any issues. I must be tempting fate by writing this!

The impeller destruction I think comes from an owner ignoring a failing water pump and the bearing play with the result the water pump advances to a stage of failure the impeller blades make contact with the block.

The blades are intended to run with very little clearance between their outer edge and the block as any excessive clearance affects the efficiency of the pump.

Any wiggle/play at the pulley end of the pump means the other end of the water pump shaft that has the impeller pressed on it also has that much movement. Any movement is risking the impeller blades will make contact with the block and fracture.

If this happens the debris can interfere with the flow of coolant if the debris gets into say the radiators, heater core, oil/water heat exchanger, or the coolant passages in the block or head.

The debris is bad enough in the radiators, heater core, oil/water heat exchanger, but much worse if the debris makes to the block or heads.

In some engines -- I do not recall seeing any signs of these in Porsche heads but I have never cut a Porsche head casting apart -- the coolant passages have tubes that pass coolant from one area to another. The tubes are positioned so the coolant under pressure flows with some force aimed at specific areas of the head. This is to help remove steam bubbles that can form.

You can see this formation of steam bubbles for yourself using a shallow pan of water on the stove. With the burner temp control set to high watch the pan -- the water will eventually get hot enough even with you watching it -- and observe the water forming bubbles on the bottom of the pan. Gently move the pan about to swish the water around and watch the water remove the bubbles. Let the pan/water get hotter. The bubbles form faster and grow in size quickly. You will notice now you have to move the pan about with more vigor to get the water to remove the bubbles.

Same thing happens with the coolant in the engine block, heads, even though with pressure the coolant's boiling point is raised. (I've driven with the coolant temperature in my Boxster up to 226F with no problems. The pressure that develops is what keeps this hot coolant from boiling.)

The hottest areas of the heads -- around the exhaust ports/valves -- can generate bubbles of steam. Even though the entire cooling system is empty of air the coolant needs to flow to collect and remove heat.

The small tubes are arranged to "blast" jets of coolant at these hotter locations and prevent the growth of those bubbles which if left alone could combine to form steam pockets which can block the flow of coolant and result in localized overheating.

However, the temp gage may not register any real increase in coolant temperature because the coolant isn't absorbing the heat. Most likely one notices there is a problem when a head cracks and leaks coolant into the combustion chamber or exhaust port. But of course by this time it is too late.

Even if you replace the water pump prematurely you still need some ability to recognize a water pump problem as a water pump can fail at any time. What saves the majority of owners that practice this preemptive replacement of the water pump is that new water pumps don't fail all that often.

OK ... so here was me thinking that the impeller failures might be age related (impeller material deterioration) but a pump bearing failure leading to impeller rub would make more sense and result in external leakage. Thanks again ... very much appreciated.

The tech editors at Excellence have been on both sides of this issue. I wrote to what turned out to be the new editor to ask about the inconsistency. Below is my original email and his response.

From: Dan
To: "excellence@rossperiodicals.com"
Sent: Thursday, October 13, 2016 12:54 PM
Subject: RE: Tech Questions - November 2016

Now I'm confused. In the May 2014 issue you warn M96 and M97 engine owners to replace their water pumps every four years or 50,000 miles to avoid small pieces of broken blades circulating through the cooling system and potentially plugging small coolant passageways in the cylinder heads leading to localized hot spots and cylinder head cracks. But in the November 2016 issue you state that you have never seen the plastic impellers break into more than a couple of pieces and have never seen an unsuccessful fix of this failure or further problems occurring in the cooling system later on.

So, which is it? Is the risk of engine failure from a broken impeller blade great enough to warrant preventative replacement of the water pump or not?

From: Greg Hudock [mailto:excellence@rossperiodicals.com]
Sent: Thursday, October 20, 2016 9:49 AM
To: Dan
Subject: Fw: Tech Questions - November 2016

Dear Dan,
Thanks for writing in on the contradicting water pump advice.
I have to say, you had me confused as well. Having fifty two years of building cars (including the fabrication work required to really build them), if nothing else, I’m consistent. Variations will of course occur as you learn or when technology gives us better tools to work with, but general principles remain the same. We are, in a large part, defined by those principles: it’s who we are.
Check the authorship of the Tech Notes in the two magazines Dan. That wasn’t me. My first published work for the magazine was September, 2014. As is the case with the previous authors, if you own a business whose purpose would, in fact, include parts replacement, then the advice you give may be effected by that doctrine. Mine is not, and I stick by my advice as presented in my response in November, 2016. Mind you, if it were a different part, say the timing belt in a 944 for instance, my advice would be different; but not regarding the water pump in your car. As I have stated in the past, water pumps give us fair warning when they are failing: they stink, they get noisy, they drip water--plenty of signals. I will also stand by my statements that I actually have never seen the plastic impeller water pumps break down into dangerous pieces that might threaten the wellbeing of the motor. Now, admittedly, not all of my fifty two years of experience have been Porsche, but twenty five of it have been. Oh, and one more thing: I own a vehicle, powered by a German V8. It has 294,000 miles on it with the original water pump, probably a plastic impeller model, still in place--I kid you not. As I said, we learn and give advice on the basis of our own experience.
Sincerely,
Mark Robles, Technical Editor
Excellence Magazine

Can you not get a metal impeller water for the Boxster? I had a plastic impeller water pump
go bad on my other car. GM updated the part from a metal one to a plastic one. I think the impeller
didn't break apart, but was slipping on the shaft. Anyway, I replaced it with a metal impeller one
and don't have to worry about blades breaking off and causing blockage. I would like to do the
same on my 2001S when I go to change it. Didn't everyone change to a plastic impeller because
of manufacturing costs?

Quote
Larry Nakamura
Can you not get a metal impeller water for the Boxster? I had a plastic impeller water pump
go bad on my other car. GM updated the part from a metal one to a plastic one. I think the impeller
didn't break apart, but was slipping on the shaft. Anyway, I replaced it with a metal impeller one
and don't have to worry about blades breaking off and causing blockage. I would like to do the
same on my 2001S when I go to change it. Didn't everyone change to a plastic impeller because
of manufacturing costs?

The failure of the GM water pump was not due to the plastic impeller but due to a manufacturing problem.

As for manufacturing costs, the plastic impeller is probably more expensive to make. It requires a mold (or several) be machined while the metal impeller is just bent/formed sheet metal. The benefit of a plastic impeller is the blades can be shaped to avoid causing any cavitation. This has a couple of benefits: One is the pump pumps a known or calculated amount of coolant under all operating speeds. Remember the engine redline is (at least for my 2002 Boxster) 7200 RPMs. That's a pretty fast spinning water pump. Cavitation can and will substantially reduce the amount of coolant being circulated. At 7200 RPMs that is the wrong time to have less than the desired amount of coolant flowing through a hard working heat generating engine. Even if the engine was still kept cool enough cavitation can destroy more substantial assemblies -- ship propellers, water turbines, control valves -- than the water pump blades. In the car even if the blades are unharmed the block can be damaged by the effects of cavitation.

The (composite) plastic blades/impeller of the factory water pump are quite up to the job. The water pump in my 2002 Boxster lasted 172K miles and the blades were fine even then. The bearings were what failed. In the case of my Turbo's 120K mile water pump again the blades were fine the seals were what failed and required the pump be replaced.

Porsche doesn't offer a water pump with a metal impeller. There is at least one aftermarket water pump that has a metal impeller. The impression I have of these aftermarket pumps is while they have the metal impeller the water pump doesn't last as long/hold up as well as the factory pump.

is that, if bearing fails and the blades impact the cavity, they scrape away the metal that forms the cavity making any subsequent pump less effective. And those metal scrapings are small and lodge in places that get very expensive if they even can get cleaned out. Some high volume P-car shops thus suggest actually using OEM pumps as their blades are going to fail before damaging the cavity and when they fail they leave big parts.

Water pump stays put until failure. I put a reminder in my maintenance binder to replace the T-stat when the pump is replaced (or vice versa). Many thanks again for your helpful comments.

The amount of what appears to be VarioCam rail wear products is not too great but I might still have the dealer check the cam deviation at some point ... perhaps when it's time for the 90K service.