mark kibort

Thats comforting until you are the 1 in 100 that failed or 10 in 1000. The sample rate is important.
plus there are so many factors too. Like the failure cause.
1. some leak
2. some impellers come off the shaft and spin (plastic)
3. some impellers come off the shaft and migrate to the block
3. some bearing casings move inward to the engine
4. some make noise/bearing failure

of those failures, i wonder what the most common one is to see.

ROG100

Version 3 IIRC - had a metal impeller with an added bronze bush.

It has been about 4 or 5 years since Laso converted to the plastic impeller. I am not where I can check the quantity sold over that period but circa 2000. Failure rate of these pumps is less than 0.5%. The only failures I have recorded where leaks from the front weep hole. No other failure modes reported or known off.

I have sold one GEBA and it failed but taking into account Mark's stats I think I (the customer) was unlucky.

You race your engine and Greg makes high performance engines so all bets are of relative to warranty with the WP supplier. Greg sticks with what he knows and I respect that 110%. However most of my customers do not have bottomless pockets or race there engines and want a product at a reasonable cost with a low failure rate.

GregBBRD

I wonder how accurate all these statics are.

I certainly don't pay much attention to statistics...I go purely by what I see and experience, directly. And because I deal with such a large volume of these cars, I think I get a good idea of what happens in the real world.

When I get a car in that has had a water pump, belt, or tensioner issue, the failure never gets reported....the crap goes into the trash can and we fix the vehicle.

I replace an alarming number of aftermarket water pumps with very low mileage on them....alarming number. Migrating bearings, noisy bearings, leaking seals, etc.

I can't remember the last time I had a Porsche pump fail that didn't have 80,000 miles on it....

There's a reason I think that if no company had ever built an aftermarket waterpump, belt, or tensioning device and everyone only had the Porsche factory parts as an option, the 928 world, as a whole, would be hundreds of thousands of dollars ahead...and that reason comes from observational reality.

ROG100

I can vouch for my statistics because I keep the records. I know how many pumps I sell and how many failures are reported. As I warranty my Laso pumps for two years (even more in one case) I know customers are going to tell me about failures.

I also trust what Mark said about GEBA and that my one failure was unlucky.

I also trust what you say even though you have no statistical backup.

I have sold probably 5 Porsche pumps in nearly 15 years. Porsche is very funny about Warranty of parts and normally will not warranty a part unless it was fitted by a dealer. I buy enough parts to navigate around that. Never had a failure of a Porsche pump though many others have.

dr bob

The inertia of the pump impeller isn't nearly as important as the inertia of the water mass it's pumping when you talk about rapid changes in RPM. The "load" includes that plus friction (resistance to flow) that changes exponentially with those same RPM changes. Regardless, the shaft shouldn't be spinning in the impeller.

dr bob

And I suspect that Greg keeps the statistics in the back of his mind at every decision. Maybe the most important is the impact it has on the customers' experience when something happens. It only takes one awe $h!p to ruin that.

ROG100

Absolutely 100% and he stands by what he sells.
So do I but in a different market category as I do not build the engines on to which parts are placed. That is why I keep statistics as I get asked about WP's, Timing Belts, tensioners, and all sorts of parts and I like to be able to give an honest answer based on my knowledge.

G8RB8

Drill it and pin it. Problem solved.

mark kibort

like the intermediate plate!

Actually, there are a lot of forces acting on the impeller. its own inertia, the inertia of the fluid, which i suspect has some cavitation properties , especially wiht quick acceleration of the rotating parts. its not a positive displacement pump, but it does a good job of the impeller absorbing the drive forces spinning it..... there is no friction, except in the bearings, but there is drag and i believe that is what you mean. the drag is the force that is acting against the pump impeller. the impeller imparts a positive momenium to the water. how quicky it does this is important, but the difference between reving a high HP engine and a low HP engine is almost immeasurable. certainly driving the cars, its even less of a factor, as the rates of rotational acceleration are quite small vs when the engine has the entire mass of the car to move. (vs neutral ) and i would think you would almost hit the rev limiter before the gas pedal hits the floor in most cases, so , thats not a smoking gun comparatively.
the main problem is how the bearing wears under High RPM and how the impeller and bearing casing act under axial load, based on flow rates. not to menton viscosity differences of different systems. 50/50 vs pure water.

If you have ever spun a water pump with the feed line open and engine running. (Tstat pegged open) .... a garden hose can feed the engine quite nicely at moderate RPM. power the water pump requires is not as much as we are alluding too, even at 6,000rpm.

G8RB8

Inertia is the highest load on the impeller. Overcoming it is it's job.

docmirror

Failure rates for something as simple as a water pump for a car should be lower than 1 in 100,000. There is just no reason outside of economics for a 1% or even 0.5%, or 0.1% failure rate. It's just annoying as hell.

QC breaks down into two general categories; Workmanship and materials.

The best machinist in the world can't do great things with pot-metal that is like butter at room temp. We are seeing a lot, a lot, a lot of crap metals(I can't call them 'steel') out of China, and India.

Next, if a fine, well grained carbonized piece of steel/Alum/Titanium is in the hands of a jack-wad machinist, it can be destroyed in a matter of seconds(Brad Webster can attest).

With these pumps, we're dealing with a corner case of a part which is basically a wear item. It needs to be changed every 40k miles(or whatever the latest belt interval is), and the rebuilders know this, so they start cutting corners with both materials and workmanship. A bearing can cost $1.11 per 1000, or $1.55 per 1000. While prices is not always indicative of quality, companies like ***, and KKK have a reputation to live up to. If a rebuilder can save 44 cents per unit copy, and they make enough of them it only makes sense that they would opt for the cheaper bearing. The same can be said for the casting, the shaft, impeller, bushing, seal, etc.

This is NOT ROCKET SURGERY! Use good materials, and machine it properly and the device will live a long and happy life. Does that mean we have to spend $780 for a Porsche branded pump? No really. I can make a decent pump from rebuilder castings for a few hundred bucks. And I would do it, if I could get some castings sent my way.

I learned about this first hand with Ferrari 308/Mondial water pumps. Ferrari wants about $2600 for a WP. It is not in the belt path, and it is not a catastrophic failure item. I did about 11 pumps, and as far as I know, all of them are still in service. Of course, those pumps have a very low duty cycle, as the cars aren't driven much. But I put in the best bearings I could find, top quality seals, I pinned the steel impeller to the shaft, dimpled the pulley as well. I sold them for $550 each, and it only took me a little over a day to finish one of them.

I swear when I take the pump off this S4, I'm going to pull it apart and make my own good one, then maybe I'll start a long chain of custom pumps with returns required for the next custom build.

We will see how this goes.

PorKen

Put a clear glass fuel filter in place of the coolant pressure sensor.

EG. Amazon, Harley filter



Except his mind is prone to making up stuff, too, unfortunately. Witness any number of technical threads where, when confronted, he justifies some silly theory of his with imaginary 'facts' and 'figures'.

Just in the last week or so he has discovered the magic of allusion, trying to sow doubt about my tensioner with less-than-subtle digs about 'aftermarket tensioners'. What's troubling is, especially in the case of the PKT, although he does it for most anything he does not make, it's not for the common good. It's just to make himself look like he knows something.

PorKen

That sentence makes no sense, but neither would using aftermarket fuel hoses that leaked...

mark kibort

Just coming back from the track on a very unusualy hot weekend at laguna (not a lot of high speed straights, so cooling is limited too vs other tracks) and the car hd some overheating issues. way too hot for my liking, so i had to back off for half the race. anyway, not to digress.............. the pump is still working well and not making the noise the pump made with the belt loose. i guess the noise was the loose belt hitting the idler roller. it still makes a little bearing noise that i can hear with a stethoscope, but you cant hear it when the engine runs by just listening to it normally. so, i guess its ok.

yes, its job is to move the water, (and its own mass) but because it is not a possitive displacement pump the, inertial value of the water is very low, even on a quick idle to high RPM blip. anyway, there is not much of a difference in the forces on the pump wth a high HP engine vs a low HP engine in this regard. in other words, the forces on the water pump do not go up any significant amount by being able to rev slightly quicker in neutral. in gear....... the lower gears all accelerate faster than the taller gears.. so one could then say that if someone spends more time in 1-2nd vs 3rd 4th, the 1-2nd driver would have a higher chance of spinning the water pump impeller off the shaft and that is not a likely condition......again, due to the moderate acceleration rates of the engine. the force at higher RPM moving the water is the most significant.

Having studied and examining these forces and analyzing failures for many years on industrial robotic electric motor systems . radial and axial loads orders of magnitude higher, With smaller shafts and lesser press fits. the problem is the force acting on the bearing axially, for which the design fails. this is strictly caused by the forces on the impeller to move the water and little to do with how "fast" it accelerates the water.

polecat702

During the many builds of the blower motor in my Pantera, we changed the pulley diameter on the water pump. The result was that the water was moving to fast thru the radiator to properly cool it. We went back to the stock pulley and it stopped overheating on the freeway. Also changed to a flow cooler water pump, flow cooler pumps are designed to pump more water at idle, and that fixed the over heating problem in traffic.