Missing water pump blades

My 2002 Boxster 3.2 that I got for my youngest son to drive on the track had a water pump fail. It never ran hot, just started making a rumbling sound after start-up, then after about 10miles near the destination the low coolant light began to flash. Returning home after topping off with coolant, temperature never went above whats is normal. The water pump is a little over 2 years old and less than 20k miles.Has the OEM black plastic impeller that is missing the tips off the blades. I spent a long time back flushing the engine and trying to catch the missing pieces, no pieces were recovered. I can only hope that the pieces made it through the engine and found a non-critical resting place in the heater core or radiators, or even a non-critical place in the coolant passage. The pieces should be relatively small compared to the passages in the cylinder and head. I fabricated some screens to catch any pieces if they return to the engine, or if another water pump happens to fail at some point down the road in the future. I would recomend this mod to anyone rebuilding thier motor or having the pump off for any reason. The metal screen is .250 welded squares with .025 wire diameter, the fiberglass screen is .125 squares with .008 thick strands ( gutter guard screen). Has been on for over a week, temps are as they should be.. Missing blades metal screen with fine fiberglass top screen fine mesh "gutter guard" test fitted on spare short block I've had sitting on a stand in my basement for years, hopefully I never need it.. spare short block showing coolant passages spare head showing coolant passages

For some reason a couple pictures didn't load, so they are below.. missing blades welded wire mesh with fine fiberglass mesh," gutter guard mesh" Boxster on track My 99 996's going head to head( me and my oldest son) I let him pass..lol after a week with new water pump with screens all temps fine

Definitely an interesting idea! That said, is there any data that shows the size of the smallest coolant passages?

Yikes, when my water pump went it just dumped coolant out of the bottom and the tank made a gurgling noise. Was two weeks after I bought it!

Quotes source post: 15930783

I don't know of any data anywhere, but from what I can see without slicing a head in sections with a water jet, the smallest passage appears to be the slots in the block deck to the cylinders, you can see them in the pictures. They range from .165 to .225, so anything that can block them will be caught by the fine mesh (.125)..

I wonder if there is a easy/good source for stainless mesh (like in reusable oil filter cartridge filters) to do this with? Perhaps a kit could be made for field testing these awesome debris filters?! I'll beta test!

So what is the theory on how the blades were taken out? How is the screen held in place? Spring tension? I take it the finer mesh screen is 'tabbed' to the coarser mesh?

I too would be concerned with the fiber/mesh softening and getting clogged up in there. Seems like a great idea though, thanks for sharing.

Quotes source post: 15930953

Yea, that's exactly what I was looking for, but couldn't find a source.

Quotes source posts: 15930953, 15931881

Try here: https://www.mcmaster.com/mesh

Quotes source post: 15931226

The sealed bearings in the water pump( think IMSB) started to fail causing a wobble of the shaft and impeller, the impeller contacted the case breaking the tips. V. Yes the welded mesh is held in with spring tension and tabs, the fiber mesh is laced/ tabed at the edges..

Quotes source post: 15931442

Yea I thought about that, but I think the fiberglass mesh will be fine in that environment. But this is a test and I will know in years to come.For me I think risk is worth not having water pump debris circulating.

Quotes source post: 15931906

The problem I found in searching this source and others , is that the fine mesh stainless steel or similar is a woven strand, NOT WELDED.The woven strand wire cloth tend to unravel at the edges, that's why I chose the smallest welded wire cloth I could find which is .250, and used the fiberglass cloth as the fine mesh (.125)

They do get lodged in the head. I have physical an photographic proof. Had a piece fall out when I had my heads off. Look up my "Dragged me home one of these" thread and you will see the picture of the piece that fell out of the head when I was lapping the valves. The piece was from a previous pump that the owner replaced some time before I bought the car with bent valves.

Quotes source post: 15932470

Yes, this was my main concern, and the reason I designed and install the catch screens. I wish I had thought about it and put them in YEARS AGO !!

I'd be really curious to see a teardown of the bearing(s) in that water pump, it would destroy it, but its done anyway. It failed really soon, a bit concerning for the rest of us..

Seems Jake tried making screens for this about 14 years ago. https://rennlist.com/forums/996-foru...with-pics.html " I did that at least a decade ago.. The issue with it, is it blocks ALL the coolant flow right out of the pump, and you have no idea that you've had a failure. Why is that bad? Because without a flow of coolant, there's nothing to get the temperature up at the coolant temp sender, which is located OUTSIDE the engine, in the front coolant console. That means you bake the engine, but you never even know it, because the gauge says the engine is normal temp. Lots of good ideas, end up being bad ideas when the whole cause/ effect is looked at in a big picture."

Quotes source post: 15932615

Yea, I have intended to bust it apart and take pictures of the bearings to show the forum. But haven't gotten around to it yet, it's only been a week or so since I have done this. The last count I had the 996/986 has 18 "sealed deep grove ball bearings" (besides the infamous IMSB)....

Quotes source post: 15932628

Well, I'm glad Jake has thought about this catch screen, but he doesn't sell parts/mods anyway, so they would just be used on his motors, or sold through LNE, and there is really no money to be made on them. But I dissagree that it's a bad idea. I don't think enough debris can be created by the broken blades to TOTALLY BLOCK the coolant flow in that large area. If the bearings start to fail, I think the water pump will leak before ALL the blades are broken off or enough to TOTALLY block the coolant flow of even the smaller passages( and flash the low coolant light.). Plus I have oil temp gauges and I think the between oil temp and coolant temp, even with reduced flow, I will know when the engine starts to overheat..Plus there is some movement of coolant even without a water pump through "convection" .The first IC engines had no water pump. The water was circulated only through the principals of hot water expands and rises, and displaces the colder water. I have installed "block heaters" on Porsches for customers who were headed up North, and with a block heater, the temperature guage would rise as the water warmed with no circulation from the water pump.I will NOT be removing the heads or disassembling the engine to remove water pump blades..

Quotes source post: 15932615

Found time to bust apart the water pump, and to my surprise there is a deep grove ball bearing AND a roller bearing in there. I was expecting to see two ball bearings, but the ball bearing is to the pulley side of the shaft, and the roller bearing is to the impeller side of the shaft. You can see in the pictures the outer race/sleeve has fatigue pits in it where the ball bearing runs. The outer race for the rollers looks fine but the rollers themselves don't. Shaft/outer race-sleeve/rollers and ***** closer view of damaged rollers and *****, they flew everywhere when I busted it apart, I'll be finding them for a year.. outer race deep grove ball bearing side, you can see the fatigue pitting outer race roller bearing side, has damage at the inner and outer running edges shaft: the brass at the top is the center of the impeller(plastic impeller broken off, then seal area, then roller inner race, then the groove for the ball bearings OEM part number

There was a big thick seal behind the impeller, but there was no lubricate that I could see in the bearings, there was also a lot of surface rust which should never be in there.I'm thinking there were never any lube in the bearings, or very little. stacked a few roller back in couldn't get any of the ball bearings back in the heart of the water pump

FYI just happened to be in PET2 and looked the water pump up - they are up to revision 57 (supersession: 51, 54, 55, 56, 57) with the latest part number: 996 106 011 57 Wonder what the differences have been...

Good job! That mesh will have some restriction to the coolant system but hopefully not enough to make a difference.

Quotes source post: 15938645

I honestly don't know what changed are on the higher revisions,we can only hope they are a better pump, not just a different supplier.

The Pierburg water pump I installed has fins that are light brown. I think the original one I pulled out had the same fins but I can't be certain. I was under the impression that Pierburg was the OEM supplier. I have another replacement on hand for the future and it has the same fins, but it doesn't have a Porsche part number on it, just their own.

Quotes source post: 15939677

I was wondering not just about pump quality, but the tension on the serpentine belt. Seems pretty darn snug to me. I replaced mine with the Continental version and hope it is the correct length. Can't deflect with my thumb at the pump. Is there a value for tension? If so is there a tool?

Quotes source post: 15939647

Thankx !! I have been driving it pretty much every day, no difference seen..

Quotes source post: 15939925

There is no data on the tension. It is controlled by the spring and belt length. I'm sure you could use a standard belt tension gauge, and get a value off a new serpentine system with new spring and correct belt to get a value, then use that to value to evaluate used systems, but I don't know of anyone who has went to that much trouble, even the Factory.

Quotes source post: 15939677

That's quite interesting actually, so even with a supplier change there is a revision?

Quotes source post: 15940580

I have been told that , but haven't seen it "officially" in print from Porsche, so I take it with a grain of salt. What's also interesting is I was told revisions always start with 00, why on earth would Porsche need/ want to revise an M96 water pump 57 times?

Quotes source post: 15940667

I found that confusing as well, didn't think there would have been 57 revisions. Also the revision numbers jump as well it's not an exact sequence. Maybe someone who works/has worked in the parts department can clarify! Although it's probably not of interest to anyone but you and I! Haha

Quotes source post: 15940675

Right!! Inquisitive minds want to know! Porsche knows the allure of their " mistique" so probably do these things on purpose.

Here is some pictures of the water pump that was taken off a few years ago. It has the brown plastic impeller, turns good, no broken blades, appears to be OEM supplier, has part number but no revision number, has 17w08 stamped on the pulley, has 007090563 cast on the back. I busted it apart just to see what it looked like on the inside. I think with the screens installed this Boxster will only get new pumps from now on AFTER they leak, the screens will catch the blades brown plastic impeller part number 996 106 101 but no revision number bearing came out intact, you can see the bearing seal still in place. of course the plastic impeller and large thick seal behind the impeller broke to pieces when I drove the shaft out lifted the bearing seal, black thick grease, bearings turn smooth, no rust.plastic bearing cage holds the roller apart 007090563 09 cast in the back side of the pump 17w08 stamped in pulley

Wouldn't just putting a -12AN dry sump pickup strainer/filter in the main line to the radiator help with the little stuff? And keep the radiator from getting plugged with schrapnel? Canton and others make them,, they flow gallons per second.. and have a fairly tight mesh. cant imagine the coolant would bother them, stainless strainer..

Quotes source post: 15942224

I really wanted to catch the stuff before it goes into the engine passages where it can block a passage into the cylinder area or one of the passages around the exhaust valves in the head. Anything smaller than .125 there isn't going to be enough of them to stop up a radiator, I think they will find a non-critical area to rest in. If you look at his picture below, you will see a big round welch plug in the oil sump area. This welch plug is where the coolant flows from the right side of the water pump, down then over to the bank 2 case half. There is some dead space/non-critical area there that small debris can sit. There is also some dead space at the flywheel side of the engine behind the 2 bolts cap/plate on the case coolant passage. Lots of places small debris can sit without causing any problems, I just want to keep it out of critical areas and from blocking a smaller passage. The large round welch plug is sealing off the unused part of the coolant passage that comes down from the right side of the water pump, then crosses over to the bank 2 case. Photo courtesy of Millennium Technologies If you have a source for a stainless steel fine mesh about .125 that is welded not woven, I would love for you to share it. I have some screens made for my other cars, but haven't installed them yet..

This maybe what you're looking for.... https://www.mcmaster.com/wire-cloth

Quotes source post: 15944289

Same problem as other sources, the fine mesh is woven not welded...

Quotes source post: 15938558

First thing I wondered too about the grease, I bet you had a faulty one with almost none or zero grease. I wish there was some way of easily adding a grease port to the water pump bearing. Be nice peace of mind to add some grease every oil change.

Quotes source post: 15942395

I see what your after,, now it makes sense,, I was just thinking of trapping whatever is in the system, Your trying to stop it if the pump fails BTW the other option is to get a cheap jewellery/dental spot welder and just weld the edges ,, the screen your looking for should be available as calibrated filter/sifter screen used to size small abrasive and test it to spec. it comes in 4" and 6" squares, google aggregate inspection screens. I haven't bought it in a long time but used to make carburetor baffles out of it. The other way to lock the screens is to use a piece that larger than you need, mark it with a sharpy and silver solder the mesh outside the line, then trim with scissors. Just leaving the edge. You can add a thick(er) wire during the soldering process if it needs to be more rigid.

Quotes source post: 15945397

That site does offer welded mesh. Called formable in the description. The smallest opening available is .218 or 5.5 mm. You could use the .286 size and overlay one piece on top of another and off set the openings. Welding smaller mesh may be too difficult to manufacture. Not sure..

Quotes source posts: 15946624, 15946852

Both of these options would undoubtedly produce a more substantial catch screen, but would be more difficult to make and require some tedious work to make. My old eyes are just not up that much small work,lol, I have to wear two pair of magnifier glasses for such things, and still end up with eye strain. A fiberglass mesh product is generally known to be high strength,resistant to antifreeze (ethylene glycol, propylene glycol) and motor oil, and is temperature resistant to boiling water (212F). It can be engineered to even greater strength, or to higher chemical resistance to harsher chemicals/acids, or even extreme high temps(900F+). Y'all have me second guessing my choice of fiberglass mesh, so I have sent a technical request to the maker for reassurance of the particular mesh I choose. I may want to make another selection for the fiberglass mesh. The .125 fine fiberglass mesh with the .250 welded wire backing was just so easy to work with compared to what it would take to secure a woven stainless fine mesh or the stacked/offset the courser mesh to catch debris in the .125 range. I do appreciate the input as I will be making and installing these on my other engines as the opportunity arrises..I simply don't like the fact that a simple water pump bearing failure can cause a cracked head and end up disassembling the engine, when even a newish OEM pump is not immune.

I simply don't like the fact that a simple water pump bearing failure can cause a cracked head and end up disassembling the engine, when even a newish OEM pump is not immune.[/quote] This! I had a brand new Peirburg wp fail after only 1000km. Luckily I noticed it was weeping out the bearing and swapped it out.

I've been considering going to a davies pump, used them on race cars before,, rock solid temperature control, more complex, but easy to monitor for function. I seriously dislike the reverse thermostat design,, may be perfectly acceptable to german engineers, but its counter intuitive that putting the thermostat at the return end of the loop is a good idea.

Quotes source post: 15948281

This! I had a brand new Peirburg wp fail after only 1000km. Luckily I noticed it was weeping out the bearing and swapped it out.[/QUOTE] Unfortunately the m96 has many design peculiarities that have little margin before failure; and they also have a habit of cascading into secondary and tertiary problems that are sometimes worse.

Quotes source posts: 15948281, 15948798, 15949633

This! I had a brand new Peirburg wp fail after only 1000km. Luckily I noticed it was weeping out the bearing and swapped it out.[/QUOTE] Good catch, glad you caught it before blades started to dissapear in the engine !! The Davies Pump is definitely an interesting option, you could remove the thermostat and the WP impeller (or just the problematic blades), if there was a good warm-up strategy, would be a nice option. Although installation space is a Premium, and it wouldn't pass the "stock appearance" test , I'm sure the 120 amp alternator could handle the extra amps. Yes the thermostat being on the inlet side just doesn't seem like a good idea to me either. It's not just a German thing, I have seen that arrangement on domestic cars also, and they seem to have more temperature related problems than the conventional designs.. EDIT:Ccome to think of it the domestic car I was thinking of was the 60degree v6 Ford, but it was actually designed in Cologne Germany, so it may very well be a German thing !! I can't think of any other right off that has the thermostat on the return/inlet of the engine. Unfortunately the m96 has many design peculiarities that have little margin before failure; and they also have a habit of cascading into secondary and tertiary problems that are sometimes worse. [/QUOTE] Very true, but there are also some very high quality material and processes that should equate to a very long lasting engine, provided the pesky little failures that can have a domino effect are avoided, but I think a lot of these are the results of poor control of quality. High quality material and processes also require high consistency !!

Purely observational ,,, but my experience is: You set the controller for a base RPM, its fairly slow but moves enough water to prevent hot spotting in the engine it speeds up as the temp increases, on the race cars it give a nice smooth warmup .. They are fairly small units, Even used once directly bolted in to the front of a 2.0 Ford Sports racer engine with a aluminum plate as a adapter in the original water pump location. Seldom needed to run the pump beyond about 3/4 capacity, the non fluctuation of the water speed makes the system WAY more efficient. The race car had no alternator we ran it on total loss just on a 650CCA battery and saw almost no difference in the capacity. So even the power draw wasn't too bad. The pump we used ran three full heavy race seasons with no issues.. The 2.0 gets run wound out the whole race,, so I'd guess its a decent simulation of the way our engines run.. This was on a very early version, 5 years ago, the car moved on...

Quotes source post: 15951156

Sounds like a good pump to use for a race car, but too complex for a street car. The simple catch screens to catch failed blades seems to be a better solution for street aplication. I've used electric water pump drive motors before on race cars, no A/c, no alternator, no P/S, no belt except a small cog belt from an electric motor to the water pump.A total lose system, saves a lot of power, but running on a street car would just move the power loss from the water pump to the alternator. The Davies Pump could be a more efficient pump than the conventional electric motor/cog belt WP drive systems, and possibly eliminate the worry over missing WP blades,but it looks so much like an electric Centrifugal Supercharger I almost expect Superpower from it..lol The Davies Pump just see3ms too complex, and overkill for street application for me. I will just stay with OEM style pumps and simple catch screens, but thanks for the thoughts and suggestion.

Been there, done that. Two issues: -Retaining the screens.. This is tough, as the velocity of the water is very high, and the screens want to dislodge. -Screens may catch debris, but if they do it's easy to block the flow of coolant to the whole cylinder bank, which is worse than a localized hot spot Due to the location of the CLT sensor, an elevation in head temperature is often not noted quickly, and blocking the flow of coolant at the pump will restrain the elevated temperature coolant from reaching the sensor quickly. We made screens years ago for this, but never got beyond the initial development due to these two major issues.Doing something as a DIY for YOUR consumption is a lot different than doing it for the market, or for an engine that's being sold. Just imagine trying to write instructions about how someone should install this mesh, and how they'd royally screw it up, and contaminate the engine with the wire after they botched the install. Horrible! Porschetech3, I made a support that would touch the bottom of the coolant passage to support the screen from underneath. I didn't like the screen supporting it's self, and I found that it would move some after a few days of high RPM usage. I actually used the same hardware mesh that you did, and it worked better than stainless mesh. One of my customers who makes medical devices did give me some awesome stainless mesh to use, but I could not get it to stay in place well at all. I hate to create a problem, by trying to solve a problem, so we never took any of this stuff to market. The LN Engineering test Boxster has been running the Davies Craig booster pump since 2008, using it is nothing new around here :-) I had the first EWP150 sent to the USA. I have used them in both Porsche and Subaru applications, with and without the factory T stat, and with and without the factory water pump in the equation. That said, the only benefit I have seen from these with the M9X engines comes from post- cooling capability on the track. Those that do back to back sessions really need this to combat heat soak. The LN Boxster would sometimes stay on the track all day long, with driver after driver borrowing the car, doing laps, session after session. The post- cooling was huge while sitting on the grid between sessions, especially with some higher speed fan tricks, and some developmental CSF radiators.

Quotes source post: 15954881

As I said, the Davies Pump looks to be a good idea for a race/track dedicated car, but overkill/out-of-place for a street car. I'm confident the welded wire mesh is secured enough to hold up to the most extreme coolant flow. It can't go anywhere, and the force would have to be enough to collapse the wire, then wad it up. Even then it would flow more than flat pieces of blades getting stuck.. As far as detecting temperature rise if/when a failure of the blades occurs, if the blades were left to chance, it would be harder to detect a localized hot spot that has been blamed to cause the head cracking, than it would be to detect a coolant low flow on a complete bank, or the whole engine after blades were missing, and take evasive action. I know you have stated many times that the missing blades getting lodged in the head was a major concern, and the reason to recommend WP replacement every few years or so, but with the early failure of the WP I had and others have seen on Pierburg WP's, seems this issue needs to revisited.

Quotes source post: 15954871

The install and setup is less effort to me than many spend on a stereo.. I've also installed it in Trophy trucks,, Really helps the temps on a engine that's revving up and down in a hot environment.. But I get what your saying,, its not "tech free"

Quotes source post: 15954909

Hell, I�m just happy because I�m no longer being called a liar in regard to these water pump impellers losing pieces. If others have noticed it, and took steps to address this, it must be confirmed. Amazing how that happens... Only took 15 years :-)

Quotes source post: 15954988

Yes, EXACTLY!! Thank you, that's a better description, the simplest solution is sometimes the best solution..

To me, it's the only solution.

Quotes source post: 15955094

You have seen more of the failed 996's than anyone. Even though I have been in a Porsche dealership as a Tech, the 996's that we saw were mostly still in warranty, or CPO cars, or had owners who were inclined to have all their work done at dealerships. Most of our work was electrical/electronic repairs, tires, brakes, alignments, maintenance, clutches, squeaks/rattles, some transmission and engine repairs, just anything and everything someone could complain about on these cars. All the cars you have seen were engine repair, so even though I have seen way more cars than you, you have seen more engine failures than I ever will. I do wish you would share more of the detailed specs that you have learned about these engines for increasing power/longevity, I am an old ex-racer who would use them on my cars and too old go through the things I did too learn on the platform I was racing and setting some IHRA world records many years ago. But I understand you not wanting your competition to know what you have learned, but they will never catch you..

You hit some very important points here, and I am glad that you did. When people say that I am a "fear monger" it's only because failure, and bad scenarios are ALL I have seen. The issues are what have driven cars to us from all over the planet. It's kind of like being a mortician, where every customer comes in stiff. What you said here about your dealer experience mimics exactly what I have said for years when telling my customers.Over the years I have employed quite a few dealer techs that told me the same thing about what they did while at the dealer. Actually as long as they play nice, that never bothers me. I have had direct competitors send technicians to my classes, and to date all of the major competitors sent techs to the classes I used to instruct for Worldpac. I even had the owners of those companies, one of which being our only real "competitor" attend the WTI class. At the time he was just getting into these engines, and he was really late to the game. I taught the same class to those people that I did to anyone else, and left the feelings out of it. And that's because they don't have the spirit to do so. Most of them are running around to different race tracks, wasting fuel, and tires, and living that part of the life. As my dad always told me "Son, a rolling stone will gather no moss" The only competitors that we have are physics, and the economy. The humans take themselves out of the equation.

Ok, have to ask... so with the �quality � OEM and Pierburg pumps failing at low mileage has anyone tried the cheap offerings on eBay. NEW ENGINE WATER PUMP FOR PORSCHE 911 BOXSTER CARRERA PORSCHE BOXSTER 1997-2004 With FREE GASKET Interchange Part Number: 99610601153, AW9475, BWP-9379, 131-2307, 180-9030 These pumps start st $33.95! And at that price you could almost replace them at every oil change and still save money😉

Quotes source post: 15956015

I haven't seen one, or know of anyone who has bought one. Maybe you could buy one and bust it open to see what's inside? Mainly the bearings and seal are the components of interest. There are a lot of manufacturers making pumps to fit the m96, but for that price point it's hard to see how they could make that cheap without it being inferior. Take some pictures and post them here. Maybe some other members have some take off pumps of other manufacturers they can dissect for informational purpose? I have already busted two open, one that had failed, and one that was replaced "just because"...

I finally found some fine mesh .125 (1/8) welded steel wire (.014) mesh at Ace Hardware and order it. It is welded steel wire, then galvanized (zinc coated) after welding. I will still use the course .250 (1/4) welded steel wire (.025) mesh that is also galvanized (zinc coated) for the backing to give the strength to stabilize the catch screen. I will be making these screens for my two 99 996's to be installed when the opportunity comes. https://www.acehardware.com/departme...-cloth/5763743

Some of the skeptical posts on here had me second guessing my choice of fiberglass mesh for the fine mesh of the catch screen. So I sent a technical request to the maker of the mesh, the response I got from Saint-Gobain ADFORS customer service/ marketing rep.? was of no help what so ever, so I decided to do some testing on my own. I wanted to take the temperature way above what would normally be encountered in normal use. I took a lid from a pickle jar and placed it on the stove top, added some cooking oil and the fiberglass mesh then turn on the eye. I monitored the temp with a laser thermometer. At 350F all was fine, at about 450F the plasticized coating/seal in the lid started to burn but the mesh was fine, I took it to over 500F but couldn't go any higher because cooking oil will self ignite around 600F. The fiberglass mesh was able to handle the 500F with no problem. This confirms my assumption that the fiberglass mesh will handle anything temp the m96 can throw at it. So for me, the fiberglass mesh or the welded/zinc coated .125 fine mesh will both work equally well. DISCLAIMER :: DON'T TRY THIS AT HOME, EXTREMELY DANGEROUS !!!! UNLESS YOU ARE A PROFESSIONAL RISK TAKER AND ALL YOUR INSURANCE IS PAID UP, LIFE, HOMEOWNERS ECT. Fiberglass mesh submerged in cooking oil At 350F all is fine At about 450F the plasticized coating/seal in the lid started to burn, but the fiberglass mesh was fine

Can it handle heat cycles?

Quotes source post: 15966052

lol, well that would be extremely hard for me to test, or maybe even impossible, but I figured you (or someone) would ask such a difficult question.I have thought about that, and it deserves a discussion/answer/,theory. And I find this type of discussion much more interesting/productive than discussing the pro's/con's of the fried egg headlights of the 996's, or the best placement of the cup holders, subjects that have all been beaten to death... Heat cycles/thermal fatigue usually are associated with metals that have internal stress trapped inside after cooling. The internal stress can cause cracking if not relieved prior to putting in service. This is not a problem with GFRP ( glass fiber reinforced polymers) because they are very flexible and no internal stress is created in the process. GFRP degradation due to temperature can be a problem. The general study is the continuous temperature (or operating temperature) that a GFRP can withstand without degradation is about half of the "short term" maximum temperature. Since I have tested the "short term" temperature to 500F ( could be much more ) , then the general knowledge is the continuous temp (working temp) of at least 250F (possibly more) without degradation is extremely plausible. The Boxster has been working fine, no changes in temp. I am pleased with the outcome of the temperature test of the GFRP fine mesh that is installed. I am still under the opinion that the GFRP mesh will work equally as well as the welded steel zinc coated fine mesh. Just depends on which one is easiest to work with... Headed out in the Boxster to dinner, with no worries that the GFRP will melt

Awesome!

Quotes source post: 15968716

Thanks !! Ambient temp today mid 90's, coolant temp normal as ever with "catch screens" installed...

Got around to making some "catch screens" from fine mesh (.125) welded wire(27 gauge) still using the .250 (1/4) mesh welded wire as the main frame, since this is what I had originally wanted but had a hard time finding the welded wire in a small .125 mesh. The one made with fiberglass fine mesh is working fine and will remain, but these will go into my other cars !!

Any concerns with flow restriction?

Quotes source post: 16067235

The Boxster 3.2 with the fiberglass mesh screens installed has been working fine, no change in temps. From an engineering prospective, the general "rule-of-thumb" is 0.2USGPM per BHP of coolant flow. This means for a 996.1 (300bhp) at WOT requires 60GPM coolant flow, at 1/2 throttle 30GPM, at 1/4 throttle 15GPM. If I get real frisky, I'll take one of my left bank 1 case halves and set it out on my driveway, install the catch screens and a water pump. Figure out a way to spin the water pump at various speeds, then hook a radiator hose to the thermostat inlet and to a bucket of water and measure the flow with and without the screens..I think it will flow the required 60GPM, we'll see.

Quotes source post: 15940667

Maybe to track failure rate of different production runs / warranty claims.

Quotes source post: 16067574

That's way more work than I would do!

Quotes source post: 16067882

I hear ya on that !! At my age (63) and physical condition, I don't have the motivation I used to have. If I were 30-35 years younger I would show ya some innovation on these m96 engines. But it requires a lot of physical and mental work. I would like to see more people here do some experiments, there is a lot of smart people here, but we are all probably all older and less enthusiastic..lol Back in the early 80's I took a 96hp 4 cyl (2.0L T88-series SOHC non-Cosworth) German made Mercury Capri to 325hp (5.8 lb/hp) street trim. Nothing could touch it !!! (even today, no factory car under 100k could touch it)...Was my daily driver.. In the mid 80's I built a race tube chassis (1800#) and took the 96hp engine to 525hp (11k rpm's). Set some records with it.. Nowadays I just do little projects. I am re-powering my pontoon boat from a 70hp to a 150hp Black Max V6 ( rebuilding the 150hp 2-stroke), and will take me all summer. It took me a week to replace the water pump on the Boxster..lol