Water pump over-speed... cavitation... is there one that can spin 10,000

[nvrstuk]

Working on a SBF which needs a larger dia crank pulley but the water pump pulley is in the way.

Currently the water pump pulley is physically preventing the "enhancement" of the crank pulley so how much dia can be "taken out" of the wtr pump pulley w/o over spinning it?

Crank pulley is 5.75" dia (appr)

Wtr pump pulley is appr 4"

Distance btw the 2 pulleys is 1.00"

Crank pulley needs to be 7.25"dia min to spin a different pulley in this accessory setup but the problem is I can't find a wtr pump that can spin 10K rpm w/o cavitation...


Currently with the STOCK pulley sizes it has now, I did some math and if the engine rpm buzzes up to 7,500rpm, the crank pulley is 5.75"dia and if the wtr pump pulley is 4"... wtr pump is spinning at appr 10,781 rpm. OK rpm#'s for guys with engines w/ cams and heads that allow those rpms. I used to spin these rpms all the time with my HP289 with solid lifters and nothing else back in the day.

For this project we need the engine spinning at a much lower 5800 rpm while having the crank pulley at 7.25" to 7.5" (min) dia with the wtr pump pulley dia at appr 3.5-4.0" in dia and with these pulley diameters and rpm the wtr pump pulley is spinning from 10,500- 12,014 rpm. Off the charts from what I've read. These aren't constant numbers like in a Nascar race or something but still...

Not an expert here but everything I read on water pumps, including Stewarts is that pretty much any rpm over 7,500 rpm is inviting cavitation and other issues...

Electric WP is possible but their lifespan seems to generally last about 1/4 as long as a mechanical... at best, so that is not a good option. Also, EWP's generally fit over the timing cover to keep it's use as an idler pulley in place and they mostly the same dia as what we have now and you can't machine down the EWP's pulley because it covers the electric motor! I've never used an in-line one but that could be an option but the fact is they still don't last very long...

How fast can a mechanical water pump spin for short periods of time w/o self destructing?

Thanks

 

[Torkman66]

Maybe Change to an electric water pump and discard the pulley all together.

 

[Broncobowsher]

That is a lot of RPM. The 289 K-code was famous for having fewer water pump vanes for high RPM use, but I don't see that working here. Even thought of NASCAR and sustained high RPM, but I am sure the water pumps are under driven to a manageable RPM. I don't know if you can get the impeller tip velocity down to a manageable speed and still have a respectable flow.

Next thought isn't the pump itself. It is just belt velocity. How fast can a belt fly around and stay in control?

Back to the pump. Electric water pumps, whole engine ones, have existed in the OEM for several years now. I don't have high regards for the generic aftermarket electric water pump. I feel most of them (and what you probably have experience with) are built to a price point. Brushed electric motors just don't last forever. There are a few nice high end pumps run with brushless motors that I feel would be good choices. Without actually searching for them again, EPM, Davis Craig, Stewart come to mind. Probably butchered them. I have done a couple remote mount pumps that were really nice units, but they came with a healthy price tag as well. For you, I would not look for engine mounted pumps. Remote mount pump that is optimized to be a pump and not package limited to fit on the engine.

 

[nvrstuk]

Remote mount EMP is an option but only if OEM reliability comes with it.. I didn't start looking into remote mount until recently but going that route creates potential replacement issues down the road for proprietary fitment, local availability, replacement time figuring shipping, etc. But spinning a wtr pump at 25% faster than a stock one might not be the best option either. lol Nascar under drives their's so at track speeds they are in the 3k-4k rpm which is ideal for cooling and same appr rpm this setup would have cruising with 7.25" OD crank pulley.

This isn't Nascar as the engine rpms for this application would be between 1,800-3,000 rpm 90% of the time... any sustained rpm's over 5,000 rpm would be for maybe a minute or less as this isn't a marine applicaation, hence the question if there really would be much potential pitting, flow issues or cavitation "problems" for that length of time.

I am not an on-line video guy but dyno testing at 7,500-8,000 rpm on stock crank pulleys and water pumps (I'm assuming) are creating some high rpm numbers. Probably higher than what I'm looking at. I am guessing the amount of time under those conditions is the issue. Maybe they are all undersized for testing?? who knows...

Ran some typical LS pulley sizes at 5.6 (wtrpump) and 7.56 (crank) at 7,000 rpm spin their OE pump at 9,000 to 9450 rpm... interesting. I don't know if the diameters posted are OD of the pulley including lip or at the ribs but ...

 

[ntsqd]

As long as the diameter referenced is all measured from the same relative feature your ratios should be unaffected by which it is.

Perhaps look instead at remote marine water/coolant pumps (Jabsco, et. al.) that you can set up a separate, sane, drive ratio for?

Or perhaps whatever it is that needs to run faster needs to either go electric or get it's own, dedicated drive system?

 

[73azbronco]

https://www.fcpeuro.com/products/bm...vZxOn5njtq5iDJRRYjamvSIClwwRrw1caAj5_EALw_wcB

pretty sure if it says bmw, it will last you more than a few years.

 

[nvrstuk]

Fact!

I will do some more checking on that one. GPM, what vehicle/model, etc

 

[DirtDonk]

And are you sure sure that the RPM statistics you read previously from companies like Stewart, are for water pump RPM specifically? And not just engine RPM?

 

[DirtDonk]

Speaking of cavitation and sustained RPM, once cavitation occurs, is the air hard to purge from the system? Creating continuous issues for awhile? Or, if in the case of the engine in question, it doesn’t spend a lot of time at that RPM and it’s still running at lower RPMs after that, would cavitation really cause a notable issue for this person?

 

[nvrstuk]

1st post is wtr pump rpm.

2nd post... that's why I"m calling Stewart pumps on Friday... I iwill post back.

Talking about posting back, I found another pc of paper with weights, yup, weights from my build on it from 10 yrs ago.

D60 housing w/ARB and 5.13 gears is 180# ea

Each D60 knuckle w/spindle & caliper bracket weighs 57# eac

Dana 60 calipers, kingpin parts and misc wheel studs weigh 32# ea

I'd have to look at my old build thread to see if D60 axles were figured in or not.

So we'd have to add whatever weight of the tie rods

D44 complete with tie rods but w/o radius arm brackets weighs 260 #

Since I found this note in with all the parts I'm selling I thought I'd post it up for ya!

 

[Broncobowsher]

Speaking of cavitation and sustained RPM, once cavitation occurs, is the air hard to purge from the system? Creating continuous issues for awhile? Or, if in the case of the engine in question, it doesn’t spend a lot of time at that RPM and it’s still running at lower RPMs after that, would cavitation really cause a notable issue for this person?

The cavitation doesn't make air pockets. They are steam pockets. They collapse and there isn't any air added to the system. The collapsing part is what is so bad, it is a laser beam of energy at a single point. That eats away the metal. The steam pocket isn't from the heat, it is from low pressure. A vacuum allows water to boil at low temps. Boat props will get cavitation, no heat involved. Heat makes it worse.

 

[ntsqd]

@jamesroney would be disappointed in me if I didn't point out that it is the same mechanism in fuel that causes vapor lock.

Re-reading the first post, I think that the way around this problem is to drive the water pump with a separate belt that has a drive ratio that the pump can live with. For example these two appear to need a 1:1 drive ratio:
https://marineenginedepot.com/johns...aft-mounted-pump-w-coolant-supply-10-24946-01
https://marineenginedepot.com/crankshaft-style-engine-cooling-pump-f6b-9
More remote pump options:
https://dieselpro.com/marine-water-pump
https://www.hardin-marine.com/c-116-water-pumps-and-accessories.aspx

 

[jamesroney]

Speaking of cavitation and sustained RPM, once cavitation occurs, is the air hard to purge from the system? Creating continuous issues for awhile? Or, if in the case of the engine in question, it doesn’t spend a lot of time at that RPM and it’s still running at lower RPMs after that, would cavitation really cause a notable issue for this person?

@DirtDonk I think you are speaking of water pump aeration. @Broncobowsher is talking about pump cavitation.

Coolant aeration in a SBF is a frothy, bubbly continuous issue that occurs when the impeller sees an air pocket, or is starved of liquid for a short time. It takes about an hour of continuous driving at steady speed, followed by an overnight cooling session to "condense out" the bubbles.

Cavitation is a much more insidious failure mode, where microscopic localized pressure events cause the liquid to "boil" at the interface, resulting in physical damage to the components. Cavitation is a peculiar failure mode, because various instigators compete for priority. So while cavitation would normally scale with temperature, there are often applications where MORE cavitation actually happens at lower temperature. The most infamous in Ford pickup land is the Navistar 7.3 IDI Diesel...which would cavitate at the cylinder wall, resulting in perforation of the block. This would happen more at cold temps, and actually dampen and lessen at higher operating temps.

@Broncobowsher is correct from a physics perspective. Cavitation does indeed scale with temp. BUT he is wrong from an engineering perspective, because cavitation does not always scale with temp.

Eventually, cavitation can result in sufficient "boiling" that the pump stops moving fluid. This is the propeller cavitation where the noise, grinding, and bubbles all happen, and go away

@nvrstuk is looking for an anti-cavitation impeller for his high RPM application...but of course he is creating the problem that he is fixing. The rest of the world sizes the crank to water pump ratio first, and other accessories second. I don't know what accessory he is driving that he needs a larger crank pulley, but the obvious solution is a different size sheave for the accessory that needs the "7.25" diameter. This is normally accomplished by optimizing the water pump pulley ratio on sheave 1, and then running an oversized crank pulley on sheave 2 for his high RPM accessory. But since I don't know what he is optimizing, I'm not here to judge.

That being said, an anti cavitation impeller on a high quality pump is more than capable of surviving at the modest 7500 RPM limit. As everyone here has said..."my 289 HiPo ran 7500 RPM all the time..." And so does every Honda S2000, with an 9K redline. And every CBR250rr motorcycle at 19K redline. But those are running 1/2 speed off the camshaft.

So yes, change the geometry of the impeller, run it at about 1:1 to the crank, add an accessory drive for your accessories, and you can graduate to the late 1960's trans Am technology. Just like every other Z28, Boss302, and AAR cuda coming down the track.

 

[nvrstuk]

Yup... down to trying to physically deal with 2 belts so the correct drive ratio can be achieved with the wtr pump & crank pulley.

Wtrpump issue is a headache when the entire water pump has to be removed & custom hose fittings to engine block "holes" must be adapted to route water if an in line electric pump is to be used.

If you use a typical electric water pump that sits in the same physical location as the mechanical water pump, those motors usually are too large of diameter to allow the correct pulley ratio - - meaning you're still not able to use the larger crank pulley.

 

[ntsqd]

Does the rest of the accessory drive system need the original coolant pump in place? I.e. does it need an idler there? If not, then it's a simple plate to put the cold coolant into the engine. That would be too easy, so it likely needs a pulley there even if it isn't driving anything. Neuter the stock pump's impeller and feed the coolant into the normal place, only now the OE pump is a only a coolant distribution manifold.

 

[Broncobowsher]

@DirtDonk I think you are speaking of water pump aeration. @Broncobowsher is talking about pump cavitation.

Coolant aeration in a SBF is a frothy, bubbly continuous issue that occurs when the impeller sees an air pocket, or is starved of liquid for a short time. It takes about an hour of continuous driving at steady speed, followed by an overnight cooling session to "condense out" the bubbles.

Cavitation is a much more insidious failure mode, where microscopic localized pressure events cause the liquid to "boil" at the interface, resulting in physical damage to the components. Cavitation is a peculiar failure mode, because various instigators compete for priority. So while cavitation would normally scale with temperature, there are often applications where MORE cavitation actually happens at lower temperature. The most infamous in Ford pickup land is the Navistar 7.3 IDI Diesel...which would cavitate at the cylinder wall, resulting in perforation of the block. This would happen more at cold temps, and actually dampen and lessen at higher operating temps.

@Broncobowsher is correct from a physics perspective. Cavitation does indeed scale with temp. BUT he is wrong from an engineering perspective, because cavitation does not always scale with temp.

Eventually, cavitation can result in sufficient "boiling" that the pump stops moving fluid. This is the propeller cavitation where the noise, grinding, and bubbles all happen, and go away

@nvrstuk is looking for an anti-cavitation impeller for his high RPM application...but of course he is creating the problem that he is fixing. The rest of the world sizes the crank to water pump ratio first, and other accessories second. I don't know what accessory he is driving that he needs a larger crank pulley, but the obvious solution is a different size sheave for the accessory that needs the "7.25" diameter. This is normally accomplished by optimizing the water pump pulley ratio on sheave 1, and then running an oversized crank pulley on sheave 2 for his high RPM accessory. But since I don't know what he is optimizing, I'm not here to judge.

That being said, an anti cavitation impeller on a high quality pump is more than capable of surviving at the modest 7500 RPM limit. As everyone here has said..."my 289 HiPo ran 7500 RPM all the time..." And so does every Honda S2000, with an 9K redline. And every CBR250rr motorcycle at 19K redline. But those are running 1/2 speed off the camshaft.

So yes, change the geometry of the impeller, run it at about 1:1 to the crank, add an accessory drive for your accessories, and you can graduate to the late 1960's trans Am technology. Just like every other Z28, Boss302, and AAR cuda coming down the track.

While the 7.3 IDI cylinder wall issue is well known, there is also a coolant additive the IH said to use to minimize/prevent the issue.

Another cavitation issue I have seen on small block Fords is the timing cover when mixing standard and reverse rotation water pumps. the high velocity water leaving the pump makes a sharp turn. I once had a used engine what was a mix of parts. Although the timing cover wasn't leaking, I tossed it as there was too much material missing. It was only that one spot. Otherwise it would have been a good cover.

 

[nvrstuk]

Does the rest of the accessory drive system need the original coolant pump in place? I.e. does it need an idler there? If not, then it's a simple plate to put the cold coolant into the engine. That would be too easy, so it likely needs a pulley there even if it isn't driving anything. Neuter the stock pump's impeller and feed the coolant into the normal place, only now the OE pump is a only a coolant distribution manifold.

That was the thought, with an idler.

 

[73azbronco]

Fact!

I will do some more checking on that one. GPM, what vehicle/model, etc

328, so 2.8liter motor, maybe run two.

BTW, my simple engineering degree says, you can't turn a water pump at 10,000rpm and expect much water movement. At most, specific race pumps can only work to 8,000rpm. Your only solution is to use pulley size to reduce rpm.

 

[ntsqd]

Can probably design a pump impeller to work at that shaft speed, but it likely won't work well when turning slower. And like most things, the more finely you tune the system, the narrower its performance range becomes.

Clearly we need a constant speed drive like some airplane propellers!

 

[73azbronco]

im thinking electric, two of them, and some switch turning second one on when hot