Background: (skip these first few paragraphs if you just want the data)
For the first 5 years of its Lemons racing career (March 2013-March 2018), my '64 Dart used its original '64 225 block. I don't remember what brand water pump we used, but we probably replaced it when rebuilding the block in early 2015. That engine always ran pretty cool (when we weren't blowing head gaskets for other reasons). On the racetrack, with or without a thermostat, the coolant temp tended to stay around 180-190°. That block went to the scrap heap in 2018, so it is not available for further study.

For most of 2018-2019, we raced with a freshened-up 1973 engine. It had stock compression with a stock cam, and we installed a brand new Gates water pump with an 8-blade impeller. This engine always ran on the hot side: 210-230° at the track, usually dependent on how hard the car was driven. We never really figured out why, and then like idiots we ended up installing a big turbo on this same short block and water pump combo before our most recent race in November 2019. We also shot ourselves in the foot by not closing up the radiator support completely after shoehorning in a giant intercooler, so airflow through the radiator certainly could've been better. Perhaps most importantly, we ended up using a slightly smaller diameter water pump pulley (from a big block, I think) so we were slightly overdriving the pump. This smaller pulley was used because it was much narrower, allowing us to press the fitting down on the water pump shaft, shorten the shaft itself, and create more space for our new gigantic electric fan. (See picture here)

At this race, the coolant temp gauge would max out at 250° after about a lap at full tilt, but it would settle back down to 230° or so after babying it for a couple laps. This was with pretty conservative ignition timing and rich AFR even under boost. One of the many things we did while trying to bring the coolant temps down at that race was test the water pump flow. We disconnected the upper radiator hose, attached an extension so we could run it into a big bucket, and then watched the outlet with the engine running while two guys manhandled a 5 gallon fuel jug full of tap water to keep the radiator full. At low and mid RPMs, the water flow out of the upper hose looked fine. At higher RPMs (say above 3500 or so), the water flow cut off and took a few seconds back at idle to resume flowing. I assumed this meant the water pump impeller was cavitating at high RPMs, so I pulled off the water pump and used a cutoff wheel to remove every other blade (leaving 4 behind). While we never repeated the flow test after this, it had no effect on the temperature behavior on the track.

One change we're making now is replacing that smaller-diameter pulley with this shiny larger-diameter aluminum one, designed to underdrive water pumps on big blocks. It just barely clears the harmonic balancer on a slant. Hopefully it will prevent high-speed cavitation. But that's not what I came here to talk about today...

Block Measurements:
In my garage I have 3 forged-crank engines in states of partial disassembly. The hot 1973 engine we just pulled out of the racecar, a 1970 truck block that we're finally building up into a long rod race engine, and a spare 1966 (3-plug) engine out of a B-body. I took the following pictures of each water pump pocket. Using a lump of play-doh and another brand new Gates water pump, I checked the clearance between the back of the impeller and the back of the water pump pocket.

Water pump pocket for 1973 block (stamping 3W 225 4525). This one has a gap of 10-11 mm between the impeller and block.


Water pump pocket for 1970 block (stamping F225 T 1112). This one has a gap of 9 mm between the same impeller and the block. So about 3/8".


Water pump pocket for 1966 block (haven't scraped the stamped pad clean yet). The gap between the impeller and the very back of the pocket is 8-9 mm, very close to the gap on my 1970 block. But there's also a raised annulus (ring) behind where the impeller sits, and the gap to this surface is barely 6 mm (less than 1/4"). The location of the ring also illustrates where the impeller is centered in the pocket.


I think I've found the reason why this 1973 block always ran so hot even before the turbo and the change to the water pump pulley. Assuming my old 1964 block has the same cast-in annulus as this 1966 block, that might explain why it had no issues keeping cool. Smaller gap behind the impeller means less space for water to pointlessly circulate behind the impeller, although there's really nowhere else for it to go (the impeller inlet is at the eye, front and center). I'm not exactly a fluid dynamics expert, but I would theorize that a bigger gap here might drop the pressure at the impeller outlet (circumference) and contribute to cavitation.

I'm guessing this 1970 block will be a little better than the 1973 block (and the underdrive pulley will probably make a bigger difference at high RPM) but I'm considering attaching a round plate to the back of the pump pocket to block out more of this space. Just need to figure out a way to attach it that doesn't come loose later. It might be easier to attach to the impeller itself, like one of these FlowKooler Water Pump Discs*. But the plate I'm talking about would be much thicker and heavier than the disc described in that article, so I'm not sure it would be such a good idea.



*FlowKooler used to offer these round plates to attach to the back of your water pump impeller; it looks like they might not be available anymore. But from the pictures I've seen, these were really thin. I think the whole point of them was just to create a solid backplate, which makes the impeller more efficient in general, not to close up excessive clearance with the block.

_________________
Somehow I ended up owning three 1964 slant six A-bodies. I race one of them.
Escape Velocity Racing

https://www.flowkoolerwaterpumps.com/pr ... water-pump is this kinda what you mean.

Since you mention fluid dynamics, you might try imagining the water pump as the compressor side of your turbo.
They are essentially similar, so that may help you picture how to direct the flow.

_________________
O==\=/==O

"A mechanic is Somebody."
- Jim Preston

Interesting.

Some thoughts -

There's probably not much difference between 6MM and 10MM in terms of flow loss / dead space. Either gap would flow 'a lot' of water. If it was me...and I wanted to pursue the 'minimum gap' path, I'd try to get it near-rubbing tight. **I haven't thought about how the pump is fed or discharges but I assume from your notes that the gap is not needed for either.**

I bet different pumps have different vane depths...I doubt they are made that precisely to be all be the same.

You could attach a steel back plate to the impellers by welding it. A good tack on each vane would be strong. I'd submerge the front half of the pump in water during welding so the heat doesn't affect the seals or bearings.

My opinion....a 1:1 ratio is want you want for high speed cooling. I wouldn't think slowing the pump down for the sake of cavitation is gonna help. Some examples:

Most Chrysler Big block V8's W/O AC run a 1:1 ratio. These engines had 6 vane pumps and didn't seem to have high RPM cooling issues.

The average big block with AC runs a 1.45:1 ratio with an 8 vane pump. The pump is overdriven by 45% to help low speed cooling....yet you never heard of any cop cars or anyone else having cavitation issues.

Cavitation on Mopars might happen but I haven't seen enough actual evidence to be convinced. I'd be curious what a Bonneville racer in 1970 ran...those engines see high revs for a long time.

I'd be looking for sediment or left over stop leak from over the years plugging up the water jackets. That one block you show is awfully rusty. That sort of thing will really screw with effective cooling.

Thought provoking observations, Frank. Thanks for sharing. I would probably try welding a plate to the vanes to close the gap some, as others said. Nice idea about submerging pump body in water while welding, Greg.

Lou

_________________
Home of Slant6-powered fun machines

I'm no engineer, but the following article lists 6 reasons for cavitation, and all 6 mention suction.
https://www.enggcyclopedia.com/2011/11/ ... on-causes/

The problem may have been restricted flow into the pump, rather than too much room in the pocket.

while I have never been a Chevy fan look at the design of their water pumps, with the plate behind the impeller, no pocket in the block... how much clearance do they have behind impeller to plate?

also... I remember with the Jeep inline 6s where the old 258 pump (V belt drive) would fit the 4.0 (serpentine) and vise versa but every pump I bought for one, had a sheet in it to make sure that you got the right one for your application, could they have mixed up the impeller at the factory where the pump was made, and put an opposite rotation impeller on it? or maybe installed to the shaft upside down? Not like we have ever seen mis made parts or anything. I don't see where you had more than 1 WP on that hotter running setup.

I don't know that this makes a difference, but you did mention something similar to this... on my 78 Fury and on my son's Ramcharger (both small blocks) we bought "flow kooler" pumps, just to try and see if they live up to the hype....these pumps have a plate attached to the impellers. IDK that they do help or not, but these engines certainly don't run any warmer than they used to.... though in the case of my Fury, when I originally had to replace the pump, (bearings started squealing, wasn't yet leaking) I remember seeing that its original pump was an 8 vane/ but the replacement that the idiots at the parts store handed me, was a 6 vane.... and unlike older years, when I would go buy a water pump, where both versions were available, that time they tried telling me "there is no such thing as an 8 vane water pump for that engine" they looked up several years of vehicles with 318s and all showed them only ever having had a 6 vane water pump (I knew better, because of what I had taken off that engine at the time) but I tell you what.... that heat gauge was noticeably higher, while I had the 6 vane WP on it.... I replaced it again with that Flow Kooler pump because of this....the Flow Kooler, besides having the plate attached to back of impeller, it is an 8 vane pump which was my biggest reason for buying it.
seems that on the 6 vane pump, that the impeller was also a smaller overall diameter across from vane tip to vane tip than the 8 vane had been.

they made HOW many /6s? over HOW many years? They never seemed to have a rep for being a "hot running engine". I'd be looking more at the pumps that you are getting, plus the possibility of a blockage, bad T stat, than I would at the construction of the block....

I would be interested to know though, within the blocks that you have been looking at, the ones with the raised "shelf" within the water pump recess, vs the ones without... was one version a "HD" truck or fleet/police/taxi or industrial application originally when new, vs the other one not?

People seem to forget that the early 60's units with the ring casing in the pump pocket had one of the circular bakelite impeller style pumps
that would take up the space compared to the star style metal impeller we are used to now...

FYI. Food for thought for the arguement...

which probably would be NLA??? or is there a WP available, with maybe a thicker impeller to make up the difference?

I just happen to have the 6- and 8-bladed (Airtex) types side by side. The used one was on my 1974 slant, but I assume it's not original to the engine because it isn't painted. One difference is the 8-bladed one has a sort of bevel in the plate under the blades. The 6-bladed one has a flat plate. In both cases the blades are shaped to match the plate, and the distance between blades and plate looks to be about the same for both, though I didn't measure.

Well, yeah, you can't go to a dealer or NAPA or otherwise like that and get one of those pumps or impellers. But that's just one item on a very long list of Slant-6 parts you can't go to a dealer or NAPA or otherwise like that and get. There's plenty of NOS and NORS out there. Old-type Slant-6 water pumps with the circular bakelite or (more commonly) cast iron impeller are usually a frequent sight on eBay, like here or here or here (dig that price!) or (maybe) here, or (maybe) here. Or this Aussie's been selling these water pump rebuild kits with the old-type impeller for years; he's probably got a warehouse full of them. If you buy an old-stock water pump, mind the one-year-only 170 pump which has a ø 3-1/4" impeller, versus the ø 3-1/2" impeller used on the '60 225 and all '61-up Slant-6s.

Me, I prefer the older-type impeller, whether it be made of iron or bakelite. This preference is not based on any data or formal testing, just using the reason for the change (stamped steel impeller is cheaper to make) as a proxy. The stamped steel impellers work fine for most purposes, but the question at hand is about optimisation, not "most cases". I'd (still) like to see a more modern impeller design—circular like the original, but perhaps with appropriately curved blades, etc. Probably easier now than ever before to design and procure such an item affordably.

_________________
一期一会
Too many people who were born on third base actually believe they've hit a triple.

Or you could always run a Hyster forklift water pump....

https://www.ebay.com/itm/NEW-WATER-PUMP ... 0005.m1851

Is there something special about those?

_________________
2 Mopars come with Spark plug tubes. One is a world class, racing machine. The other is a 426 CI. boat anchor!

12.70 @ 104.6

No. It's just funny they sell them as forklift water pumps...then of course it's 'cuz the forklifts have Slant sixes.