Questions You Should Know about cast iron Impeller

I have a discussion with some engineers regarding the selection of the impellers material for horizontal booster pumps that will be used to pump clean water, with the possibility to have small debris passing through the pump (like sand and silt).

The selection is between Cast iron and Bronze impellers, Bronze doesn't rust but cast iron does, BUT at the same time the impeller material must be chosen for both chemical compatibility and wear resistance so we may need to consider one of the duplex metals because most corrosion resistant materials are too soft for the demands of a pump impeller

From your experience with impellers material selection, what would be the best choice for the pump impeller? Is it bronze which is soft or the cast iron which will rust??

We are in the process of finalizing the technical specifications for the required pumps. Your inputs for the disadvantages and advantages of both choices are highly appreciated...

The largest and oldest pump in our refinery is a cooling water booster pump. It has a bronze impeller that has held up incredibly well. But, it is large and slow (low rpm). We have deep well pumps that are known to have some problems with passing sand. Bronze impellers did not perform well in this application. We changed to stainless steel (316) and the impeller life went up drastically. But, these well pumps are running at rpm.

The correct selection depends on the water chemistry (as already noted), the exact composition of any solids, the flow velocity, etc. If I had to pick a readily available, hard, durable material with good corrosion resistance, I would choose CA6NM (11-14% Chrome). But, for a very large impeller, this could be extremely expensive. If the velocities are relatively low, I would tend to go with bronze as my next choice. I would only select stainless steel (316) if there was an overriding need that could not be satisfied with the other options.


Johnny Pellin On all our pumps, both fresh and salt water, we have Duplex SS material. In terms of rust, they stand up extremely well and show no signs of corrosion or erosion. Why don't you leave the choice open and ask the pump suppliers what they would recommend and offer in their package? If the liquid handled is clean water with very little sand or silt, bronze generally offers the most cost effective solution. Bronze does not rust, is eay to cast and machine and generally offer superior surface finish of the waterways leading to efficiency gain. Duplex stainless steel such as CD4MCu will have superior erosion / corrosion characteristics but you will probably pay a much higher price for your impeller, if it is not an off-the-shelf item for your pump supplier. Also, delivery lead time for duplex impeller will be much longer compaed to bronze. You may select gun metal - BS LG2 and this will be the most cost effective. BS CT1 is more expensive but has a higher hardness and abrasion resistance.
If the sand content is negligible - bronze is the best answer.

It seems the cast impellar is generally considered to be better than the stamped steel ones.
Since the remanufactured cast units have had the long edges of the impellar ground down to make them look better/newer they are less efficient since this process increases the distance from the impellar to the pump body and thus allows water to bypass the vanes.
I can get new stamped steel pumps but no new cast ones.
It seems since the steel version has less mass/thickness to it it also allows more water to be moved. Of course this depends on the shape of the vanes.
Can someone explain why the cast ones are better???
Thanks
Darrell This is mostly my opinion, but...

One arguement is that the stock units were cast, at least if you look at the original manuals. I tend to think that the people that created these engines seemed to know what they were doing, based on their longevity and performance.

I'm not sure your remark about the cast ones being reground to look nicer is accurate. The last one I bought sure looked new to me, not only because it was pretty, but also it differed in slightly in design. The rake of the vanes were not the same as the original factory one I had, shaft it rode on was a different size, and the vanes reached a smaller diameter than the original. Plus, when I installed mine with a gasket on the benchtop for curiosity's sake, I noticed it fit very tightly at the vanes and their mating surface. The stamped steel unit, by contrast, was more of a paddle than anything else, that sorta fit in that spot. Also, if you notice, the vanes create a shape that maintains consistent volume across the travel. The unit I bought would slightly widen the path as it became shallower, moving from center to perimenter.

If you ever open a high volume pump for moving water or similiar liquids (ie, not hydraulic) you'll see the same impeller design. Same thing goes for biomedical grade pumps that are extremely accurate in their volume rates. Generally, any lab pump that costs real money has that design.

Bottom line, though, is that when I installed it, I noticed that the engine ran a little cooler at low rpms, and when it did get warm, it would drop dramatically quicker when the load decreased. And I think that tends to be the experience most members on this board would attest to.

Hope that helps.

scott I've seen NO difference in performance as related to engine cooling. I've used both. You could measure say with a flow meter, an electric motor with a watt meter for energy draw, and a glycol mix at operating temperature to accurately see a possible difference. IH spec'd 80 gpm at an engine rpm of 4K. Even then I'm sure the tolerance is fairly wide with 80 gpm being the minimum and that may account for subtle differences. It's rolling stock, not a race car. Impeller performance is based on the number of and curve of the vanes, diameter, and clearance to the volute etc. Brand X my have a wider clearance, but maybe for THAT particular impeller it works the same as a similar one that needs a tighter tolerance. It doesn't necessesarily point to quality; though tolerance issues could equally apply as in how well they're made. One point, I made a similar modeling clay deal to see the clearance to the volute. The old cast vane and the new steel one both had the same .062" or approx 1/16". There are too many variables to say claims of superior design as we never know the root before and after, a person's cooling system health and so on.

If a lot of guys like the cast, and a lot of guys like the stamped, it ain't the pump; jes' guys lik'in to jaw.

This reminds me of the old argument about how many angels can dance on the head of a pin!

My procedure? Buy a water pump. Install it. 150,000 miles later, replace it.

Who cares which is better? They all work well if your cooling system is in good order.

If this was the case I would not post it! I've got better things to do.
I like Greg's idea of building a test fixture to test the pumps capacitity at a given RPM. (Oh FYI my cooling system is as new as you can get with "everything" having been rebuilt or replaced, yes including the engine) But with what I would call marginal cooling. As for testing pumps, I even have an old block that's in good shape I could use to provide the proper test bed, but I think I going to buy an Airtek pump first and see how that works first. R290 has brought up a good point, running clearances. The pictures from the manual shows that both styles were used. What was NOT ever discussed in older threads was the FITMENT There could be a variance in assembly. Both styles work, but a new one not seeming to move enough coolant could be traced to a root cause in assembly specs whether a reman or new. Running clearance in any pump is key to where it'll place on it's curve. I misquoted my measurements. I looked up the notes I made back in '06 or so and they were actually closer to book spec, maybe the .062 was a s moment on the shim. If the clearance is large, then the impeller is pressed on too far. Seeing that the steel impeller can get ruined taking it off to start over, that's an oppurtunity to fudge it.

so the shims are not in correct i guess

There are no shims in the final assembly, they're for setup.

Edit... What was NOT ever discussed in older threads was the FITMENT. Both styles work, but a new one not seeming to move enough coolant could be traced to a root cause in assembly specs whether a reman or new. Running clearance in any pump is key to where it'll place on it's curve.

As you can see mine is not even close to spec. On a ScoutII the cross flow radiator prevents you from getting a good visual inspection of the flow, unlike a down flow radiator.

So all you folks living in colder climate and the heater is not working so good, might just need to check the pump impeller clearance. I've read a few of those post and thought it was the heater core, but maybe its the pump

Once I get either a new pump or fix mine I will post up. (anybody know where I can buy a new impeller? online would be nice.)