Material Selection for Blender Blade

[Bunt23]

I'm looking for some help selecting a material for what's called a Rushton Turbine impeller (if you do a quick google search, you'll see it). The application is to mount the impeller to a motor shaft and use it to blend soil and water to make a slurry to use for soil analysis. We've currently been using one that is 316L SS, and we find that the blade wears very quickly (after only 500 cycles, it had shed 5% of its mass). So I'm thinking we need a harder material, but I'm not a materials expert, so I was hoping for some help. As I said, it will be used to blend soil with water at 3450 RPM speed, and so it is a very abrasive application. Any suggestions on a better material to use? Thank you!!

 

[arunmrao]

Silica in the soil and presence of water , makes it very abrasive in nature. You could try Hardox 500/550 for your application. I am using a steel blade with WC inserts brazed on the contact surface. This will last long.

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[Bunt23]

Thanks! What about A2 tool steel? That's technically harder than Hardox 550, right? WC is even harder, but would the brittleness of that be a concern with this abrasive application?

 

[arunmrao]

Yes you may try, any of the available tool steel ,I have not . No WC inserts if properly seated and brazed do not fall off or get broken.

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[EdStainless]

I would suggest looking at a harder stainless alloy, such as one of the PH grades (17-4PH).
These could be fabricated from sheet material, welded, and then heat treated.
There are many other options, but that might be the most straightforward.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Bunt23]

Thanks, Ed. I should have clarified this from the start because your responses may be different knowing this: the Rushton impeller we currently use cost $800, and so we intend to design one ourselves with replaceable paddles. The idea being that we could buy a sheet of whatever material we end up using and waterjet a bunch of spare paddles out of it and then bolt new paddles onto the wheel head as needed. Kind of like this one, but more heavy duty -->

http://www.directindustry.com/prod/lightnin/product-24564-1281349.html

 

[Bunt23]

Would 17-4PH grade work for that idea?

 

[EdStainless]

Do you really want bolted construction for these rotational speeds?
You may want to consider blades that are either "L" shaped and bolted through the leg or "T" shaped (flat with welded fin on back) that could be bolted through the back leg. These would keep the bolts out of the direct impingement of abrasives.
I could see a design where the "T" shaped blades fit into a slot in the hub and were then used a pair of bolts for retention.
You could jet cut sheet, have the back tabs drilled and welded, then have them all aged to high strength.
You will likely need a fairly beefy hub to support these.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Bunt23]

I can't think of any other way to have replaceable paddles without bolting them on. We have also thought about the T-shaped concept you are describing. Bolts to be on the backside of the paddle so they don't take the brunt of the impact. Here's a quick sketch of what we were thinking. Just need to have a material such that tabs could be welded onto the paddles, heat treated for hardness, and then we could bolt the paddles into the wheel head. The impeller only spins in 1 direction, so we can ensure the bolts won't take too much wear if we place them on the backside of the paddles.

 

[arunmrao]

Foundries face the same problem. High intensity sand mixers for green sand preparation,, where the ploughs are constantly in contact with wet sand.

Hard facing the blades was not a long term solution and every week, it needed to be replaced.

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[EdStainless]

Bunt, that would work, but you need two bolts from each side in order to brace the blades.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Bunt23]

Ya I had planned on that.. just drew the first one too big and didn't have enough room to draw the second without ruining the sketch haha

 

[dgallup]

Turbine blades are replaceable and don't use bolts. Might be overkill.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[EdStainless]

I would design the hub with slots so that the blade fit into the hub a bit to help with support. If these were deep enough a single bolt should work fine.
You can bend the 17-4PH to sharply, don't try to make square corners or you will crack it.
After forming age the 17-4 at 1025-1050F for 4 hours. This will give good hardness but won't be too brittle.
When you assemble it I would use a non-hardening sealant to coat all mating surfaces. You don't want grit working its way in.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Tmoose]

Got some pictures of a badly worn impeller ?

 

[kingnero]

For abrasion (with low impact; low temp and low corrosion), WC would be the preferred choice, with martensitic stainless a good alternative. If you're looking at hardfacing, look for a Co-based filler.
I'm sure your local filler material supplier could offer you some samples. This would be a perfect test case, you could apply different fillers on the different blades and do a relative comparison.

 

[Bunt23]

Sure. Here are pictures of a new impeller, after 500 cycles, and then after 2000 cycles. Sometimes we will have small rocks in the samples we are blending (1/2 inch diameter, maybe?), and so I would be worried about shattering WC. I like the 17-4PH suggestion, and I think we could slot either the paddles or the hub, like EdStainless suggested, then weld tabs to the paddles and put 2 bolts through them to hold the paddles to the hub.

 

[Bunt23]

Excuse my super-non-artistic sketch, but this what I am thinking.

 

[Tmoose]

What would an acceptable life be?
I'd start by increasing the paddle thickness 3 or 4X.

Or, is the eventual blade bending a "good" thing?

 

[Bunt23]

Valid point. I think 10,000 cycles would be ideal... once the blade wears enough to where it looks like it does in that 2000 cycle picture, the mix quality of the soil/water slurry starts to suffer and affects our downstream testing. Maybe 10k is unrealistic, but increasing paddle thickness is an interesting idea that I had not thought of yet. Thank you!