Cantilever or submersible?

[DubMac]

Finally I can send in a request for info instead of making snide remarks to others.....please no snide remarks.

Friend asking me to help select a scale pit pump for steel mill expansion (contact water, lots of tough solids). I don't have any details or specs yet, he is just drawing up. He has a cantilever design shown on initial P&ID.

My question is more of a general nature about deciding between cantilever design or submersible. During the mini-mill boom in the 90s I recall most users preferring submersibles such as a Flygt or similar; and that is what I would recommend. I don't recall seeing any cantilever designs, such as a Lawrence or Hazelton, on those projects. Wondering if there has been some change of heart, or new designs in cantilever technology.

Anyone have particular likes/dislikes between the two? My dated experience recalls the cantilevers being more expensive first cost and astronomical to repair if you couldn't do it in your own Maint. Shop. The submersibles on rails were much easier to get in and out of the hole for an inspection or seal changeout.

I'm guessing materials will be hard metal on wetted parts but maybe there is something new on the material end as well.

 

[bimr]

Yes, submersibles are cheap on the first cost. And steel mills are known for being cheap, so it is not a surprise they went for submersibles.

However, it is common wisdom that the submersible have a much higher maintenance and repair cost than any centrifugal. Yes, the submersible are easy to pull out, but then the money starts to roll.

My preference is to stay away from submersibles unless you have to.

What is the application.

 

[Artisi]

Depends on sump depth for cantilevered units - 1.8 maybe now-a-days 2 metre is the absolute max for the unsupported shaft. In deeper sumps it is possible to lower the bearing assembly into the sump to increase the maximum setting - provided sump full level is below the bearing housing. Try Weir - Warman or Toyo.
Alternately submersible- but must be designed with hard metal case and impeller - don't think Flygt would cut the mustard here - the only unit I am aware of that would fit the bill is Toyo from Japan

http://www.toyopumps.com/product-info.html.

Best I can do - over 15 years out of the pump game and many things have changed re manufacturers and I would assume pump ranges / models.

One other unit used in Australia in the 80's / 90's for a duty such as this was the Sala (Sweden) cantilevered Ni-hard, it was manufactured in Aust. for the Aust. market -- have no idea now if / where / who manufactures this unit (if at all).
Aware of Lawrence or Hazelton but can't comment - no exposure to either.

Further, I wouldn't consider that a submersible was any easier to withdraw from a pit than a cantilevered unit mounted on a removable base-plate, disconnect a pipe flange, 4 holding down bolts and power disconnect - lift it out, and certainly cheaper to maintain - standard electric motor, no mech. seals or anything below water level other than a casing and impeller and maybe wear plates.

Would like to hear the outcome of the investigation.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[DubMac]

Thanks to the both of you.

Sump is 10' depth so I guess that does make for a long, fat shaft on the cantilever. I just can't imagine why this would be a preferable design against a submersible; seems like it would be harder to manhandle with a crane due to its length. DOn't know why I didn't think of Weir; I'll give them a shout.

I now recall hearing about the weak link of the submersible being the motor; if it was a cheapo design, it would be repaired early and often. This is going to be a design build by a 3rd party firm supplying an entire skidded heat exchange package to the mill. Somehow they got the scale pit pump in their scope. Seems weird to me, but as you say Bimr, mills usually look at first cost first, and they want it cheap.

Thanks for the ideas and I'll report back on results soon.

 

[Artisi]

You can't cantilever to 10 ft., it is possible to fit a water flushed intermediate bearing but in a steel mill - would think trying to keep up water supply/ maintenance would be impossible.
I would look at Toyo -good robust pumps designed to tough duties.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[MJCronin]

What has not been mentioned here is the Achillies heel of the submersible pump design, maximum fluid temperature limits.(this can run ~150-180F for these pumps)

Submersible pumps must reject heat to the surrounding liquid. Exceed a certain limit and the submersible pump will shut itself down. The cantilever design can tolerate much higher liquid temperatures.

My experience has been that scale pit and other steel mill applications have very hot abrasive liquids pumped. Steel mills have no real idea of the temperatures.... Project managers, of course, say "assume something, I have to go to a meeting.."

Steel mill maintenance departments want their job to be easy. They like submersibles because you just haul them out of the pit and send them out. Cantilevers are much more difficult to maintain, but have realistic limitations on pit depth.

IMHO, the cantilevers are susceptible to vibration and accelerated bearing failure in this service because they pick up slugs of heavy tramp solids and are momentarily "out of balance". This "slug flow" vibration is made worse by a larger cantilever. Rugged submersibles have a small advantage in this service.

There are a lot of subjectives in this discussion.

I have always tried to use the experience of the steel mill client in this decision and tried to encourage the pump vendor to add his operating experience here.



MJCronin
Sr. Process Engineer

 

[DubMac]

The temperature is listed as 113 degrees F, which I thought was a bit low also, but thats it.

Based upon the 10' depth, I think I will heavily recommend the submersible; probably a Toyo.

Thanks for the suggestions.

 

[TenPenny]

Make sure you design for 2x100% duty, so that you still have a system when one pump is out for repair.

 

[DubMac]

Was considering to recommend 3 pumps w/ 2 running at normal service.

 

[Artisi]

I agree, 3 x 50% units rotating would be a good option.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)