Line Shaft Bearing Clearance for VTP in Molten Salt?

[UhWhat]

Greetings All,

First post here, so hello and thanks for reading! I've got a question regarding proper line shaft bearing clearance for a vertical turbine pump running in molten salt. We have an application in our plant, using a sigle stage VTP, supplying molten salt to a furnace and reactor. I've been working with an established repair shop as well as the OEM, to document all critical dimensions/tolerances to conduct a proper rebuild on these units. One area that really concerns me is the clearance on our line shaft bearings. The pump operates at steady state at 875F, 1100rpm, and was originally designed to deliver 2400gpm of molten salt (sg 1.8, Visc ~ 8cStks) at approximately 75ft. The pump is constructed entirely of 304SS (bowl, impeller, columns, discharge head, shafts,) and we're running segmented cast iron bearings, with a stellite 6 laser cladding on the line shafts in the bearing journal areas. The current bearing clearance is 0.007-0.008 inches diametrically. I've questioned the OEM and they confirm it's proper. But from my past experience, I'm used to larger clearances (.010-.030) for VTP's running water. Tolerances for teh register fits on the dischrge columns is .002 concentricity, so with 4 flanges, that will eat up the clearance in a worst case scenario. Does anyone have experience with VTP's in molten salt? What are your thoughts?

Thanks!

 

[UhWhat]

As a follow up, I forgot to mention we've had some recent failures, one of which was a seized lineshaft bearing. No indication of dry run, NPSHa issues, or vortexing. The pump mounting flange is severely out of level (0.238 inches across 30 inch flange), so we're working to relevel the tank. Beginning an RCA to try and determine the mode of failure, but we don't have the pump apart yet....

 

[Artisi]

My initial thought would be to modify the pump to an enclosed line-shaft arrangement, of course the expense of this would need to be balance against the ongoing maintenance costs and down-time of the current arrangement.

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.)

 

[UhWhat]

Artisi,

Thanks! However, there would have to be significant modifications to convert this pump, that it would be more cost effective to just purchase a new one, with enclosed line shaft. Not outside the budget if need be, but then the challenge is still lubrication for the bearings. With a pump temp around 875-900F, what would you use to lubricate bearings? I'm guessing it would require a complete lubrication skid, supplying positive lube oil flow and equipped with a cooler, to remove heat soak into the lube oil.......

 

[1gibson]

What is the pump length? Do the trig to find out where the bottom of the pump is (based on the 1/4" out of level at the flange) compared to where it wants to be (hanging straight down.) Then consider the bending on the column and the rotor. This may be a major part of your problem, especially if the pump is long.

I wouldn't worry about 0.002" concentricity of the columns "stacking up." If there was perfect concentricity, the bearings would not offer as much radial support to the shaft, you wouldn't want that.

Because the bearings are metallic, the minimum running clearances in API either apply (if an API pump) or can be used for reference (if not an API pump.) I would agree .007-.008" is tight, for diameter of rotating member 2" or less, the minimum API clearance is 0.010" *plus* 0.005" for services over 500*F.

In any event, get the pump level first and look for design changes second. Vendor may have experience showing that design to be reliable in a proper installation.

I don't understand the possibility of enclosed lineshaft; what would you use to lubricate at 875*F, and how would it affect the process?

 

[Artisi]

UhWhat, sorry missed the 875F temp. makes it a different story altogether.
What are the implications of increasing the clearance? - and agree getting the pump vertical is important and probably should be step 1.


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]

Without a doubt, level the pump before blaming clearances; this is the most overlooked reason for lineshaft bearing failure in VTPs.

You have a very tough service; is this pump truly designed for it?? I typically think of a cantilevered verticals for molten services, such as a Lawrence or Hazelton pump that don't have the lineshaft subjected to product.

If your pump is in fact designed for the service, then I would have to trust the manufacturer....initially. What type of bearing spacing do you have, column length, speed are you running, velocities through the column, etc would be good to know. While you're somewhat tight its not unheard of and may be necessary to control vibration.

Can the manufacturer point to other successful installations in similar services?

 

[UhWhat]

Gents,
Big thanks for the reply's and thoughts! This forum is really great. To answer some questions, the original data sheet makes note that the pump is indeed designed for high temp service in molten salt. Whether or not it's an actual API pump (VS1) I don't know. It's a heritage (Bingham-Willamette) Sulzer VCN, single stage. There's about 20 ft of line shaft, 2 sections. Bearing in the suction bell, and another in the bowl aft the impeller. Then there are 2 more line shaft bearings, about 86 inches apart, which is right on the limit according to API 610 spacing with a 2.75 inch line shaft running at 1100 rpm. The running clearances in the current version of API 610 don't apply to the bushings according to section 9.3.4.2 (11th edition). So I really don't have a warm and fuzzy reference. Discharge column is 18" ID, so at 2400 gpm we should be around 3 ft/sec velocity

I agree with everyones concensus to level the mounting flange...and thats in teh works right now. Pumps are mounted in a horizontal tank, and we're cutting the mounting flange off (also has some pitting corrosion) to replace and level. Stamped vessel, so it's an R stamp job....big fun!

Guys.....Again...Big Thanks!!!!

 

[DubMac]

Bingham made a great pump and was surely configured at the factory, but how many times has it been rebuilt, by who, and was it built back to original specs?? The bearing spacing seems a bit far a part, I would have expected 5' sections.

Also, what is the velocity through the column? Should be somewhere around 5-10 feet/sec. If it is much more than 10'/sec then you could have washout problems.