Magnetic Drive Pump Leak at Containment Shell

[jsasada]

Hi everyone,
We have experienced weak acid pump failure (again). This pump type is magnetic drive pump (Innomag C4 3x1.5x10) The fluid pumped is nitric acid 40% with temperature of 55 degC and differential head of 1700 kPag. This is the assembly of the pump:

The start up went well however we saw abnormal spike on the electric current of the "A" pump. Shortly after we saw NOx fume and decided to switch to "B" pump & it ran normally up until now. After inspecting the A pump, we found the following pics:

Containment shell worn out due to scratching with magnetic outer drive.

Worn out outer magnetic drive

Broken wear ring

This failure mode happens after the pump was changed from C3 to C4 type. Prior changing, the failure mode was different (broken casing and scratching impeller with intact containment shell). After changing, the pump last a year before breakdown, while before changing, the pump only last 1 month before breakdown.
I would really appreciate for any thoughts or comments. Thank you.
Best regards,

Yosep

 

[jsasada]

Here are the summary of the failures. The yellow-coloured texts are the failures that happen when the pumps were still C3 type, while the green-coloured are failures of C4 pumps.

 

[pierreick]

Hi,
Did you check with the vendor?
Is this the right the technology for your operating conditions and material (40% Nitric acid)?
Is your solution clean, free of particles ( abrasive) ?
My 2 cents
Pierre

 

[jsasada]

Hi Pierre,

We already checked with the vendor various times. By design this pump is capable to pump nitric acid 40%. We also have reported the failure of March 2021 and their response was due to a manufacturing defect. However, a manufacturing defect should not be the root cause because the failure mode kept happening until now.
Thank you.
Regards,

Yosep

 

[pierreick]

Hi ,
You need to investigate the reason why the coating is gone ( abrasion, temperature, Operation (overheating) , acid concentration (variation) , impurities with the acid ) .
It seems to me that the temperature is already high !
Is it a continuous process or batch process ? Some info about your process could help (simple PFD for example) and description of the operation
My 2 cents
Pierre

 

[jsasada]

Hi Pierre,

The process is continuous. Two pumps with one as standby pump. Attached below is the process flow diagram of the system.

Pump tag number are 12P007A/B. The pump has duty to pump nitric acid condensate from 12E007 to 12C003 column. They are equipped with kickback line for minimum flow as well. We put temperature transmitter on the discharge of 12E007 prior entering 12P007. So far, the temperature transmitter did not show any abnormalities. Range of the temperature was 53 - 55 degC.
Best regards,

Yosep

 

[TugboatEng]

Can you provide a bill of materials for the pump? It appears to have leaked which causes exposure of the iron based components of the mag drive resulting in that brown color and the running. Did it leak through the seals or did the coupling perforate?

 

[LittleInch]

Tug et al,

See the links below for details.

I am particulalry taken with the "revolutionary thrust balanced design". This eem to use some small orifices and a balance chamber to avoid thrust.

The bearings seem to be alpha sintered silicon carbide.

What do the bearing surfaces and wear rings look like.

A potential issue for me is that the internal rotor is clearly now impacting the containment shell which means you bearings are shot.

https://www.flowserve.com/sites/default/files/2016-07/ps-10-36-e_0.pdf

Sectional drawing in section 8

https://www.flowserve.com/sites/default/files/2016-10/26999988-e.pdf

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jsasada]

@TugboatEng

Hi, attached below are the bill of materials of the pump

I have also attached below the working principle of the magnetic drive pump (YouTube video):
Link

I agree with you, there's a leak inside the outer magnet and the blue-paint lining has completely eroded. The liquid being pumped should not enter that area. For comparison, the picture below is the new spare of the containment shell:

The pump has no couplings as it is mag drive type. The seal was completely destroyed but I am not sure whether the nitric acid entered the outer magnet area through the seal first or through the eroded containment shell.

Currently, we are performing 3 monthly visual inspection of the pump's internal parts during plant online condition and we always find a tiny deposit of nitric acid on the containment shell. (the date stamp in the photo below is wrong, it is on March 2021, 2 months after the incident above happened).

We have no idea what's causing this problem as this pump is stated compatible with the fluid.
Really appreciate any thoughts on this.
Best regards,

Yosep

 

[jsasada]

@LittleInch

Hi LittleInch,
Thank you for your response. Does it mean that "thrust balanced" system is the culprit of this failure? Based on the video, it seems that the containment shell does not move when the thrust balance was activated.

If the internal rotor scratched the containment shell during its axial movement, there should be wear in the inside part of containment shell, yet it did not present.

The picture above also shows a dent in the balance shaft, possibly because of friction with the other parts due to thrust balancing system?

Best regards,

Yosep

 

[MJCronin]

"tiny deposit of nitric acid" .... Hmmmmmmm

MJCronin
Sr. Process Engineer

 

[LittleInch]

That's a very interesting video you linked to.

I'm just speculating that maybe the impellor assembly is moving too much bac and forth or can move very suddenly causing the containment sheath to crack or get stressed and allow a little bit of fluid out.

Do the pumps see any form of surge or shock loading?

What do the wear rings look like?
What do the bearings look like?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EdStainless]

And I would like to see data showing that that particular resin is good at this temp in this acid.
I am concerned that the fiber is so visible, there should be a substantial layer of pure resin on the wetted side.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Compositepro]

I would say it is highly likely that magnetic particles from your process fluid are accumulating on the magnets and wearing through the containment shell.

 

[jsasada]

@LittleInch
Yeah, the axial movement corresponds to the suction pressure of the pump, in this case - the Cooler Condenser 1 pressure. We put a pressure gauge in the suction of the pump and so far it does not fluctuate too much. It stays on 430 kPag. However, when the pump running, we often see fluctuations of the pump motor's ampere. You can see the example below. Yet we are not sure whether we should worry about it or not because the fluctuation is so small, below 1 A.

Red line is the amps of A pump, blue for B pump, and black for the discharge flow from the pump to the Absorption Tower.

Below is the picture of the back rotating wear ring

and below is the front rotating wear ring

below is the bearing bushing photo from the backside of the impeller

@Compositepro
We had suction strainer removed from the pump because of the action of the past RCA which stated that the strainer created turbulence flow that would impact the pump. However, the manufacturer of this pump said that this pump could handle solid particles up to certain size.

 

[pierreick]

Hi,
What about the control valve downstream , do you notice any corrosion , erosion ,.... which may come from the nitic acid solution?
Pierre

 

[Compositepro]

There is a big difference between "particles" and "magnetic particles". Particles will pass though without any issue. Magnetic particles will stick to the magnets and accumulate until they destroy the pump. You need a good magnetic separator, not a strainer basket. Every car has a crude one on the oil drain plug. In fact, the deposits on your magnets look just like those on a drain plug magnet.

 

[MJCronin]

Compositepro is correct .... "magnetic particle (rust) seems to be killing this pump over and over.... IMHO

What did FLOWSERVE tell you about the need for a magnetic strainer upstream of your pump ?

I do not see one on your PID

MJCronin
Sr. Process Engineer

 

[jsasada]

@Pierre
Hi Pierre, so far we do not have any big issue on the downstream control valve (LV12005 as in P&ID below)

@Compositepro & MJCronin
So far during our discussion with Flowserve they did not mention about the need of magnetic strainer upstream of the pump.

FYI, we just have performed a pressure test of 13 bar on the containment shell that has been in service for 2.5 months. The leak is confirmed coming from the inside of the containment shell.

Leak coming with yellow mark liquid on the tissue

Pressure dropped from 13 bar to 4 bar overnight

This finding is consistent in the past year since the failure on January 2021. We always find the containment shell leak during pressure test.

My thought is the nitric acid leaked from the containment shell would start to corrode the outer magnet which was ductile iron. Overtime, the corroded parts then start to accumulate inside the outer magnet chamber and amplified the initial damage (leak). We know what was happening but we do not know on why it started and how to stop it.

Best regards,

Yosep

 

[Compositepro]

Magnetic particles will always cause problems on either side of the can in a magnetic coupling. However, it is also very possible to have voids or porosity in the composite pressure seal due to a manufacturing defect.