vapor pressure and dissolved gasses

[navi26]

Does anyone know if it is standard practice to include the effect that dissolved gasses has on the vapor pressure of a fluid? All that I have been able to find is a specification sheet for potassium carbonate at one concentration and temperature. I know that we have been having problems with not enough NPSH, but the cavitation damage on the impellars has significantly accelerated over the past failure rate and we are trying to track down why.

I am currently having a failed impeller metallurgially analyzed and am going to have a brand new impeller checked for casting imperfections. Does anyone know if impeller surface roughness may contribute to increased cavitation damage (more nucleation sites?)?

I am leaning towards thinking that we are either getting bad impeller castings, or the amount of dissolved gasses in the potassium carbonate is higher than in the past. Let me know if anyone has any thoughts.

thanks!

 

[BigInch]

Standard practice and specific fluids other than cold clear water are mutually exclusive. If you have a specific fluid with any characteristis other than cold clear water, you are responsible for accounting for whatever different characteristics it may have.

I'd think dissolved gas would not be responsible for pushing you over the cavitation limits and on into the severe area. You've got other problems that I'll just bet you can attribute to typical causes, small diameters, low tank pressures, pump elevations too high in relation to fluid surface levels and hot fluids. I don't think potassium carbonate has an overly high vapor pressure at normal temperatures.


"I think it would be a good idea."
- Mahatma Gandhi (1869-1948),
when asked about Western civilization

 

[navi26]

The pump suction pressure I have seen to be 21 psig and the vapor pressure of the potassium carbonate @230F is 19.1 psia. At the flowrate that the pressure transmitter is displaying we are running at about 4' above vapor pressure. This is a really small margin and I am trying to figure out if the dissolved CO2 in the potassium carbonate is eliminating this head. Also, as a side note, we are seeing slightly more cavitation damage on the inboard side of the double suction impeller.

We are out on the right side of the curve and to reduce discharge recirculation we have tightened the clearance on the wear rings and changed the sacraficial ring to PEEK. We are still seeing the classical signs of cavitation from the failed impellers and also on the vibration data. The only thing that is strange with the vibration data are some sidebands at turning speed that we cannot identify.

This has been a problem pump for the last 40 years with the impellers being replaced approximately every 1-1.5 years. Over the past year we have been having impeller failures about once every 3 months and I am trying to find out what has changed. I am also pursuing replacing the pump with a slower rpm multistage pump, but I don't know if the Life Cycle cost will work out. There will not be an opportunity for a couple years anyway.

 

[BigInch]

Suction pressure is way too close to vapor pressure. I'd say that cavitation is guaranteed to happen somewhere in the pump. If your pump has an NPSHR of more than 4 ft (I'll certainly bet that it does), its just not going to work out. You should have no reason to expect this pump to ever perform well under those conditions.

"I think it would be a good idea."
- Mahatma Gandhi (1869-1948),
when asked about Western civilization

 

[Artisi]

Firstly I would investigate very carefully what has changed - having impellers metallurgially analysed will not overcome the problem, look for the cause - don't concentate on trying and treat the effect, as you can bet it is a cavitation problem tied in with NPSHa/r coming from temp, gas entrainment, tank levels etc. are a combination of all.

You say that impellers are changed 1 - 5 years, what happens in the 1 year period compared to the 5 year period -must be a story there somewhere?

Are flow rates, power inputs etc monitored - this can also give you a lead into what is happening. Is pump thru-put the same now as over the past 40 years.

The unbalanced cavitation on the inboard side of the impeller also needs investigation- has this always been the case or has something been charged recently with the pipe work close to the pump inlet, ie, new valve fitted etc.

 

[Artisi]

correction- "don't concentate on trying TO treat the effect"

 

[Artisi]

Can you post a few photo's - this will give a better idea of the "damage".

 

[navi26]

Here is a picture of the outboard end of the impeller. The failure have been every year to year and a half. For the last 9 months we have had 3 failures and the element in there now needs to be changed.

 

[navi26]

Here is the inboard end of impeller. The wear rings also have more damage on this impeller than some other failed ones I have seen.

 

[navi26]

There is a flowmeter and the flow has stayed roughly the same. There has been more demand since the plant production was increased 20 years ago. The one thing about the flowmeter is I don't think that the orifice has been checked in 5-10 years. The inlet block valve has failed in 3 months and there could be the same kind of thing going on with the orifice where we are getting a lot more flow that what we think. I don't think this is the case because the flow is really not as important as the amount of CO2 and CO that the potassium carbonate dissolves (what we are actually monitoring, the flowrate is just for trending). We are testing for suspended solids and dissolved solids, although I don't know how successful that will be with the potassium carbonate. Also, the failures look like cavitation (not in the classical location though?) and not erosion, but I am not an expert. The next thing to check that has been suggested to me is to do a scanning electron microscope x-ray or something like that on the pit areas where the cavitation is occurring to check for chlorines and other things I guess. Out of my experience and knowledge, but we'll see.

 

[unclesyd]

Here is the Discflo pump that solved a lot problems for us with hot NaOH with solids and high gas levels of H2 and N.

http://www.discflo.com/autoslct.htm

 

[Artisi]

From your photo's I couldn't say that there isn't any cavitation but I can see a lot of erosion in many areas of the impeller. Is there damage in other areas of the impeller?

To me, the damage looks to be erosion and could ever be erosion/corrosion - and repeating myself, I do not think it is predominately cavitation.

What material are the impellers, are they OEM or pirate parts, if pirate parts -- have you changed supplier recently.

In my first reply I suggested not to have the impellers metallurgically analysed - seems I will now change my mind and suggest that this could be a good idea to see if the material is to spec. etc.

A closeup pic. of the damage would be helpful - this will allow us to zoom in and not lose definition.

 

[btrueblood]

It is also hard to tell from the photos, but:

There appears to be damage localized to one side of the rotor on the inboard side, does the damage on the outboard side happen in the same clocking location (same side of the rotor)? Dunno why that would happen, but I can envision sludge buildup when pump is stopped...or a gas bubble...either case adding another element to the corrosion/erosion concerns noted by Artisi. Or, it could just be a red herring caused by variation in the material of the rotor. In any case, the uneven damage could certainly cause unbalance on the rotor, and would likely explain the vibration numbers you say you are seeing.

 

[navi26]

Artisi, the impellers are OEM. I can probably get a material origin from them. I will investigate the erosion issue further. I am out of town this week and I will put more pictures when I get back.

btrueblood, the damage is focused more on the inboard end, but the location on both the inboard and outboard end is the same. This has caused us unbalance and that is what is making us change the impellers right now, instead of the lack of a lack of flow a little later. The sludge buildup is not a problem because this appears on the pump during continuous operation.

 

[Artisi]

The localised damage is a bit of a mystery, why isn't even across the entire impeller? Is it typical for all failed impellers?
As for gas entrainment, if not excessive and causing vapour locking within the pump can in fact be benifical with regards to cavition as it tends to soften the implosive damage on the impeller surface. An old "field-trick" for a cavitating pump is to introduce some air into the inlet to reduce the noise/damage.

I miss read your first posting re the time span for impeller changes, disregard my comment of 5 years.

 

[navi26]

A little bit of an update. We took out the impeller today and I have spent a little more time analyzing it. Actually looking at the high pressure side of the vane on the suction. As Artisi was saying, it looks more like erosion. My question to Artisi is, why did you think it was erosion when the pictures only show the low pressure side? I will do some google picture searches on cavitation and erosion and try to see the difference. I have also attached a picture of the latest impeller. As a side note, btrueblood, I guess the last time they replaced the filter for this fluid, there was a bit of sludge build-up in the filter housing. We are replacing the filter today and I am going to have them increase the PM on it also, at least for the short term until we get the fluid cleaned up again. Thanks all, and if there is any further developments I will put out a note.

 

[Artisi]

what is the flow/head of the pump in relation to the performance curve curve.

The problem might be the result of either under or over capacity resulting in mismatch of flow into the impeller which can result in damage similar to insufficient NPSHa..

Question for consideration - the lastest pic you posted certainly looks like cavitation -but is it the result of flow mis-match.

As for erosion. I have cut a section of one of your pics - if you look closely at the left hand of the pic you can see the signs of erosion on the outer area of the impeller eye.

Lets assume that it is typical suction cavitation - the question is then - why? As stated above you need to investigate the pump performance in service compared to the original design / pump capabilty.

 

[navi26]

We are on the right side of the curve, as we have been over the last 40 years. The erosion damage is also visible on the wear rings, but I don't think I included a picture of that.

We know that this pump is not in a suitable application, but the reason why we have all of a sudden seen decreased impeller life is what we are trying to figure out and correct. We are sending out the fluid to get analyzed. Since we are so close on the NPSHr and NPSHa it can be something as simple as the specific gravity has changed slightly. Also, there may be a slight increase in suspended solids. We measured that yesterday and got something like 2mg/L on a 1.5micron filter.