DIAGNOSE THIS 2

[PUMPDESIGNER]

Please go to the following link:

http://www.irrigationcraft.com/diagnose_this.htm

Please examine the photograph and make your diagnosis.
We pondered the situation for a long time before guessing correctly what the cause was, which we then verified.
However we were never able to figure out exactly how the damage was inflicted in detail, it remains a mystery to us.

Thank you.


PUMPDESIGNER

 

[PUMPDESIGNER]

Kawartha - Close coupled, horizontal, end suction, only one pump on the station.

hydrae - Creative guess. We have seen the results of iron bacteria and mostly that is a deposit issue with little or no damage caused to iron.

Wear ring was bronze, no damage visible. We see bronze wear rings frequently on pumps with iron impellers and no galvanic problems.

I'm ready to give it up if you guys are. The reason I stuck with this was because that impeller sat on my desk for about 1 day with me thinking the problem was cavitation, and yet that impeller just does not look like cavitation, I could not relax. Then I guess at the problem and had someone go to the site and check which verified I was correct. Easy solution too.

PUMPDESIGNER

 

[BobPE]

lol pumpdesigner, give us the scoop!!!!
BobPE

 

[PUMPDESIGNER]

My guess was that the damage appeared to be chemical.
The pattern of the damage was gradual, the damage appeared to be rather even in many areas (but not the back side of the impeller). It appeared that something was attacking the impeller other than an a turbulent process which cavitation is. The only thing left is chemical attack.

I had someone go to the site and look at the system. There was a little tube someone had installed into the pump intake line. That tube ran to a chemical storage tank. We inquired and found out that the owners had been sold a fertilizing system where very dilute fertilizer was constantly being pulled into the system. We called the guy that sold them the system and he said (as most people will say when questioned), "we have had no problems".

Anyway, here is what I think I know.

1 - The nitrates and/or phosphates were very dilute.

2 - The impeller is susceptible more to chemical attack because of its relative high velocity, whereas the volute experiences less velocity.

3 - After the eye shroud of the impeller deteriorated sufficiently serious recirculation began, which caused the front side of the impeller to be attacked more by the recirculating water. It is possible that the front of the impeller was thinner than the back side explaining why the front only is missing. However I think that the front is missing because the water behind the impeller would be far less turbulent than the water on the front side under heavy recirculation conditions, and that additional turbulence exacerbated the chemical attack.

Am I correct that high velocities and high turbulence around impellers can exacerbate or multiply the effects of chemicals?

If so, then how does that process work? How can higher velocity increase the effects of chemicals.



PUMPDESIGNER

 

[BobPE]

Pumpdesigner:

You know how I am, so I have to comment....

we pump stuff like that and worse all the time so I would find it hard to believe that the type of primary damage was caused by chemical attack. I think one of the keys is that the impeller exhibits non uniform holes. Chemical attack would be strictly uniform as a function of the mechanism and would not form holes, leaving other metal surrounding the holes more intact.

I hear arguments all the time about this velocity thing. You know what discredits this, look at the other areas of the impeller like the suction, they are under high velocity as well, and no one that prescribes to this thoery can explain why damage did not occur there. Also look at the diffuser section, again high veloity, but no damage. So velocity can not be a contributing factor in chemical attact in this case.

What are the emands like as I hnted to in my original post? This guy was using this pumps as an adjustable tool to acomplich many jobs it seems. I would think, like Quark started to agree, that this would be the root of the problem.

I agree that all hell broke lose when the impeller selfdestructed.

But none the less, this was a fun one.....

BobPE

 

[Kawartha]

PUMPDESIGNER,

I'm having some problem understanding your explanation for this damage. I guess BobPE is also.

I think what you are saying is that the chemical attack occured because of the high velocity, and only in specific areas. Are you making a comparison to erosion-corrosion, where the liquid passing the metal surface continuously erodes the protective oxide layer. If that is the case then as BobPE suggested the impeller back shroud should also be corroded since velocity and inertia force is very high in this area also. Even the volute casing, especially the cut water (also a high velocity and high inertia location) should show signs of this damage.

I like the chemical attack theory but have trouble with the lack of damage to other parts of the pump.

 

[PUMPDESIGNER]

BobPE - For one thing, hold on a second so I can get some band-aids, peroxide, and a brace of whiskey. Going against 'ol BobPE has got to be thrilling or scary.

I bet that they keep 'ol BobPE around even if he doesn't do anything just so that he can scare the bejeebeees out of the younger engineers and keep them on the straight and narrow.

Kawartha, hold me up and I'll talk to him.

Anyway, I understand the concern about why damage is so uneven on the impeller and very little damage on the volute.

However, explain how this impeller continued to pull cavitation as damage progressed, and long before it got to the station we see in the photo. Severe cavitation would eat away until the pump's ability to create low pressure at it's eye was just insufficient to continue creating cavitation.

Also, Since then we came across this guy one other time. This chemical guy beats his competitors on price because he does not have to install an injection pump. I spoke with him and this is exactly what he said: "OK, so our stuff does eat up impellers, I'll grant you that. But replacing an impeller every 2-3 years is still cheaper than installing an injection pump".

I spoke with him in a friendly way. He had just installed a fertilizing system on another pump station of ours. A Parks Department technician called and said that they had low pressure. Our dealer (different dealer this time) investigated and eventually pulled the pump apart and found the same thing. But this time we knew right off what the problem was.

Best I can figure is that a stream of chemical in the intake line damages one area of the pump but then as the impeller is damaged that stream hits different areas.

PUMPDESIGNER

 

[BobPE]

lol pumpdesigner, don't be scared.....lol....they keep me around and they pay me lots of money to figure things likethis out LOL....so I have fun and enjoy my job...go figure lol

what your missing is how cavitation occurs in pumps, and a lot of people do miss it....its not an easy concept...even for me sometimes lol

As the impeller wore away, the pressure went away, lowering the pump curve, but the system curve stayed the same. As such as the impeller wore, the operating point continued to move more and more to the right of its curve (or what was left of its curve) which at 3550 rpm could still induce cavitation. This continued until the owner could not take the pumps operation anymore. I could tell from the picture that it was an irrigation pump based on the damage, thats how good I am at this.

Think beyond the pump to help you figure what no one on the oners side was telling you. The pump was in a lake, the water was free, the demands were real, so....use as much water as possible from the lake...this is a typical response in systems like this. In operating like this the owner routinely cavitated the pump albeit at varying levels of demand over BEP based on the pic.

You chemical vendor sees this damage and associates it with the chemical. this is a natural response since all his clients do the same thing with their pumps, by operating them off BEP.

I feed strong oxidents to pump suctions to pleae owners, pumps are great mixers, so thats not the solution. When they hydraulics are correct, the oxidizers eat the impeller and pump uniformly. You almost had it in realizing that the damage was only restricted to the impeller. That was an important piece of information you gave. To me that confirmed the story the picture told....

I am not the creepy engineer you may think, you would be surprised lol. I have seen this before and I learn from experience. I am also very well educated in hydraulics. I dont scare my junior engineers LOL....But I do make them learn how to do this stuff and they enjoy it as much as me.....

and don't get me wrong, these are great posts!!!! But you have to be commended too....in posting these you have to give you opinion just like the rest of us.....

engineers can run, but they can never hide....

BobPE

 

[PUMPDESIGNER]

BobPE I still don't see it.
The impeller being the only damaged part is trouble I admit.
However, as the impeller degraded it became less able to push water out, which lowers the suction pressure, which makes the cavitation go away. At least that is the picture I get over time.
If you are speaking of turbulence causing the rest of the damage I will not say that cannot happen, but turbulence cannot cause the type of damage that cavitation can because cavitation creates shock waves which can eat metal whereas turbulence (from bizarre shape of damaged impeller) cannot.

I don't think you too tough, just making fun.

I was very busy today so I have not had time to ponder your stuff yet. I just posted so that you would not think you won easily.

Still think I'm right.
Sorry.
Laugh out Loud like you always say.

PUMPDESIGNER

 

[scalleke]

As the impeller gets damaged the discharge pressure decreases but also the NPSHR goes up. Hence the cavitation will actually eat its way through the impeller as the impeller gets more and more damaged.

However I do not believe we are facing cavitation here. Why, because the damage is too smooth. This is water that would normally eat it's way roughly through anything. Secondly cavitation starts after the inlet of the impeller and the impeller is more eaten away at the beginning than at the end. If this would be cavitation more of the leading edge of the blades would be left over. And to throw on some moree arguments: cavitation happens where pressure increases, and the compression side of the blades is not more damaged than the opposite side.

Is it possible that the fertiliser causes the water to become an electrolyte and that we are looking at galvanic corrosion here? It would be logical that the material would be carried away with the electrolyte water that exits the pump.

Have a great day.

Scallake

 

[BobPE]

cavitation can happen just about anywhere in a pump from the suction bell or housing to the cut water to the discharge flange...thats what a lot of people dont realize about cavitation and tend to default to corrosion...believe me... alot of peope think this....scalleke is right, as pressure is increased, cavitation stops, unfortunately as vapor bubbles collapse and cause damage...low pressure does not eliminate cavitation, and thats where your thougs may have went astray pumpdesigner....what was this guy doing with this pump? was it irrigation? Its not a matter of thinking your right, its a matter of telling the owner what really happened...which sometimes can be a never ending process...

Don't confuse my statements with turbulance...I am not talking about that....shock waves from cavitation do not cause the damage, they are a seconday effect that you hear commonly called pumping rocks.....

scallake, that pic didnt look smooth to me, but smooth or rough, cavitation does many different things to pumps, and varying the flow characteristics of the pump will cause cavitation to occur in may places over and over again, making smooth damage if varried enough, making holes if flow conditions dont vary much....

pumpdesigner....you poseted this hear for a reason...I hope its not to prove your right, but rather to learn...If you take anything from what I said, open your mind and think outside the box (or pump in this case) its very rarely the pump that is broken, its the system that is misunderstood...your picking up on the fertelizer input was a good thought, but look at the industry of pumps, we move diluted and non diluted fertilizers all the time with normal iron pumps....that simply cant be the reason.

BobPE.....LOL.....

 

[PUMPDESIGNER]

I agree that cavitation can occur anywhere.
I cannot get by your experiences with oxidants, got me there. But if those impellers of yours were failing would you know? Do you have long term relationship? Hey, just a thought now, you don't need glasses to you BobPE? LOL

I agree with scalleke. That impeller is smooth on the 15 cm scale, rough on the 1 mm scale. Cavitation damage is usually localized.

I am also concerned about the Ns of 9270. Could that negate the NPSHa over NPSHr of 9.4?

Pump is used strictly for irrigation. Not used for hose bibs, or anything else. Perhaps I should send someone out there now and take a vacuum reading. But then if they call with same problem and there is no chemical system, I will be able to release quickly on this to cavitation.

I still get stuck on the process over time, cannot see this impeller causing cavitation as it went away. Even if cavitation did the initial damage, I think that the impeller eventually got to the point where it could not move enough water out the discharge to continue creating sufficient low pressure at the eye to cause vaporization. If you believe however that turbulence caused cavitation then perhaps I could go there. Something like recirculation cavitation?

How 'bout both. I could write a book on "Cavitation Exacerbated Chemical Attack"
I'll let you guys write the introduction. I could go down into the history books for ...

Anyway, stick with me, I'll get there if you don't wear out.




PUMPDESIGNER

 

[hydrae]

Pumpdesigner
Back on June tenth you said
"NPSHa exceeds NPSHr by 9.4 feet worst case" did you measure or calculate that?
How complicated is the inlet piping, such as, a compact street ell tight to the inlet or a strainer that was never cleaned?
Is there a valve on the suction line the operator would throttle to get more chemical draw?
Hydrae

 

[BobPE]

no problem pumpdesigner, i'll stick with you, i enjoy your posts....

you stated that the impeller is causing cavitation, this is where you may be stuck. Nothing in a pump causes cavitation, its the system that the pump is connected to that causes cavitation.

You would have seen discharge cavitation (recirc cavitation) on the pump caseing, cutwater, and discharge. You looked at this part of the pump from what I can tell of your posts, and you didn't see this damage. So that can't be it.

I can discredit most cavitation enhanced chemical attack. I just commissioned a replacement pump where a leading national consulting firm claimed corosion attack. It turned out to be cavitation damage from an improper pump selection. They still believe it was corossion attack, but the new pump (exact materials as the pump that was replaced) is not showing damage now.

You stated the pump was for irrigation (now how did I know that???? LOL), sprinkler heads? mobile or fixed? golf course? farm? I will make another bold statement, there is no flow meter on this system, maybe a pressure gauge on the discharge, but none on the suction. I can't tell you how many of these systems I have designed fixes for....Ask the owner how he uses the system the pump feeds...you will get a wonderful story....and it will tell you why the pump is damaged, trust me!!!!

As for my oxident pump systems, yes I keep in touch with them, they all good return clients of mine...the pumps dont cavitate...the system they are working in will not let them. they do corrode away, in about 1.5 years. From casing perforations.....it seems this part of the pump is the cheapest produced part, the impellers are made in a better process....the client is ok with this dmage becuae I told them it would happen.

No, i dont need glasses, i eat my carrots!!!! LOL

BobPE

 

[scalleke]

"Nothing in a pump causes cavitation, its the system that the pump is connected to that causes cavitation." Boy you really must be into pump design!

NPSHR is a figure that will tell you how much the pressure in a pump decreases before it will start to increase. and yes cavitation damage takes place where the pressure starts to increase. that is one of the reasons why I co not believe we are looking at a cavitation problem here, the high pressure zones on the impeller are not damaged (unless if a higher res. picture could show otherwise).

The issues that cause pressure to decrease in a pump are design related, Shape of the impeller surface finish of the parts dimensioning of the waterway etc. These are all under the control of the pump designer/manufacturer. this makes that NPSHR is a designer manufacturer issue.

The issues that determine NPSHA related are all under the control of the system designer.

Both NPSHR and NPSHA are equal contributers to cavitation problems. I have seen several cases where there was ample NPSHA and there was cavitation damage in pumps.

I do keep on believing that we are seeing a corrosion problem here. For one thing the extensive material removal would never happen gradually in case of cavitation and the unit would have devellopped vibration issues over it's lifetime.

Has the problem been solved by relocating the fertiliser feed?

Should we not focus our research on the reason why the front of the impeller is completely gone and why the backshroud is still intact?

Scalleke

 

[BobPE]

scalleke:

we are focusing on why the impeller disappeared....so far everyone but me thinks the fertilizer did that.... If a pump manufacturer designes a pump to cavitate in its applied operating range, I think they will be out of business rather quickly....

I am really not into pump design, its kinda boaring actually if you ask me...I am into hydraulics....much more exciting.....

NPSHr is not a contributor to cavitation, its just a known constant...NPSHa contributes to cavitation and it is a variable. And as you hinted, NPSHa is a system thing...system things cavitate pumps....

I guess I must be looking at a different picture than everyone, that has to be it....LOL

BobPE

 

[scalleke]

OK, OK I had another look.

You are right we are focusing on why the impeller disappeared my question is just why is only one portion of the impeller gone. The portion of the impeller that is gone is the front shroud only, the back shroud remained intact even if there are erosion marks.

I do question the design of the pump! Wat about the recirculation flow that took place at the front side of the impeller? Was there a clearance at the tip of the impeller that was too small? That created low pression on the front side of the impeller?

What make and type of pump was this? How is it built?
What are the operating conditions both on the suction and the discharge side? And Bob, you are right we should know more about the construcion and funtioning of the irrigation system? I do not believe this was cavitation but I would not rule that possibility out though.

I'l try to eat some more carrots, maybe that way I can get a better look at the picture.

Scalleke

 

[PUMPDESIGNER]

hydrae - Sharp question. NPSH is calculated, which obviously can be wrong. However intake is very simple: 20 feet straight with one 45 ell, stainless screens were inspected and were clean. No valves in the intake except high flow swing check. Can the Ns of 9270 cause the 9.4 NPSHa over NPSHr to be insufficient?

I was joking about the cavitation enhanced ...
BobPE, you are correct to be suspicious about the conditions, and I cannot put you at ease there because they "do things" I agree.

I have not heard anything about this pump in probably two years. I will inquire about it.

I will attempt to find more photos, or the impeller. People think I'm nuts to save that stuff and they throw it away when I'm not looking, or hide it somewhere.

BobPE - I agree with all your statements about cavitation, especially the ones intended to get a rise out of me.

I am still stuck with scalleke in not seeing how cavitation could possibly produce this. Overall impeller is smooth, although some cavitation damage can be seen at the discharge side at the bottom of one of the vanes. However my not seeing it does not convince even me.

BobPE, How can cavitation continue as impeller performance drops: flow rate drops dramatically, cavitation drops dramatically. If impeller performance against system pressure drops then suction energy required drops also.



PUMPDESIGNER

 

[PUMPDESIGNER]

I have noticed on many pumps that the front vane shroud is thinner than the back shroud, don't know why this is.

scalleke - I've thought about that recirculation also. I I posted way back up there in my original explanation that as the eye shroud disappeared (fron chemical) then recirculation had to become huge, which could then add to the disintegration of the front shroud while nothing happens to the back shroud. Now you have the front shroud being eaten on exterior by recirculation cavitation.

PUMPDESIGNER

 

[scalleke]

The back shroud is the main structural component of the impeller. Behind the impeller and certainly as in this case when the impeller shroud is closed pressure will be high and the impeller will be pushed forwards in the direction of the suction side by this high pressure acting on the backshroud of the impeller. This force can be very high depending on the discharge pressure and the impeller diameter. We are talking tonnes of force here. the back of the impeller must be able to withstand this.

Pressure on the front of the impeller is more balanced and can be controlled partially by contriolling the gap between the impeller front side and the casing. The front side of the impeller thus needs not to be so strong.

Can you point us to the pump design on the web? A cross section would tell us a lot?

Scalleke

 

[d23]

Pumpdesigner:

I wonder how many times the chem tank ran dry and gave you an air line to the intake of the pump?

Just for grins did you sent the impeller to a metallurgist for analysis? The pump manufacture may be willing or capable of doing this for you.

Just a thought. Like your new site, it's simple like me.


D23