Sundyne pump model LMV 311 3

[rota1]

Hi guys!
Can you guys explain to me what in the design of this pump is so unique that allows for the high head and flow developed ? some one mentioned it is all in the discharge diffuser exit port?

 

[pennpiper]

There is a lot involved in making the Sundyne pumps so good.

Check out their online catalog at:

http://www.sundynepumps.com/Files/S...ndPump-SingleStageIntegGeared-16page-2005.pdf

Read all the data and then contact them with your questions.

 

[JJPellin]

The reference from pennpiper gives the Sundyne explanation for their design very well. Unfortunately, I am not as enthusiastic about this design as he is. They are able to achieve high heads with very low NPSH(a) using their conical diffuser and straight vaned impellers by running at relatively high speeds. The cost comes in efficiency and operating range. Their curves tend to be very flat and drop off sharply at the "knee". This makes them very sensitive to operation away from BEP and very sensitive to the way they are controlled. In the Sundyne literature, it suggests that their pumps should be flow controlled rather than pressure controlled. This is because of the shape of the curve. They also suggest that you locate the control valve very close to the pump. They suggest 10 feet from the pump to the control valve.

In my experience this model of Sundyne pump can run well if everything is perfect. But they are not tolerant of problems. Running this pump in parallel is not a good idea. Running them below 90% or above 105% of BEP can be a problem. The last set of similar Sundyne pumps that I installed cavitated badly enough to destroy the diffuser throat within a month running at 105% of BEP. We had to reduce the flow rate to exactly the rated point to stop the cavitation. I had problems with others when the control valve was installed too far away. All similar Sundyne pumps that I have routinely run in parallel have been extremely unreliable. If I can come up with any reasonable alternative, I will not buy a Sundyne. Unfortunately, they will quote services that very few other pump companies would even consider. For low NPSH(a), high head, low flow applications, I would look at Roto-Jet first.

 

[bingopin]

After living with about 20 Sundynes for the past 25 years, alas, I too am not a big fan. Most of the reasons are already listed but I include these:
- There is almost no flexibility in the operating range. You run it on design or you pay.
- They are very sensitive to the surface finish of the diffusser and diffusser cover. If you ever experience high current draw, check the finish...it must be very smooth.
- Mechanical seal choices are very limited. You just don't have room to install a "real" API seal.
- Because the high speed is achieved with a gearbox, they are inherently more complex for the maintenance mechanic than other centrifugals.
- 311's and 331's used a removeable diffusser. While this is handy, it creates additional potential leak paths that are usually sealed with o-rings.
- I seen design efficiencies as low as 12%. This is problematic when pumping some fluids. One momentary loss of flow and the pump gasses up. Many times when this happens the seal is next to go.

The Sundynes fill an application niche. The choices for some applications boil down to a vertical turbine pump or a Sundyne. I try to convince the project engineer to select the vertical turbine but with the difference in the cost of the pump and the cost of installation, the Sundyne wins out many times.

 

[sms]

Amen to all of the above! Running two in parallel is just asking for trouble, running three or more in parallel is a nightmare.



-The future's so bright I gotta wear shades!

 

[rota1]

Thanks guys for the excellent information.We have a similiar problem whereby there are four pumps running in parallel.Even so, designed head and flow cannot be achieved, we have found that the diffuser discharge exit port is 10 thou larger than what it is supposed to be, contractor is telling us this is the source of the problem, as the pump depends heavily on this to deliver head amd flow.Originally this was not the case,but it seems strange to us, if we take what the contractor says, that such accelerated wear can occur in four years time, pumping gasoline.

 

[JJPellin]

Depending on the flow and head for the particular pump, these can be very sensitive to diffuser throat diameter. The last one that I purchased failed to make the required rated flow point on the test stand. They took it apart and honed about 0.002" from the bore of the diffuser throat and retested. On the second test, it passed. The diffuser throats can erode or can be damaged by cavitation if the flow is above the rated point.

 

[rota1]

Thanks JJPellin
A test done on one of the pumps running solo revealed a discharge head of 2200 head ft and a flow of 76 gpm
(running below the rated point) The pump is designed to run at discharge press of 2748.0 head ft (899.8 psig) and rated flow of 88 US GPM pumping propane (SG 0.5).
Based on the above test results is this pump sensitive to diffuser throat diameter?
Is it reasonable to conclude that the throat diameter has increased?
Visual inspection of the diffuser shows no signs of normal cavitation damage.

 

[JJPellin]

For an LMV-311, this is not very high head. For the amount of degradation in performance that you describe, in propane service, I would be surprised if it was all attributable to diffuser throat erosion. There are several other things that could cause a loss of performance in this model:
* The o-ring at the bottom of the diffuser to the inside of the case could be leaking by. This results in internal recirculation which robs performance.
* The internal cyclone separator (if so equipped) could be washed out. I have seen this several times. They make two styles of orifices for these. One of them has an o-ring at the tip and the other relies on a metal-to-metal fit. The one without the o-ring tends to wash out and erode. This is often missed by mechanics since it is not routine to pull this orifice out. The Sundyne literature does not give you the sizes for the orifices in the internal cyclone. If you need them, I have a table that decodes the part number and gives orifice size.
* As noted above, the surface finish on the underside of the diffuser cover will cause a loss of performance and an increase in BHP. The surface finish in the diffuser is also important. If you compare the motor amps to the horsepower curve, this may point you toward this issue. If the amps are high, check the surface finish.
* If it is a very large impeller, they sometimes install a diffuser throat extension insert inside the case. This little half moon shaped block acts as an extension of the conical diffuser throat. It is made symmetrically and can be installed upside down. If it is left out or upside down, performance will be severely degraded. This is probably not your issue with the relatively low flow you are pumping.
* Diffuser throat cavitation tends to happen just downstream of the leading edge on the impeller side of the throat. For small throats, this can be hard to see. If you run a piece of wire along the inner surface of the throat, you can feel the damaged area. Since this normally happens if your flow is above rated, you are not likely to find this problem.
* A washed out throat is still a good potential issue. It is very difficult to measure some throats.
* Some of the older cast diffusers can be installed in the wrong position if the alignment pin in the bottom of the case drops off into the hollow section of the casting. If the diffuser throat is not lined up with the discharge flange, a slight reduction in performance could be expected.
* Impeller diameter can be tricky on some of these. The part number of the impeller has the diameter imbedded at the end for most styles. But they make one impeller type that this is not true. If the impeller part number looks something like this "J10ND33GDB219BS", then you cannot tell the diameter by the number. With this style, they change the face angle and diameter on the test stand and sometimes the documentation gets screwed up. If you replaced the impeller recently, recheck the diameter. I got an undersized impeller for one pump. I notified Sundyne of the mistake and asked for a replacement. They sent me a second undersized impeller. It took three times to get it right.

 

[behar]

Hi Rota1, JJPellin and members all!
I have been following this thread with keen interest, we had a similiar problem some years ago.A new Sundyne LMV 311 was installed in our MTBE plant with severe overloading problems, designed head could not be achieved.Checks carried out found the impeller front and back clearances to be 0.060" and 0.070" respectively.Manufacturer specifications for these clearances were 0.035" and 0.045" respectively.The impeller boss was machined to give 0.045" impeller back clr, and the casing splitline was machined to achieve 0.045" impeller front clr.Care was taken not to move the location of the centerline of impeller to diffuser throat.
This achieved major improvement.
Maybe Rota1 you should look at these clearances?! JJPellin your views on this approach will be most welcome !

 

[bingopin]

My experience has been that the impeller clearances mentioned are, within limits, not critical. I've seen a number of instances where rough, pitted surfaces were causing high current draw and the problem was remedied by machining smooth the diffuser and diffuser cover ("front" and "rear") surfaces. My memory is that Sundyne blessed removing .035" from these surfaces (total) without performance problems and our experience proved this out.

Any chance you cleaned up these surfaces when you did the machine work?

 

[behar]

This pump was new,overloading on commissioning,I think chances are that the surface finish would be " factory installed",further the areas machined were non wetted areas, surface to surface contact.
Is it the as found impeller clearances you found to be within limits?
It is a little baffling to me that you say that the impeller clearances obtained are not critical, and as JJPellin said earlier a 0.002" on the diffuser throat made all the difference.Therefore the most critical dimension for this pump is the throat diameter.

 

[JJPellin]

I did not include impeller clearances in my list of issues because I have rarely seen any problem there. The allowable range for clearance is quite large, as bingopin noted. In any other open face impeller, these clearances are key. I have had similar results by cleaning up these surfaces to improve finish. The improvement with smoother surfaces seems to be more important that the increase in clearnce because of the machining. But anything has a limit. I have no doubt that increasing these clearance too far will cause degraded performance. I find it very interesting that these clearances were excessive on brand new pumps. Quality and design problems are chronic with Sundyne. The point is well taken. Assume nothing. Recheck impeller clearances, front and back. Verify diffuser throat size and orientation, surface finish, flush piping configuration, seal housing porting, impeller diameter, inducer design (pitch and number of vanes), inducer throat clearance (between inducer and diffuser) and gear ratio. If any of these were manufactured incorrectly, performance could be lower than expected.