Impeller Trimming 1

[RefineryRR]

I have a few questions regarding impeller trimming:

1. Many end-users attempt to use trimming of impellers as a strategy to reduce throttling the pump due to the pump being oversized. By reducing the impeller diameter, the efficiency of the pump is also reduced. Can this strategy of trimming the impeller also generate more heat in the volute casing due to increased fluid circulation? If so, is there any way of quantifying this increase in temperature?

2. Trimming the impeller also changes the impeller blade angle (usually increasing it). The affinity laws do not take this into consideration. How would one determine the new pump curve following an impeller reduction if the blade angle has changed, and the affinity laws aren't valid (as the original & final impellers are not geometrically similar).

3. In a shrouded (and semi-shrouded) impeller, what would be the impact of trimming the vanes of the impeller ONLY, leaving the shrouds in tact? That is, what would be the impact on Head, Flow Rate, Vibration, Internal Recirculation, Heat Generation, etc?

 

[ashtree]

RefineryRR

Excellent questions.

1) There is no doubt that trimming the impeller may increase fluid circulation at least in theory. In practice however i suspect this is relatively small if you stay within the maximum permissible change in dimensions. Some of the "lost" efficiency is energy used for that recirculation being converted to heat. Short of fitting a pump up with a very comprehensive array of sensors it is difficult to quantify, but the change in efficiency is the maximum extra energy used for heat generation in the circumstance.

2) What you are saying is correct. However unless the impeller has been drastically trimmed the pump affinity laws would still give a close approximation of performance. But at the end of the day if this is not good enough and you need a new curve representing the actual pump you have you can develop a curve by setting up flow, power, rpm and pressure sensors and taking the readings over a number speed , flow and head scenarios. Likewise there are specialist labs that do this type of testing and many pump manufacturers are equipped to do it as well.

3) Difficult to quantify, but it would be likely to reduce the re-circulation that otherwise might be increased by impeller trimming.

Just to put some practical experience around this whole discussion i have attached a photo of a pump that had damaged an impeller twice in about 10 years due to cavitation. As you can see the impeller is badly eroded and the impeller shrouds are cut back and the whole shape of the impeller has changed. After the previous event the impeller was even worse. On the first occasion some performance had been lost and i dont remember the exact numbers but i don't think it was anymore than 10 percent. On the occasion where the photo was taken the performance loss was only about 5% at the most and not sufficient to attract the operators attention. On both occasions it was only the fact that the pump made a scraping noise when being turned slowly that attracted attention.

The point is that , at least for some pump designs things can change a fair bit and will only have modest impacts on pump performance.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[Artisi]

Good question and the response sums it up perfectly, can not see the need to add anything further at this stage.

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.)