Designing a Circulator

I can not make sense of your design numbers with any certainty. 400g/s (variable) is not a number that can be used for design purposes. You need to state a maximum and a minimum mass flow rate.

"working ambient pressure - 40 - 150 bar, design pressure head - 10 bar." is also a bit hard to understand. I do not know what working ambient means. Give a number of design cases specifying P1 (suction pressure) and P2 (discharge pressure) with corresponding T1 (suction temperature and m (mass flow). Does design pressure head mean differential pressure?

If you give us these numbers I am sue that we can tell you if a 1 st centrifugal will do the job or not.

what i meant by 'ambient pressure - 40 -150 bar' is that the circulator operates in an environment where the pressure range is from 40 - 150 bar, which is set separately by a compressor (probably a reciprocating compressor). The pressure drop in the whole loop considering all components is about 10 bar, which the circulator has to overcome.

The closed loop system has to provide an adjustable mass flow rate with a maximum design mass flow rate of 400 g/s and a minimum of 2 g/s (which is quite a low value). I would like to know if such a pump could be designed. If not, how else could u achieve a variable mass flow rate in my loop at a given fixed pressure (valves? piping?)

I hope this clears some of the ambiguity regarding my previous question. Do let me know if any further clarification is required.

If I play with your numbers, it becomes quite interesting. Using different pressure ratios, (41 to 51 or 151 to 161 bara) and different mass flows (from 2 g/s to 400 g/s) you would need compressors varying from 65 kW polytropic power rating down to less than 1 kW and from 6 to 2 impeller stages.

Apart from being exceptionally expensive, a centrifugal compressor will not be able to cope with your massive turndown ratio and variation of pressure ratios. The impeller and diffuser channels would have to be impractically narrow for some design points.

This is a job for a variable speed positive displacement compressor.

Start by looking at the shaft work needed, which is the volumetric flowrate, at whatever pressure, multiplied by the pressure increase through the compressor. What does that come out to?

the volumetric rate is got by dividing the mass flow rate with density. the working fluid is gaseous helium (at 60 deg C) and the density varies with pressure. Also the mass flow rate is variable.

wat pressure and mass flow rate should i take? As i see it we are looking for the max shaft work, and consequently, we calculate the max volumetric mass flow rate which is got by dividing max mass flow rate(mfr) with the min density (which means at the lowest design pressure- 40 bar). this works out to be 0.0703 m3/s....... am i proceeding in the right direction???

Looks good from here. Now consider the system pressure drop at that flowrate. What is it?

The answer in N/m2 multiplied by the flowrate of 0.0703m/s gives a power in Watts. Divide that by, oh, say 70% to get a motor power for this circulator in Watts.

Why is Helium, such an awkward material, being used as a circulating fluid?

Ok, the problem now is that dont know my pressure drop before hand as the pressure drop is not fixed. but i know for sure that the pressure drop will not be more than 10 bar.

to calculate the shaft power, i had referred to the following formula posted in another thread - W = P1*Q1/n*((P2/P1)n - 1) where P1 is the inlet pressure and P2 is the outlet pressure... now what should P1 and P2 be? 40 and 50? or 140 and 150? because i am currently as u had suggested calculating the maximum shaft power if u remember....

The Helium here is used as a coolant - to extract heat.

vm289

For all these calculations you need to use absolute pressures and temperatures. You will need the most shaft power at the highest compression ratio and highest mass flow. Your highest compression ratio assuming that you always have a 10 bar pressure rise will be at 41 to 51 bara or P2/P1=1.24.

I repeat what I said in my previous post, you can not do this with a centrifugal compressor. You will battle to achieve any of the points with a single stage machine and I doubt very much whether any centrifugal machine is capable of this turndown ratio.

You have to resort to a positive displacement compressor (probably cheaper but slightly less reliable). Whether this should be reciprocating or rotary I have not thought about. Having said that, here is how to go about the calculations for a centrifugal machine. If this is all familiar to you then I apologise. Some of it has been covered by others.

As we do not yet know the physical attributes of the compressor, we will use only polytropic compression and an assumed efficiency. The formula are all on one page as this is the only way that I know to import them, so if anything is not clear, or I have left anything out, let me know.

Actual Power = mass flow x polytropic head/efficiency or mass flow x actual head.

R is the gas constant 8,325/Mol weight (or should that be mass?)

Assuming an efficiency you can calculate the actual head h_act.across the machine.

By assuming a machine Mach number of close to unity, say 0.95 and a head coefficient (relationship between head and impeller speed) of somewhere between 0.45 and .55 (this range is proposed by many OEM's) you can calculate the polytropic head per stage (D2 is impeller OD and N is revs/second). Assuming all stages have similar impellers then you can calculate how many stages are required to achieve the actual head required and the rpm of the machine.

Use a PD compressor.