Filter dryer before rotary pump?

[heinasirkka]

Dear forum users,

We have built a system however the sliding vane pump seems to be unable to maintain the flow. It starts making strange sounds (maybe it is cavitation). At very low rotation speeds (10% of nominal rpm) there is some flow but when we increase the pump speed, there is even reverse flow. Eventually it becomes overloaded and stops working (it is controlled by a frequency drive). There is a filter-dryer at the inlet of the pump. Can that cause a problem?

Thanks a lot in advance and kind regards,
Alihan

 

[heinasirkka]

Thanks! But could you please explain why?

 

[Compositepro]

Many sliding vane pump designs rely on centrifugal force to fling the vanes out of the rotor and will not function at low speed.Vanes can also stuck due to any grit in the fluid.

 

[LittleInch]

Alihan

"We have built a system... " That's very nice, but without supplying some more details we can't see it the same as you do.

Details such as

Purpose of the system,
Inlet pressure,
Outlet pressure
Fluid being pumped,
Temperature of fluid
Vapour pressure
pressure drop across "filter dryer" ( is this a gas?)
Schematic drawing
Size of pump
Flowrate
etc etc




Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[heinasirkka]

Thanks for your answers, I really appreciate it.

Many sliding vane pump designs rely on centrifugal force to fling the vanes out of the rotor and will not function at low speed.Vanes can also stuck due to any grit in the fluid.

So does that mean that we only can use at given nominal full speed (1450 rpm in our case)? Because our system is for research reasons and we'd like to be able to change the rpm for operating at various flow rates.

1. A "sliding vane pump" requires a flooded and NPSHr to perform properly.
2. A Suction drum, properly sized would be helpful.
3. Pay attention to what Compositepro said.
4. Contact the pump vendor where you purchased your pump and ask them.

1- The pump is flooded with liquid. But I suppose the NPSHr should be enough to overcome the system's pressure drop as given by the vendor.
2- Can a suction drum be used for pure liquid flow?

Details such as

Purpose of the system, University lab research on refrigerants' boiling heat transfer capacities
Inlet pressure, %%Pump inlet 5 to 7 bars
Outlet pressure %%Pump overcomes 1-2 bars of pressure drop
Fluid being pumped, %%Refrigerant (Liquid at atmospheric conditions)
Temperature of fluid %% That incident happens around 40°C we would like to go higher around 100°C
Vapour pressure %% 3-4 bars
pressure drop across "filter dryer" ( is this a gas?) %%Pump fluid is liquid
Schematic drawing %%I don't have drawing. The cycle begins with a pump, accumulator (7bar pre-charged), preheating unit, evaporator, condenser, filter-dryer, pump...
Size of pump %%370kW
Flowrate %% around 0.1-0.4 kg/s

At room temperature, the closed cycle is filled with liquid refrigerant. So we ran the pump at low rpm to cause some flow. Then we heated up the refrigerant at the electrical heater unit to pressurize it. Then we gradually increased the pump speed and the heating power. The flow rate increases at first, then the pump makes sounds, then after a while the flow rate becomes negative. The condenser has full-on subcooling capacity, however there is still cavitation

 

[LittleInch]

Ok, that gives something to bite on.

375 kW?? Seems far too big for the duty listed. Are you sure??

You need to look at where you have pressure gauges and provide running data.

Ideally after each bit of equipment but especially at the inlet and outlet of the pump.

As you increase flow my guess is that the pressure drop across all your equipment and piping connections etc will add up to a lot more than 2bar. Hence inlet pressure to the pump, assuming the outlet stays at 7 bar, will get much closer to your vapour pressure than you think, hence vapour in the pump.

Also check temperature at the various points. The vapour pressure might be a lot higher if the return temp to the pump is more than you think it is...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MikeHalloran]

We have built a system ...

You might have gotten better answers sooner if you had mentioned the use of REFRIGERANT in the system. Badly designed refrigeration systems are capable of many strange behaviors, because of the multiple phases present, and because of the behaviors of lube oil, dirt, fibers and water that may also be present.

If you filled a closed system with liquid refrigerant and then started pumping it around, you are lucky to be alive.
... so far.


Mike Halloran
Pembroke Pines, FL, USA

 

[heinasirkka]

The refigerant is neither flammable nor toxic. The dirt and fibers should be with the another filter (not a filter-dryer) in the loop. Furthermore, there is no water in the system. It is also not a refrigeration system. It is an experimental system to test refrigerants heat transfer properties. Anyway, I'd prefer not to die from this.

The pump is of course not 370kW, it was my mistake. It is 370W
I think as well it is best to take pressure and temperature readings from inlet and outlet of the pump. I'm afraid the NPSHr of the pump that was given by the vendor did not match our system. It can be our fault during the order, or their fault, that I don't know. Because according my preliminary calculations, the NPSHa in the cycle is not enough in some of the conditions of the pump.

Thanks again everybody for the answers. I'll write here when we find the solution, for the ones who might still be interested.

 

[LittleInch]

We're always interested in solutions. Why don't you just increase the system pressure by increasing the pressure from the accumulator?

Might be enough to stop the inlet pressure vaporising



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MikeHalloran]

You're running LIQUID refrigerant through a VANE pump?

That doesn't sound like a real good idea.


Mike Halloran
Pembroke Pines, FL, USA

 

[heinasirkka]

Sliding vane pumps can be used with liquids. Especially low-viscosity fluids (such as refrigerants, petrol).

http://www.michael-smith-engineers....o/Pumping-Principles/Vane-Pump-Principles.pdf

Sounds like a good idea, however our accumulator is bladder type and has a fixed precharge pressure and only the vendor can change this (special custom fittings required).

We are planning to increase the pressure by applying heat to the stationary fluid inside the heater. And then minimize the subcooling to ensure there is no cavitation. This should provide enough pressure at inlet and outlet to help the pump to keep on.

 

[LittleInch]

That might be pre charge but what is its max pressure rating? Can't you just pump some more liquid in to raise pressure? Are there any relief valves you'll need to yup rate?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[heinasirkka]

The pressure rating of the system is 35 bar. A relief valve is set at that pressure also. The system is completely filled with liquid, and there is no space left.

 

[LittleInch]

There's something very strange going on here.

"The system is completely filled with liquid, and there is no space left. "

That's what the accumulator is there for to allow you pump up your system. It might start at 7 bar but as you pump in more liquid then the pressure will increase as the bladder distends and the gas pressure behind it increases. - No?

however in use there is clearly gas and liquid - "preheating unit, evaporator, condenser, filter-dryer, pump

The normal position in a refridgeration cycle is for the gas to be compressed just before the condenser where it is then cooled and condensed in your condenser. To put the pump on the d/s side of the condenser doesn't make sense to me.

I think you need to re-think the entire system here as it seems to exist in two states - one completely liquid filled and the other half liquid half gas as you also refer to "refrigerants' boiling heat transfer capacities" implying that the heater is boiling the refridgerant??

IMHO You really do need to sketch this out as a schematic and insert the initial pressures / flows / temperatures with the pump in its current position and then also during your experiment.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[heinasirkka]

"That's what the accumulator is there for to allow you pump up your system. It might start at 7 bar but as you pump in more liquid then the pressure will increase as the bladder distends and the gas pressure behind it increases. - No?"
-Yes, I think that can be possible. Pumping more liquid for also utilizing the accumulator sounds like a good plan. I need to talk to our technician. He didn't fill more because he just set the system in vacuum, and then let the liquid refrigerant be sucked from the canister until it was completely full.

"The normal position in a refridgeration cycle is for the gas to be compressed just before the condenser where it is then cooled and condensed in your condenser. To put the pump on the d/s side of the condenser doesn't make sense to me."
-There are indeed two parts of the cycle. The lower side is filled with liquid, which begins at the condenser, continues to pump as liquid, and then somewhere in the heater it starts evaporating according to the heat power we apply. At the higher side, there is a vapor liquid mixture. It can be also fully evaporator before the condenser, depending on the heat we apply. The condenser is to make sure that the liq-vapor mixture (or full vapor) is subcooled liquid at the inlet of the pump. Because the pump can only work with liquid, and there cannot be any vapor.

I suppose that this system sounds a bit weird to you. The reason is, it is not a conventional cycle. It is for research and just to mimic some of the real life industrial evaporator conditions. The system is made for research to improve the designs of the evaporators.

 

[LittleInch]

Ok.

Is there any sort of restriction or valve on the inlet into the evaporator?

Only when you can find pressures and temps in critical locations will you have an idea what is going on.

Let us know if it works.

LI

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[heinasirkka]

There is no restriction before the evaporator.

I'll keep the thread updated. Thanks for your interest!

 

[heinasirkka]

Dear all,

I wanted to write an update for who might be interested. We have realized that a liquid receiver is necessary at the liquid suction line (i.e. between the condenser and the pump inlet). Apparently the pump did not have enough liquid supply and probably it was rotating inside a halo of cavitation vapor. So the liquid receiver should provide extra liquid. Now there is an issue of controlling the liquid levels inside the condenser. Since there is a liquid receiver, all the condensate will be drained to the liquid receiver and the condenser capacity will be drastically reduced (so unable to subcool more than few degrees). I was wondering what would happen if we connect the liquid receiver with just one tube from the bottom, so the condensate does not go inside directly.