Capillary tube size!

[mkpo]

Can anyone help me find the size of capilary tube for a very small system, that consists of about 47 watt of cooling load, 67 degree C of condensing temp, 8.5 degree C of evaporating temp and 8 degree of superheat.......Refrigerant is R134a.......

Thanks and regards,
Martin

[Brian_UK]

Download the DanCap program from Danfoss here.

http://www.danfoss.com/United_Kingdo...ning/Software/

[Chef]

You may also find this link interesting.

http://www2.the-acc-group.com/java/X...4a&applic=HMBP

If you buy a Danfoss BD35 (rated at 140 watts at your conditions) it comes with a 3m long tube 0.63 diameter.

The program suggests using 1.77m of 0.63 tube.

The BD35 is rated at nearly 3 times your duty yet it is supplied with a tube of 1.23 metres longer. It may wise to be very careful with the results of the program.

If you did indeed use the 1.77m length then the quality at the entrance to the tube would be 0.46 and you will lose a lot of performance.

Other sites suggest lengths from 4 to 6 metres which seem more appropriate - one site suggests 7.8m but this may be too long for start up loads on the compressor.

Chef

[mkpo]

Thank you Brian and Chef, but the Dancap program is limited for 60 degree C condensation temperature. I need little bit more i.e. 67 degree C..In this context is it good to use that software......

[Chef]

Thank you Brian and Chef, but the Dancap program is limited for 60 degree C condensation temperature. I need little bit more i.e. 67 degree C..In this context is it good to use that software......

mkpo - it is important to read the replies you get to a post.

On the link I gave you it explains a rule of thumb to get a length when the condensing temperature is higher than shown in the table.

I have run your case through 'another' program and it shows 9.1 metres of 0.6 dia tube, as I mentioned earlier there are several other sources that put the length at 6-8 metres.

If you use the length recommended in the program have you will have a quality of about 0.46 at the entrance to the tube and a very (very) poor performance.

If you send me datails of the compressor - its specs at your conditions - I will run a full simulation for you and send you a PH diagram. If you tell me the charge I may also be able to give you the best sizing of the condenser.

Chef

[mkpo]

Thank you Chef, I got u. Actually, I have planned to use 1.4 cc miniature compressure from ASPEN and is looking for small heat pump system rather than refrigeration, so I need higher value for condensing temperature, in my case it is 67 degree C.....The quality of 0,46 is okie with me because i am working on the upper part of the p-h diagram....BTW, the capacity of condensor is only 72 watt....If capillary tube is so long...what could be other options...may be variable needle valve....what do u think.....I dont have detail idea about these accessories.......would u plz help me......

[Chef]

OK so you have a 1.4cc compressor, what is the RPM so I can work out the mass flow.

Also when I say the quality will be 0.46 for your too short tube - this is on the upper part of the diagram ie it only goes half way across so you lose 1/2 of the condensor capacity.

Your short tube is the blue line and a proper tube length is the black line. (See Attachment)

Send me as much info as you can and I will run a full simulation for you for tommorrow.

Chef

[mkpo]

Hi,
Thanks for the pH diagram.
The Rpm is 2000...I dont need more than that i guess...I think a suitable volumetric and isentropic effency should also bee considered....Am i right.....

[Peter_1]

You will only find the correct length after several trial and errors.
How you found 67°C?

[Chef]

You will only find the correct length after several trial and errors.
How you found 67°C?

Peter I think it is now possible to get close enough to the right length that the cut and try method maybe relagated to the past, it still works but its a lot more time and effort than calculating it.

mkpa The attachment is a simulation of your system based on the data you gave me, it was a bit light in details and lack of system charge and condenser details reduce the amount of information that gets output.

The results show a big differance to your initial 47 watts of cooling load.? It is difficult to size a cap tube from cooling load alone as it should always be sized to the compressor.

The condenser temperature has been raised to 71C to account for temp differance in the condenser and I assumed 67C was the ambient.

First - I had a look at the Aspen site it only shows RPM minimum as 3000 so are you sure you can run it at 2000?

Second - the condenser should be around 150watts and the evaporator around 90-120 watts to make sure the configuration shown here works.

Third - the condenser volume should be sized at around twice the charge volume to ensure the evaporator is starved during startup and controls compressor overload on suction throttling. You are already running a very high discharge pressure!

Fourth - The tube length for the 0.6mm diameter is the best option as it will have better control if the evap' or cond' Temps change from design condition.

Hope this helps

Chef

[mkpo]

Dear Chaf,

Thank you very much for your explanation. The explanation is of course in-depth and it gives me some initial guess to start my further work. Thank you very much once again.

Peter, actually i would like to avoid the cut and try method, that is why I need more closer initial value of capilary tube.

Just one more inquisitive, Is it possible to use a variable flow control component ( I heard about needle valve) rather than capillary tube so that i can avoid cut and try optimization with capillary tube?

[Peter_1]

Mkpo, you're anyhow at the right address with Chaf, sorry...Chef. He already traveled through those tinny tubes. The algorithms he made are amazing,...
I know a lot about refrigeration but this is beyond, far beyond my knowledge. Chef combined complex mathematics with complex fluid dynamic theory.
My hat of for Chef and his work.

Chef , I have a perhaps a crosscheck for you where we have a cascade running on propylene/ethylene and the ethylene side is fed by a capillary tube
I can search once the readings if you wish
Can you give in also a specific diameter where your program calculates its specific length or is it both it calculates.
Did you ever cross-checked it Chef?

[mkpo]

Chef, BTW, can I get access to the simulation software that u are using for estimating the capilary tube lenght......

[Peter_1]

Mkpo, Chef made it himself some months ago out of scratch.
I have a strong feeling that it isn't just filling up 2 or 3 boxes to get the result straight away, isn't it Chef?

[Chef]

mkpo - my appologies for the typing mistake - I must pay more attention to what I write.

Peter - I would very much like to do a cross check with you on your ethylene system, thanks for this. I have done extensive cross checks on R134a fridge and also freezer systems and so far I have only been using R134a as the refrigerant.

It will be interesting to see how it works on ethylene as I have never worked with this. I really dont know if the program can support it as the R134a program uses a renormalisation of the kinematic viscosities as the velocity profile across and along the tube changes, this might not work with ethylene without some modifications. But we can make them
if needed.

The tube program calculates the length given the diameter plus other essential operating conditions.

If you send me the basics to begin with:- comp volume and RPM, discharge pressure and condenser temp plus evap pressure and temp and superheat if you know it and a range of tube diameters I can get started. BUT - it will give the optimum tube length for conditions when there is a liquid seal at the tube inlet.

If your system is running, as I suspect it may be, with a quality at the tube inlet of say 0.1 or 0.2 how are we going to cross check the results then? As far as I know there is no way to measure what the actual quality is?

Let me know
Chef

[Chef]

Chef, BTW, can I get access to the simulation software that u are using for estimating the capilary tube lenght......

The simulation software is seperate from the cap tube program.

It looks at the 5 main components in a system, compressor, condenser, evaporator, tube and refrigerant over time depending on the load. It shows the systems pressures, temps, fluid distribution etc. So you can change a component, say the size of the evaporater and see the Qe, COP, Pressures, temps, superheat, flow etc change with it.

It does not calculate the tube length but actually needs it as part of the input data.

At the moment the program is in development as time marching is quite complicated and tends to complain if you put wrong data into it. Also it does not have a nice friendly user interface on the front of it and its a bit tricky to use! Its really not ready yet.

If you a specific task you need to solve maybe I can try and run it for you.

Chef

[mkpo]

The simulation software is seperate from the cap tube program.

It looks at the 5 main components in a system, compressor, condenser, evaporator, tube and refrigerant over time depending on the load. It shows the systems pressures, temps, fluid distribution etc. So you can change a component, say the size of the evaporater and see the Qe, COP, Pressures, temps, superheat, flow etc change with it.

It does not calculate the tube length but actually needs it as part of the input data.

At the moment the program is in development as time marching is quite complicated and tends to complain if you put wrong data into it. Also it does not have a nice friendly user interface on the front of it and its a bit tricky to use! Its really not ready yet.

If you a specific task you need to solve maybe I can try and run it for you.

Chef

Its okie...Hope that you will share about the simulation tool, once it is ready...I have used EES to optimize the components in my project....I will definitely knock on your inbox if i get confused in my prototype building, as I dont have practical experiance of making prototype..thanks and Hats-off to your detail explanation....

[Peter_1]

Chef, the system is indeed running since many years

What I have here with me:

2 compressors Copeland DLSGB-301
Swept volume = 22.46 m³/.h at 1.450 RPM

Low stage R-1150 Ethylene /C2H4 10,5 bar Hp and 0.7 bar absolute LP
Temperature at the entrance of the compressor service valve is +/- -5°C, so a very high SH

High stage Propylene/C3H6, 11.5 bar HP and 0. bar absolute LP
After 6 min running, -109 °C at the entrance of the test coil
High stage is water-cooled and maintained at 1.5 bar

Both cap tubes are 1.5 mm ID, wonders what length you will calculate

I don't have an idea of SC but I could measure it once if I'm there (plant is +/- 250 km from my home, so we only go if it's needed)

[Chef]

Its okie...Hope that you will share about the simulation tool, once it is ready...I have used EES to optimize the components in my project....I will definitely knock on your inbox if i get confused in my prototype building, as I dont have practical experiance of making prototype..thanks and Hats-off to your detail explanation....

For practical build help you will be better calling on Peter as he has years of experience.

If your interested in writing some modules or subprograms for the simulation project then let me know. You can send me a PM if you wish.

It seems Peter and I may be hijacking your thread with a different thread, shall we stay here or start our?

Chef

[Chef]

What I have here with me:

2 compressors Copeland DLSGB-301
Swept volume = 22.46 m³/.h at 1.450 RPM

Low stage R-1150 Ethylene /C2H4 10,5 bar Hp and 0.7 bar absolute LP
Temperature at the entrance of the compressor service valve is +/- -5°C, so a very high SH

Both cap tubes are 1.5 mm ID, wonders what length you will calculate

Peter - Can you confirm the power consumption of the compressor so I can be sure I have the correct flow rate. (4Kw?)

Also the inlet temperature is very high at just -5C from what your saying the superheat is almost 100C ??? if that is so the discharge temperature will be quite big also? I dont know how these cryo systems work so please explain.

Can you confirm the evap pressure is 0.7bar abs and what is your best guess at the suction temperature.
As the tube is balanced to compressor mass flow any error in the suction conditions will manifest itself in the length.

Do you have the condensor temperature - it may help.

Thanks

Chef

[Peter_1]

it's a cascade system, I don't know for the moment the running power but 4 kW seems ways too high Chef.
It must be something around 1.5 kW.
The ethylene side cools +/- 15 m of 5/8 tube in an almost perfect vacuum chamber.
The SH is indeed very high for the ethylene side.

Propylene is lightly vacuum 0.74 bara and temperature around -5°C (a guess)
Pictures, see second post on http://www.refrigeration-engineer.co...ead.php?t=5203

We have 3.6 m x 1.5 mmID

[Chef]

Peter - youre not really supposed to tell me what the tube length is before I try to calculate it, now you have told me it makes it more difficult to be objective and I have to be really vigilant I am not influenced by the results.

I have tried to use the data you gave me and with 10.5 bara Discharge and 0.7bara suction and 22.4m3/hr I get a mass flow of about 6g/s to 8g/s.

For a 1.5mm tube that gives an exit velocity of about 2 times sonic i.e. 720m/s. You dont even need the program to calculate that, just get the volume of the gas at h4 and multiply by the mass flow and divide by the area. So it is not possible to have that velocity in the tube. So something is amiss here.

If it is .7barg and therefore 1.7bara it makes a huge differance - can you check for me.

Otherwise I am busy modifying the program for ethylene and it should be ready soon.

Chef

[mkpo]

For practical build help you will be better calling on Peter as he has years of experience.

If your interested in writing some modules or subprograms for the simulation project then let me know. You can send me a PM if you wish.

It seems Peter and I may be hijacking your thread with a different thread, shall we stay here or start our?

Chef

No problem. You can continue. I am also interested to know the cross checking....

[Chef]

Peter - Having modified the code to use ethylene I have run the scenario you gave me. I am not even sure if it is correct yet but here goes on the first attempt.

Flowrate 22.4m3/hr, discharge pressure 10.5bara, suction pressure 0.7bara and superheat 104.8C. (I assume no subcooling or fractional quality at the tube inlet so h3 is on the liquid line.)

It is quite a slippery refrigerant and its viscosity is very low even at the low temps your using so this has caused a small delay - sorry about that.
With the conditions you gave me the exit to from the tube is supersonic, the speed of sound is around 312m/s but the flowrate out of the tube gives a velocity of 579.9m/s.

This is a condition which is difficult to reconcile. [I have studied the relationship of sonic speed in a two phase flow and it may increase the speed of sound by a small fraction and so at x=0,99 it will affect the results but at x=0.3 it is mainly irrelevant]. Even so I have plotted the results on the attachment.

It is also a sensitive refrigerant in that every degree of subcooling adds 60-100mm of tube length to the cap tube so even moderate subcooling can change the length significantly.

But I tried another scenario with 0.5 bar of suction pipe pressure loss and an evaporator temperature of -100C and this fits very well - it does not have sonic at the outlet of the tube and the results are much more reasonable.

Condenser and evaporator temperatures would be a great help in solving this riddle you set me.

I will try and get the revised version posted soon.

Chef

[Peter_1]

Peter - ....

Condenser and evaporator temperatures would be a great help in solving this riddle you set me.

I will try and get the revised version posted soon.

Chef

Chef, the first time I'll be there - but this can be months - I will measure all this for you.
It's the same place where the bottle with water was placed in a high vacuum chamber and the technician filmed this vacuum process.

[Chef]

No worries Peter, but if you come across another system nearby that you can get some data from I would really like to try to match that as well.

I want to get the program quite solid and reliable so it can then be used as part of the simulation program, any help at all is greatly appreciated.

If any other RE member has a set of data I would also be intersted to hear from you.

Chef