Calculate cap tube

**herefishy** · 27-05-2003, 06:32 PM

Any tool (online) available to calculate cap tube size/lenght for specific application.

I've got my 100' roll of BC-1 on it's way for the -40F+ R-409A 100w video card chiller application on its way, but I'd rather not have to cut the dern tube down fourteen times before I am happy with it.

:\

**herefishy** · 29-05-2003, 03:01 PM

I called SUPCO, they couldn't help me. I need to determine cap tube length (preferrably for .031" cap tube I.D.) for approx. 250BTUH @-40F R-409A.

**Gary** · 29-05-2003, 04:42 PM

How long is the cap tube on the Prommie?

**herefishy** · 29-05-2003, 05:04 PM

How long is the cap tube on the Prommie?

I'd guess about 48". The actual suction line is about 3' total length, and the way it's wrapped it......

47" of .031" cap tube (according to the SUPCO info) is applicable to a 1/4 HP LBP (-10 SST). SUPCO couldn't give me any recommendations fo <-40.

I asked the guy if he had the equations to determine the cap tube length required. He said that the guy that had those equations retired.... and took 'em with him!!!!!

**Gary** · 29-05-2003, 06:01 PM

I wonder if Chipcon (mfg of prommie) would be willing to share that info?

**herefishy** · 29-05-2003, 06:20 PM

Are you going to call them?..... or should I?

**Gary** · 29-05-2003, 06:33 PM

Which of us is getting paid for this job?

**herefishy** · 29-05-2003, 06:59 PM

Originally posted by Gary
Which of us is getting paid for this job?

Well, 'ought dern it! I don't feel like calling Denmark.

I think I'll start with 60" of .031.

**Gary** · 29-05-2003, 07:36 PM

There's a new thing called "email".

http://www.chip-con.com/index.php?pageid=35

I think I'll start with 60" of .031.

I was thinking 120". Easier to cut than to add.

**Dan** · 31-05-2003, 12:57 AM

I think cap tubes are a guess and bygosh arrangement.

Lol. To some extent, Dan, you are correct. But there has been decent tables set up over the years which make it relatively simple for the design engineer to zero in on a proper capillary tube sizing. Also, the Prof has noted some interesting papers published in recent years describing refrigerant flow thru capillary tubes, so there are still some people studying the subject.

Posted by the prof in March of 2001. LOL.

Perhaps the prof would share one of these tables with Herefishy! Recently I posted in another discussion that the subject was discussed in the Fridgetech pages, but it wasn't there. Who would think to search this site? Now all I have to do is recall what that discussion was so I can search the Refrigeration-Engineer website. I bet that's where it lies!

On a lighter note, when you dig up a discussion 2 years ago and see Gary's picture. He was holding the same grandchild.

Maybe it is not so much a bitch getting old.

**Prof Sporlan** · 31-05-2003, 02:46 AM

The Prof suspects that there hasn't been extensive study on cap tube sizing for the R-12 service replacement refrigerants... other than charge the system till is works right...

When he gets back to the office, he'll look up low temp R-12 cap tube sizing in his Harry Busby reference, and see if it provides a suggested cap tube size.

**herefishy** · 31-05-2003, 05:10 AM

thanks, Prof. It is a real honor to have you get off the bench for me. LOL!!!!

Yeh, Gary.... after thinking about it some more... I was thinkin' about starting at 10' too! (we must be phsychotic)

Particularly since I'm welding one end of this cap tube into a small copper block.... I don't want to be doing and undoing that connection..

see ya'

**frank** · 01-06-2003, 03:42 PM

I've just been looking up cap tube sizing in one of my reference books and came across some info under "Expansion Devices". The calcs are quite prolonged and complicated to post but the book is by McGraw-Hill - Refrigeration & Air Conditioning (ISBN 0-07-061619-1).

It also states that there are some selection charts in the Ashrae Manual - can't seem to find my Ashrae CD so I can't verify this

Frank

**DaBit** · 03-06-2003, 10:09 AM

I do have a few scanned pages from the ASHRAE handbook, with a few covering captube sizing. I won't post them here in public, but I am willing to share them through E-mail.

**herefishy** · 03-06-2003, 02:37 PM

Hi DaBit,

Since you mentioned it, I just pulled my ASHRAE handbook off the shelf, and found the information you reference. I am reviewing it now.

Thanx

**Prof Sporlan** · 03-06-2003, 02:54 PM

The Busby reference suggests 12 ft of 0.031" capillary for an R-12 low temp (-10°F evaporator) system using a 1/6 hp compressor. Doesn't provide a Btu/hr rating, though.

It also suggests 12 ft of 0.026" capillary for an R-12 low temp 1/12 hp compressor, which probably better matches your application.

The reference doesn't give a conversion from 0.026" capillary to 0.031", but the Prof would estimate 24 ft of 0.031" capillary would be equivalent to 12 ft of 0.026".

The 2002 ASHRAE Refrigeration Handbook provides capillary tube capacities for R-134a and R-410A.

R-409A at 100°F liquid and -40°F evap will have a net refrigerating effect of about 55.5 Btu/lb. For a 250 Btu/hr load, we will need 250 / 55.5 = 4.5 lb/hr flow rate thru the capillary.

According the the ASHRAE Handbook, R-134a at 150 psi liquid line pressure, zero subcooling, and 130" of 0.034" capillary will give us a 9 lb/hr flow rate. And according to ASHRAE, 200" of 0.030" capillary will give us a correction factor of 0.65, giving us 9 * .65 = 5.85 lb/hr.

Mmmm... this should put us in the ballpark...

**Gary** · 03-06-2003, 03:51 PM

I've found sizing tables at this site under technical infomation, but I think we will need an interpreter, as it's written in metric technospeak. I have enough problems trying to figure out SI technospeak.

Perhaps Prof could explain it to us.

http://www.cubigel.com/english/frprod.htm

**herefishy** · 03-06-2003, 04:12 PM

Most excellent, Professor.

I had determined a required flow rate of approximately 5 lbs./hr.

In conversation with the dude at Supco, we figured an equivalent of 1/8hp (for cap tube purposes) capacity out of the 1/4 horse compressor @-50F.

In reviewing the chart for Pressure-Drop for Capillary Liquid, If I reference the 5LB PER HOUR line, and hit the .031 ID department, the indicated pressure drop per foot = 3.0

If I assume a condensing pressure of about 170 R-409a, and a suction of 10"Hg vaccum (-5 psig?), at 175 psi pressure drop, I would require 58 feet?

__________________________________________________

I went through an example in the book.

I determined:

flow rate = 5lb/hr
absolute inlet pressure = 175psia
mass flow rate from a chart 38 = 72lb/hr
flow factor (5/72) = .07

and using this information, referencing a chart 39 (the book was saying reference chart 42... but that was an error) I come up with about 375" of .031.

...or 31.25' of .031" I.D. cap tube...

if we change the enthalpy to the prof. suggested 55.5 btu/lb and the 4.5 req'd lbs/hr that changes the flow factor to .06.... that would make the cap tube longer (4.5/72).

hmmmmm. 24'..... or ...... 31'

**Prof Sporlan** · 03-06-2003, 06:54 PM

In reviewing the chart for Pressure-Drop for Capillary Liquid, If I reference the 5LB PER HOUR line, and hit the .031 ID department, the indicated pressure drop per foot = 3.0

What chart might this be? If this chart estimates pressure drop for liquid flow only, it would underestimate pressure drop for a capillary tube used as the expansion device, as it will see two-phase flow somewhere along its length.

If I assume a condensing pressure of about 170 R-409a, and a suction of 10"Hg vaccum (-5 psig?), at 175 psi pressure drop, I would require 58 feet?

In the Prof's humble opinion, 58 ft won't be necessary here...

and using this information, referencing a chart 39 (the book was saying reference chart 42... but that was an error) I come up with about 375" of .031.

...or 31.25' of .031" I.D. cap tube...

Not too far from the Prof's guesstimate...

hmmmmm. 24'..... or ...... 31'

It will always be easier to shorten the capillary...

Perhaps Prof could explain it to us.

The ASHRAE Refrigeration Handbook has a rather complicated eight term correlation to predict capillary tube flow rate... It includes the refrigerant's vapor and liquid specific volume, vapor and liquid viscosity, surface tension, specific heat, among other variables. This model apparently have been found to work well with R-22, R-134a, and R-410A. Other than the surface tension variable, the Prof has an Excel spreadsheet capable of generating the necessary thermodynamic data for this correlation. Mmmmm.... But it would hardly be appropriate for the Prof to be developing a cap tube selection program..

**herefishy** · 03-06-2003, 07:08 PM

Originally posted by Prof Sporlan
What chart might this be? If this chart estimates pressure drop for liquid flow only, it would underestimate pressure drop for a capillary tube used as the expansion device, as it will see two-phase flow somewhere along its length.

The chart to which I refer Fig 43 of section E19.29 of my....1988.... Ashrae handbook (equipment). Whose title is "Pressure-Drop Chart for Capillary Liquid (R-12 and 22)".

on the "x" axis is Mass flow - lbs per hour, and the "y" axis indicates pressure drop per foot - psi. The linear relationship of the cap tube size if charted for .20" to .020" I.D.

And now, if I look at it closer, a more accurate estimation of the reference point would be more like 1.5 psi per foot.

so 175psia pressure drop would be more like 262.5". or 21.875'

I think I'll start with 24' of .031. I have a 100' roll, and I seem to have two or three sources, now that point to around the 24' department.

**Prof Sporlan** · 03-06-2003, 09:32 PM

The chart to which I refer Fig 43 of section E19.29 of my....1988.... Ashrae handbook (equipment).

The Prof will have to root around the Sporlan archives for that Handbook sometime...

**herefishy** · 04-06-2003, 01:46 AM

DOH!!!!

Now, if I determine Mass flow, and required flow rate (5 lbs/hr) and the pressure drop is 1.5 psi per foot, I require 175 psia pressure drop.... maybe I should divide

175 psia / 1.5PD/FT = 9.72' of cap tube....

This makes things scary...

Previously I had multiplied, not divided...

Maybe Gary is back in the running with his 12' figure...

....

PS.. Ihave digital pictures of my heat sink construction.... but I would have to have non-disclosure statements from anyone, before viewing.. hehehehehehe

**Gary** · 04-06-2003, 02:21 AM

PS.. Ihave digital pictures of my heat sink construction.... but I would have to have non-disclosure statements from anyone, before viewing.. hehehehehehe

That good... or that bad?

I'll show ya mine if you'll show me yours.

**Gary** · 04-06-2003, 03:43 AM

I think this might work for sizing cap tubes:

Sections of cap tube separated by access tees. Altogether 10ft. Connect A to B bypasses 1 ft. Connect B to C bypasses 2 ft. And so on.

When you get the right length, pull the testing rig out and install the right length cap tube.

**Gary** · 04-06-2003, 03:50 AM

Hmmmm... should add a 5ft section to make it more versatile.

Could make those sweat tees, run a piece of tubing between each, with a shut off valve for each section.

**Gary** · 04-06-2003, 05:35 AM

There... that's better.

**DaBit** · 04-06-2003, 09:50 AM

Originally posted by Prof Sporlan
But it would hardly be appropriate for the Prof to be developing a cap tube selection program..

[/B]

Put some effort in developing small 1/16 ton valves which operate down to 10% of rated capacity first

But Prof, as usual I am very interested in that Excel sheet and the correlations used. I might be willing to write the captube selection program....

Also a useful tip: Tecumseh / L'unite Hermetique has a compressor selection program which was also capable of calculating captube sizes given a certain compressor, condensing temp, subcool, superheat and load. The program was called DISK UH (DISK_UH.ZIP). A quick Google seach did not show the link, but with some browsing it should show up.

What I did: I selected a compressor that matched the one I was using and ran the captube calculations. When it came up with some odd diameter, I used a mapping table to convert ID #1 to ID #2.

**DaBit** · 04-06-2003, 09:55 AM

Originally posted by Gary
image
There... that's better.

Make it a binary system by doubling the lengths of the captube. Thus: 1ft, 2ft, 4ft, 8ft, 16ft.
This allows selection of the tube lengh up to 31ft with increments of 1ft. Say we need 5 ft, we would dial in the 4ft and 1ft section. If we need 27ft, we would dial in the 16ft, 8ft, 2ft and 1ft section.

**Gary** · 04-06-2003, 01:31 PM

Make it a binary system by doubling the lengths of the captube. Thus: 1ft, 2ft, 4ft, 8ft, 16ft.

Yep. That works.

**herefishy** · 04-06-2003, 02:24 PM

I thought that you could not "break" the cap tube in such a manner, that it must be one continuous tube to even represent proper operation.

**Gary** · 04-06-2003, 06:03 PM

I suppose the turbulance in the breaks could cause some small friction-like loss, but that would seem very minimal. I could be wrong.

**herefishy** · 04-06-2003, 06:14 PM

I NEVER repair a broken expansion capillary tube (splice with 1/4" copper say), I always replace the entire critter.

**Gary** · 04-06-2003, 06:26 PM

Me too. But we are not talking about repair, we are talking about saving countless hours of trial and error sizing adjustments.

Perhaps Prof Sporlan could provide some insight here

**Gary** · 04-06-2003, 07:21 PM

Hmmmmmm...

Lose the handles. Recessed allen stems. 1/4 flare cap for each. Wrap the cap tubes around a piece of tubing. Insulate.

Voila!!! The universal cap tube w/ suction/liquid HX.

Wanna hire me, Prof? I think I can take this idea one or two giant steps further.

**Prof Sporlan** · 04-06-2003, 10:20 PM

Voila!!! The universal cap tube w/ suction/liquid HX.

Wanna hire me, Prof? I think I can take this idea one or two giant steps further.

The problem the Prof sees with this device is it will probably cost more to make than a TEV, and OEMs who use capillary tubes aren't generally keen with TEV prices as it is.

**Gary** · 04-06-2003, 10:46 PM

Maybe. Maybe not. A TEV is far more complex. If you built one by hand it would cost a fortune.

**Brian_UK** · 05-06-2003, 10:25 PM

Sorry to come in late on this one, I've been out of the loop for a while.

Looking back over an old thread on cap tubes reminded me of the Electrolux site which has some data on cap tube sizing which may be of use...

It also has some tables for R12/22/134a/404a.

Try this link -
Electrolux

**herefishy** · 06-06-2003, 03:41 AM

Well, now Mr Brian_UK......... "JUST WHERE IN THE H*LL HAVE YOU BEEN?"

This darn thread has been goin on for a WEEK!!!

That link is most very highly appreciated......

My cuthtomer left with his prototype thrown together video card chiller.

Hey, Gary..... just to let you know... I ended up putting about fourteen inches of 1/4" tubing in the 64' of .031" cap tube and put 27 crimps in it......... LOL!!!!!!! ROFL!!!!!!!!!

I had a 205psig discharge, 5"Hg vacuum, and the screwed up thermometer we were using (not my fluke) that is mounted into the evaporator was reading -50F to -53F.

We had a card attached to my evaporator block, so I kinda conclude that we loaded up the system just a bit.

Also, I had three coils of 20' each cap tube which I tie-wrapped all together..... which I think helped.

We know the compressor really sucks for the application...... so we're just going to overclock the video card and let the compressor fry!!!! LOL

While Donebalp is having fun blowing the thing up, I am going to research a proper compressor/expansion device application for our new set-up.

**Gary** · 06-06-2003, 04:07 AM

Hey, Gary..... just to let you know... I ended up putting about fourteen inches of 1/4" tubing in the 64' of .031" cap tube and put 27 crimps in it......... LOL!!!!!!! ROFL!!!!!!!!!

Have you thought about a hand expansion valve??

While Donebalp is having fun blowing the thing up, I am going to research a proper compressor/expansion device application for our new set-up.

I think I would go for a 1/2-3/4hp methanol chiller with a TXV. Maybe a modified window A/C. Then run waterblocks to everything.

For a second project, I would build a small cascade low stage with a direct die block for the CPU. The chiller would be the high stage for the cascade unit in addition to cooling everything else.

**DaBit** · 06-06-2003, 09:08 AM

A bit offtopic, but for me it is great fun to see you guys breaking your heads over the problems of such small-scale refrigeration.

When I came here first, many of you thought I was insane, and recommended me to just buy a faster processor.

Herefishy: you might want to check out phase-change.com. For professionals it is always good for a good laugh and a bit of head-shaking, but you might find some useful information about overclocking related refrigeration.
Not to mention that a second knowledgeable professional could prevent Gary from getting overworked

**Brian_UK** · 06-06-2003, 09:57 PM

Originally posted by herefishy
Well, now Mr Brian_UK......... "JUST WHERE IN THE H*LL HAVE YOU BEEN?"

Sorry,... family illness, having to work for a living doing assorted hours.

Oh you know what we're like, whinge, whinge, whinge.

Anyway, glad to be of help.

Brian

**Gary** · 08-06-2003, 02:29 PM

Apparently, the prommie uses roughly 68" of .7mm cap tube.

Going by the electrolux chart for R134a, they are aiming for just under 150 watts of cooling.

Switching to the R404a chart, we would need about 3 meters of .7mm cap tube for the same wattage (at a lower temperature).

The .7mm cap tube equates to just slightly larger than .026 inches (.0276), therefore to achieve the same effect we would want to go slightly shorter in length using .026 cap tube.

In any case, 64 feet of .031 just doesn't make sense.

**herefishy** · 09-06-2003, 11:08 PM

I just bought 100' of .028" cap tube. ;-)

**Gary** · 09-06-2003, 11:26 PM

I wonder if a small needle valve (hand expansion valve) wouldn't be just the thing to fine tune it with.

Henry Valve Co makes a 1/4 ODS hand expansion valve. 6291N is the number.

**herefishy** · 10-06-2003, 12:52 AM

The "manual" expansion valve "is" an idea. I'm starting with 12' of .028 (got a vacuum on it as we speak).

Gary, I might have to buy some of your books! (for the input)

LOL.

(I've got a good apprentice who needs it).

see ya'

**Gary** · 10-06-2003, 02:39 AM

175 psia / 1.5PD/FT = 9.72' of cap tube....

I think I may have found a problem:

175 / 1.5 = 116.67

Better buy some math books, too

**herefishy** · 10-06-2003, 04:09 AM

DOH!!!!!!!!!

.... and you thought 64' was crazy...............

This compressor sucks (for the application) My swing was too high with the R-404A.... I charged R-401A and kept a card temperature (other side of the processor on the video card ) of minus 3C to +1C. very stable, and properly charged with up to 142 fps on the 3D benchmark program..... however, I think we're looking for better than that... tommorow I'm buying a Danfoss NFX10....

see ya'

**Gary** · 11-06-2003, 02:34 PM

Correction on information provided earlier:

The uncoiled and measured length of the prommie cap tube is 96 inches (of .7mm ID).

**herefishy** · 11-06-2003, 03:36 PM

Hi Gary,

I hooked up 12' of .028". That copeland 1/4 horse rated at -40 R-12 250btuh was just not doing the job.

My vendor with the Danfoss NF10FX compressors couldn't sell one to me, because the pallets of compressors were already sold to Coca-Cola. DOH!

I selected a Tecumseh rated at about 870btuh @-10FSST R-134A, and employed it with the 12' cap tube and R-404A. The compressor was very happy.

whilst running the benchmark program, the backside of the video card never got over 15degF and I had a suction return temp of about -34F running around a 5"Hg vacuum.

Before booting the computer, I had chilled through to the other side of the card (other side from the evaporator/processor) to -18F

**Gary** · 11-06-2003, 04:35 PM

That's much better. I'm beginning to think that about 8' of .026 with R404a is the ideal for these.

Your superheat is 23F under full load. I wonder if it might flood back a little under minimum load?

BTW, did you get those cheap guages calibrated yet?

Thread: Calculate cap tube

Thread Tools

Display

Calculate cap tube

Similar Threads

oxidation scale in copper tube and fittings

metric capillary tube

The amazing Mr. Cap Tube

Help to calculate Heat Exchancger

Tube staggered/squared

Posting Permissions