Brazing long 6mm copper tubes

I have no suggestions on the brazing front, but if its for some sort of heat exchanger why not have tubes of different diameters and put one inside the other. The ends would need a bit thinking about and would probably be best brazed after forming the helix.

Ian Phillips

message

inside the

Ian,

Interesting suggestion, but the coil will form a transformer winding with coolant passing through (up one down the other) so that the flow and return are at the safely grounded end of the circuit. I dont think I could push 5 metres though a pipe !

AWEM

Andrew

If you went the coaxial route the inner tube would not need to be copper. You could use thin wall plastic tubing, PTFE would be ideal.

I agree, pushing copper inside copper would be a tad hard unless both were dead straight, however I've made up custom harnesses containing wires, tubes, cable, and heating elements inside braided conduit some of them over

30m long so PTFE through a rigid tube should be easy.

Ian Phillips

straightener that you can borrow. Run it along the tube and hey presto...straight tube! Doesn't solve the prob of soldering it though.

In message , Clive Dive writes

Use wire as a temporary tie to hold tubes together while you braze section by section. With care the wire should be well clear of the seam and can be snipped away after.

Arrrrrrgh! Problem! Ideally you should pickle [note 1] the tube between each bout of brazing and I have just realised that you will need a 5 metre long trough.

[Note 1] Citric acid is a safe simple pickle, obtainable from your local home-brewing shop - but to buy a large quantity you may need to be prepared to convince the people in the shop that you are not a junkie!

On Wed, 10 Aug 2005 19:38:54 +0000 (UTC), "Andrew Mawson"

Good god you'll put all the lights out in Bromley. Was down your way tonight, collected this

Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

winding

It's only a 15Kw load John - it'll just melt the snow off the pavements over the cable in winter

- - - down my way - . . . good heavens that's >Essex< not Kent - need a passport over that side of the river

AWEM

On Thu, 11 Aug 2005 06:43:26 +0000 (UTC), "Andrew Mawson"

Yes but I still had to stop at Watford Gap, change my money and have my passport stamped.

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

In message , John Stevenson writes

Just as Wogs begin at Calais so does woad begin at Watford - 'twas ever so!

Wouldn't it be better to soft solder the tubes ? I would think there would be problems forming the coil after silversolder/brazing as the whole lot will be fully annealed and ridiculously soft, and probably not consistantly so over the full length. If you used electrical solder there would not be clean up probs either, just dont use lead free solder as it is more brittle than the lead sort. Mark G.

FOR VARIOUS COMPLICATED REASONS I NEED TO BRAZE (OR SILVER SOLDER) TWO

5 METRE LENGTHS OF 6MM COPPER PIPE NEXT TO EACH OTHER TO FORM A 'FIGURE OF EIGHT' CROSS-SECTION, THESE PIPES WILL THEN BE COILED INTO A HELIX OF ABOUT 4\" DIAMETER.

Would it not be better to coil the tubes on a former before joinin them. The two can be a continuous operation

If you want to lay the tubes on the surface of the coil, like a mult start thread, then you start with a round former, possibly with thread form turned into the surface. The former is free at one end an has a crank at the other.

Use two fixed guides to bring the tube from the rolls to the surface o the former aligned with the threadways. Turning the crank will form th tubes into the coil side by side. Like winding thread on a bobbin o two hoses on a reel.

With a 4 inch diameter, you will get about 16 turns in the finishe coil, so it will be about 8 or 9 inches long. You could turn th entire former of that size neatly out of wood. You can then join th tubes either as they are put onto the former, or while they are on th former or after removing the former.

It makes life easier if you hang the source rolls using an inverted Te as a turntable so that the tube comes straight off the roll.

I presume you want to join the tubes down their entire length t increase the heat transfer between the tubes, since once the tubes hav been coiled a simple strut down the outside will hold them in place.

If you want to lay the tiubes as two concentric coils, then you have t form the coil before joining because the inner coil radius will be s much less than the outer coil.

Regards Robi

-- rss

----------------------------------------------------------------------- rsss's Profile:

this thread:

message

consistantly

Mark,

Thanks for the suggestion. It certainly would be easier to soft solder, and in theory they should never get to the temperature where soft solder will melt - but theory and practise have a habit of diverging just at that awkward moment

These tubes will have in excess of 1000A passing through them at times, so if the water cooling fails I want sufficient time before melt down for the safety interlocks to close things down in an orderly manner!

AWEM

snip

I've come in late so I've probably missed important bits but, if I've understood it correctly, this is a bifilar copper tube coil with the two ends joined together at the "high" end and with go and return water connections made at the "low" end.

Provided that the flow rate is sufficient to ensure that the coolant never boils I cannot see any thermal or electronic reason for providing a continous metallic joint between the two helices - any convenient mechanical location method should be sufficient.

Have I missed something?

Jim

Andrew

Another thought occurs to me, If I understand your requirement correctly what you are trying to do is cool the helical tube that is connected to your low voltage, very high current source. You intended to do this by having one tube for coolant thermally bonded to your heating tube.

Why not just use one tube connected to the electrical supply and pump the coolant through it (the same tube)?

Ian Phillips

where

orderly

return

coolant

It's a 'time' thing Jim. The control system needs a finate time to recognise that the cooling has failed and take shut down action. If the supervisory system is set too fast then there will be false trips.

AWEM

coils -

correctly

Ian,

It is a high current HIGH voltage high frequency source - there will be up to 1500 volts AC potential across the ends of the helix when at resonance, so water entry and exit at one (the grounded) end really is essential without going to a recirculating de-ionised water system. Total loss was what I had in mind to avoid the pump etc

AWEM

OK - so the continuous metallic joint is not needed for normal operation but is only intended to increase the thermal time constant to give the supervisory system time to decide that the cooling system has failed.

Two cases seem appropriate

1 Flow rate inadequate

With conjoined helices the maximum copper temperature will be at the inaccessible "high" end

With twin helices the maximum copper temperature will be at the "low" end coolant exit.

2 Total coolant loss

Since both helices are carrying the same current they will both increase temperature at about the same rate whether or not they are in thermal contact. The additional mass of the conjoining medium will slightly slow the rate of rise but it hardly seems worth the effort.

It could be a matter of philosophy. If it's been a long and painful process to produce an effective coolant circulation system it certainly sounds logical to monitor its operation. However, for my money coolant sensors are almost irrelevant. The only thing that counts is the copper temperature. For primary protection I would settle for twin helices with one or more temperature sensors in thermal contact with the copper near the "low" end coolant exit.

Jim