Thin-walled brass tube

Information on sources for hard to find or unusual items.
Alan Knox
Posts: 4
Joined: Sat Oct 31, 2020 10:47 pm

Thin-walled brass tube

Postby Alan Knox » Mon Nov 09, 2020 2:55 pm

Hello
Can anyone suggest a source for 3/4" brass tube which has wall thickness around 0.3mm?
The (commercial) tube suppliers don't seem tobe able to offer anything with wall thickness thinner than 26swg which is 0.46mm. 19mm is near as dammit to scale for the diameter of my boiler, and would be ideal if I can get it as a tube, but it's a bit of a heat-sink for soldering if it's nearly 0.5mm thick! Or is the only alternative to 'roll my own'?

Alan

Jeremy Suter
Posts: 221
Joined: Tue Apr 14, 2009 6:56 pm

Re: Thin-walled brass tube

Postby Jeremy Suter » Mon Nov 09, 2020 7:26 pm

Hi Alan
I think its a bit of a tall order getting tube that Diameter and that wall thickness I would say that .5mm is a probably the thinnest you can get. Some of the early Brassmasters kits supplied copper tubes with a 2mm wall thickness. which I have used a blow torch to fit smoke box rappers and boiler bands on.

We were only talking about this at the Crewe area group zoom meeting yesterday. Will L was telling me he uses 100 degree solder to fit the brass fittings to tube boilers with a very hot iron, at least 50 watts. These days I tend to throw tubes away and roll my own from .3mm Nickel sheet.
Things to remember when rolling your own boiler is to roll it along the natural curve of the metal. I use half hard sheet and do not anneal the metal as it can cause distortion when rolling if not fully annealed.
When you work out the diameter for the length to be rolled you need to minus the thickness of the material you are rolling twice that is .6mm from the Diameter for a .3mm sheet or .3 off each side. This is because when you roll the metal it stretches round the outside but does not shrink on the inside or if it does shrink it is very minimal.
Other things I do are to scribe where the boiler bands should be which helps to put them on straight either in metal or as a transfer after painting, it also helps to line the sheet up when rolled. I also use some strips along the join on the inside so that when soldered together end to end to gives a stronger joint.

Lindsay G
Posts: 171
Joined: Thu Aug 07, 2008 2:16 am

Re: Thin-walled brass tube

Postby Lindsay G » Mon Nov 09, 2020 9:17 pm

Hi Jeremy,

A rather topical issue for me at present as well. Are you saying it stretches over the width of the diameter but not the length of the boiler piece? Logic tells me it's just the former, but logic doesn't always seem to come into things when I manipulate metal. Mmm, pastry stretches in all directions when rolled.

Getting a bit more into metallurgy, might brass stretch differently from N/S?

As regards rolling along the natural curve of the metal, if the boiler is to form part of an etched sheet, presumably it should ideally be positioned so that the front and rear ends are to the outside of the roll of metal that the etchers will be using (often a 300mm roll). What is the reason for rolling with the curve?

Lindsay

Jeremy Suter
Posts: 221
Joined: Tue Apr 14, 2009 6:56 pm

Re: Thin-walled brass tube

Postby Jeremy Suter » Mon Nov 09, 2020 10:48 pm

A correction to my earlier post its not .3mm I use its 10 thou or .25mm so the numbers need to change from .3mm to .25mm and.6mm to .5mm
One other point I thought about since posting earlier, is roll or curve the outer edges before the middle that is trap the edge and pull round a bar of a smaller diameter than the boiler size required as it is harder to do that part later, the metal will work harden when rolling.

A rather topical issue for me at present as well. Are you say it stretches over the width of the diameter but not the length of the boiler piece? Logic tells me it's just the former, but logic doesn't always seem to come into things when I manipulate metal. Mmm, pastry stretches in all directions when rolled.


Hi Lindsey
Going into a bit more detail.
when you pick up a length of sheet metal it has a natural curve to it, this is due to the rolling process when its made. You can roll it either direction but it rolls easier to go with the curve already there.
I quickly learnt in the early days of rolling boilers that something was wrong. I measured the diameter I wanted, multiplied it by Pi. Cut the metal and rolled it round. Only to find the boiler too wide and needing just over 1.5mm cutting off to get it to the right diameter. When I thought about it, it was .5mm x Pi. and I was using 10 thou sheet or .25mm I still do make them slightly over sized to be on the safe side and file back before soldering the two sides together. Using half hard Nickel Silver which I get from Eileens Emporium the metal stretches on the outer radius but if it shrinks on the inner radius of the curve of the boiler it is very little.
I have only once had a problem with the length of the boiler which did stretch when rolling and it distorted as well. It was a piece of soft grade nickel silver and the rollers had squashed it slightly in the corners.
Getting a bit more into metallurgy, might brass stretch differently from N/S?

I have rolled both brass and nickel boilers and encountering the same problems in both it might be because the nickel silver we use is brass with nickel added which makes it a slightly harder and silver in colour.
As regards rolling along the natural curve of the metal, if the boiler is to form part of an etched sheet, presumably it should ideally be positioned so that the front and rear ends are to the outside of the roll of metal that the etchers will be using (often a 300mm roll). What is the reason for rolling with the curve?

I doubt if it really matters in the end its just easier to roll with the curve. Going sideways to the curve I have not had a problem but going opposite to the curve there is a bit more resistance to the metal it can be done but takes more force.

Alan Knox
Posts: 4
Joined: Sat Oct 31, 2020 10:47 pm

Re: Thin-walled brass tube

Postby Alan Knox » Tue Nov 10, 2020 8:39 pm

Thanks Jeremy. Looks like thin walled tube at such a big diameter is just not available. I'm sure it's technically possible to produce, but suspect there isn't an application for it that we can grab on to by the coat tails!

Alan

User avatar
John Bateson
Posts: 709
Joined: Wed Jul 02, 2008 6:39 pm

Re: Thin-walled brass tube

Postby John Bateson » Wed Nov 11, 2020 11:08 am

On a 5' 6" boiler using 12' (0.3mm) brass the boiler measurements are shown in the attachment with the main boiler shown in red. There is almost a 2 mm difference between the outer and inner circumferences.
Bolier Measurements.PNG
GCR Class 8F TC2020 DL drawing
Slaving away still on GCR stuff ...
Avoiding the soaps ...
https://www.greatcentralmodels.co.uk

davebradwell
Posts: 382
Joined: Fri Jan 18, 2019 3:48 pm

Re: Thin-walled brass tube

Postby davebradwell » Wed Nov 11, 2020 5:49 pm

Yes, the difference will be exactly pi times 0.6 = 1.885mm.

At work, the sheet metal drawing standard assumed a neutral axis 1/3 of the way through the material from the inside and this gave very good results even round multiple bends. I've used this in all my kit design. However, it all depends on what you use to produce the bend and anyone in a workshop would usually bend a test strip first before bending the real part. Jeremy is using the inner diameter to calculate his blank length so it seems reasonable that his rollers might stretch the material a little. All that matters is the result.

It's very easy to calculate the length of a flat metal blank to produce a folded part although to achieve the required dimensions, the bends must be formed to the radius intended in the design and not be like a transition curve. It's good practice to make the inside radius not less than the material thickness but we usually used half this figure. No bending round sharp vice jaws if you want predictable results.

DaveB

Alan Knox
Posts: 4
Joined: Sat Oct 31, 2020 10:47 pm

Re: Thin-walled brass tube

Postby Alan Knox » Thu Nov 12, 2020 9:35 pm

Ooh - I can detect a can of worms opening! If this effect comes into play every time there is a bend radius, does that mean there is a danger of cumulative error where there are several bends on the same part?
I have a specific example in mind: forming the 4 bends at the corners of a tender in 0.3mm brass. The bend radius is 2mm for all bends. So applying Dave B's formula, the increase in length would be due to the effective diameter being 4.2mm instead of 4 and the increase would be 0.15mm or 0.075mm on each side of the bend. But there are four bends to be made progressively - does that mean that cumulatively the tender body would end up 0.15mm bigger in both length and width? Or is the magnitude of the error less because only a quarter segment is involved with each bend?

Alan

Jeremy Suter
Posts: 221
Joined: Tue Apr 14, 2009 6:56 pm

Re: Thin-walled brass tube

Postby Jeremy Suter » Thu Nov 12, 2020 11:24 pm

Alan Knox wrote:Ooh - I can detect a can of worms opening! If this effect comes into play every time there is a bend radius, does that mean there is a danger of cumulative error where there are several bends on the same part?
I have a specific example in mind: forming the 4 bends at the corners of a tender in 0.3mm brass. The bend radius is 2mm for all bends. So applying Dave B's formula, the increase in length would be due to the effective diameter being 4.2mm instead of 4 and the increase would be 0.15mm or 0.075mm on each side of the bend. But there are four bends to be made progressively - does that mean that cumulatively the tender body would end up 0.15mm bigger in both length and width? Or is the magnitude of the error less because only a quarter segment is involved with each bend?

Alan


Well this all depends on how you work it out.

Using .3mm rapper. So to get the length we need.

If you are folding around a shape we know the length width and radii so we get our starting distance say its 100mm by 30mm which is 260mm all the way round if its a square box. But as you are curving the corners you would need to minus those. You said 2mm radius and there are 4 so that would be 16mm ( 2mm off each end of each side) leaving 244mm then you would need to add the curve of the corner which is 4 quarter circles or 2mm radius or 4mm diameter x Pi ( using 3.142 for Pi) = 12.568mm add that to the 244mm gives 256.568mm total length of rapper. but the external length and width is now 100.6mm by 30.6mm with a 2.3mm radius externally.

On the other hand if the external shape has to be 100mm x 30mm with 2mm radius curves we need to minus the thickness of the rapper on all sides so the internal measurements are now 99.4mm and 29.4mm so are starting distance is 257.6mm. Then we need to subtract the corners again 1.7 x 8 =13.6mm leaves us with 244mm then add the corners 3.4mm x Pi =10.6828 add that to 244mm gives us 254.6828mm on the inside the metal will stretch when you bend it to form the 2mm radius corners and give the external measurement of 256.568mm as above.


Just need to note that the length of the sides and ends did not change only the corners do. The difference between the two is 1.8852 as Dave mentioned earlier is 0.6 x Pi.

Just need to remember that the metal stretches on the external part of a curve and always bend round something of the right radius or less as most sheet metal does have some spring in it.

davebradwell
Posts: 382
Joined: Fri Jan 18, 2019 3:48 pm

Re: Thin-walled brass tube

Postby davebradwell » Fri Nov 13, 2020 10:19 am

Yes, you should get accurate results and the only variation is in the bending so it's worth doing a test strip first to see if you've got it right for the way you do it - the strip from the edge of a fret is good for this. You should be able to mark out and drill holes for handrails, etc before bending and get them in the right place. It's also worth marking the start and finish of each bend to assist in setting up. I often use the shanks of drills to form bends and buying these in sets gives a convenient progression of sizes enabling allowance for spring-back. They make good gauges, too, for things like checking slidebars are parallel to frames and equally spaced each side.

Actually for 90 deg bends found in basic bracketry it would be normal to add up all the external dimensions of the part as if it had square corners and look up a bend deduction for each bend in tables in order to arrive quickly at the developed length. There's a load of stuff on-line detailing how to do this and you'll find further variations in the position of the neutral axis.

DaveB

Alan Knox
Posts: 4
Joined: Sat Oct 31, 2020 10:47 pm

Re: Thin-walled brass tube

Postby Alan Knox » Sat Nov 14, 2020 3:54 pm

Thanks for that, gents. Very informative.

Alan


Return to “Where to Find?”

Who is online

Users browsing this forum: No registered users and 1 guest