Pprojecting a circle onto a boiler barrel

[Vincent Shaw-MortonParticipant @vincentshaw-morton14296]

Hi everyone,

I’ve started on my first loco and have just finished flanging all the boiler plates, and I’m quite pleased with the results. Anyway on looking at the dome bush I can’t for the life of me figure out how to project the cylinder diameter of the bush, ( a circle of course) down onto the boiler barrel. Simply scribing a circle on the barrel with a compass will of course distort the circle. How do people approach this (what appears to be simple) but for me baffling problem?

thanks

Vincent 

Edited By Vincent Shaw-Morton on 07/12/2010 21:05:27

Edited By Vincent Shaw-Morton on 07/12/2010 21:05:53

Edited By Vincent Shaw-Morton on 07/12/2010 21:06:26

[Vincent Shaw-MortonParticipant @vincentshaw-morton14296]

[WeldsolParticipant @weldsol]

H i Vincent

I normally set the barrel on the mill (having alreadt marked the centre) then hole saw followed up with the boring head.

 

Paul

[JasonBModerator @jasonb]

Did this quite early on at school technical drawing, its refered to a circle development. Its quite easy to do on a bit of paper with a simple drawing board, maybe search the net as its not easy to explain in words.

 

But as said above if you can set it up someway to bore on a machine then its easy.

 

Jason

[John OlsenParticipant @johnolsen79199]

I suspect we are making this a lot more complex than it really needs to be. The “circle” drawn with the compass will be the right size fore and aft, and will be under size at either side. So, assuming a minimum of tooling, eg no boring head or whatever, we mark out the hole with the compass, chain drill carefully just inside and then clean up with a file. Try the bush and file away where needed until the bush is a nice fit. Silver solder is better if the fit is not too snug, 2 or 3 thou clearance is good.

 

 

regards

John

[Les Jones 1Participant @lesjones1]

Hi Vincent,

                    You could make a tool similar to the hole cutters that Toolstation sell but with just one blade sharpened as a scriber. To see a picture of the hole cutter on Toolstation’s website enter “hole cutter” in the search box. If you drill a pilot hole and mount the mount the tool in a pillar drill or some kind of guide you can let it move up and down to keep the scriber in contact with the surface as it is rotated. I used this method to cut a hole for the burner in a furnace that I was making. For that the scriber had to be quite long at the hole was tangential. I was cutting a hole about 2.5″ dia. in a cylinder of about 13″ dia. It has just occurred to me that if you use two blades then keeping both blades in contact with the surface would remove the need for a guide IF the axis of the hole being cut goes through the centre of the main cylinder.

Les

[KWILParticipant @kwil]

If you look in My Album, you will see a FC3 3 flute throwaway cutter making the holes in a boiler barrel by chain milling [like chain drilling but you can do it on a curved surface].  Once done this way, you can open it out to exact size by use of a boring tool.

Edited By KWIL on 08/12/2010 10:21:34

[Ian S CParticipant @iansc]

I needed two 0ne inch holes in the side of a piece of two inch copper pipe, so I made a one inch hole saw. Heated an inner race from a ball race of suitable size,welded a shaft with a flange onto the peice that was to be the cutter, put it in the lathe and took it down to 1″ dia, then filed some teeth on it, then rehardened it, it worked well and has since been used on steel. I had tried drilling and fileing, but mucked it up.isc

[Clive FarrarParticipant @clivefarrar90441]

There are two ways of doing it.>>

 >>

on the centre line of the boiler where you want the flange to go mark the true OD of the flange on the line.>>

Then using a circular object or hole and a very stong light project a shaddow onto the line on the boiler, at 90 degrees to the line on the boiler. Adjust the distance of the object untill its shaddow touches both of your marked points on the centre line. Then draw in the rest of the shaddow out line and that is the development of you pipe projected onto the boiler.>>

 >>

If you want a paper template to follow have a look at the attached diagram. It should be easy to see how the pipe has been divided up into sections and then those lines are projected in two directions to get points of intersection on the boiler, the brown dots in the diargam.>>

Link the dots and that is your outline.>>

The diagram is basic so you dont get a good set of points to join up. Easily solved just add more lines and intersections to fill in the blank bits.>>

 >>

I hope that helps regards Clive

>>

[Vincent Shaw-MortonParticipant @vincentshaw-morton14296]

Thanks for your help everyone, there are some great solutions here that I can use

Best

Vincent 

[Clive FarrarParticipant @clivefarrar90441]

Hi Vincent,

                   its been bugging me all day that my diagram may NOT correct. As its 30 odd years since i last did this sort of thing I have missed a bit out.

So I think if you use this method you need to test it out on something unimportant like a toilet role core.

 

So what have i done wrong?   the red dot blue dot on the circle is the boiler diameter but when you develop that out into the flat it becomes half the circumference i.e 3.142 times longer.

Some how there needs to be an intersect line intersposed between the boiler circle and boiler development such that as you transfer the lines across they correct the positioning by a factor of Pi  but at the moment I can not remember how.

Not sure if I have made that better or worse but it gives you a starting point for a net search.

 

If you give me diameters and aproximate location I will try and get some time to do it properly but with work at the moment i can not promise.

 

Regards Clive

[JasonBModerator @jasonb]

Have a look at this google book, about 6 pages in it gives you what you need

 

J

[John OlsenParticipant @johnolsen79199]

Clive, what your projection produces is the view of what the tube would look like while it is round. If you want to develop the shape of the tube as it would be if flattened out you have to do more work. You could draw radial division in the circle on the right, then project them across your developed shape. The radial lines would be getting closer together there as you get further from the centre. You then draw a similar grid of lines, but with the horizontal lines at a constant spacing, equal to the circumferential distance between each radial line. Transfer the sizes from the original set to the new set, and it will stretch the shape at the ends by the right amount.

 

However, unless the plan is to cut the hole and then roll the tube shape, which I would not advise, that approach should not be needed. The problem is not to find what shape the hole would be if the tube were flattened, it is to make a hole which is reasonably close to round in the actual tube

 

The other thing is your smaller tube seems to be meeting the larger tube at a tangent. The original problem was for a dome bush, which would mean that the centre lines of the two tubes (extended) would meet. I did a tube like yours once for a model water tube boiler, of a simple Babcock and Wilcox type. Simple because it only had one bank of tubes, which used a cross tube across the drum for a header. I made the semi elliptical cutout by just filing until I had a good fit…

 

regards

John

[Clive FarrarParticipant @clivefarrar90441]

John and All as stated I realised my  original drawing was not the full story.

 

And yes it does look like a tangent application but it was the only way i could get the template profile that would need to laid over , curved round the boiler to scibe a line that when cut would be the circular hole required.

 

Rather than fiddle about and misslead some more i had a quick look on the web and came up witht he following link.

 

http://www.alamo.edu/sac/engtech/CADD/Rstrube/TECHNCAL_DRAFTING/Student%20Tutorial%20Power%20Points/CH23.PPT#293,1,Slide 1

 

about half way through it does the intersection of two cyinders.  That is what i was trying to convey but failed to do so correctly.     I was close though

 

Regards Clive

[Author]

Posts