Mixing fractions and decimal units in an imperial drawing

Hi Guys,

Thanks for all your replies. I haven't been on my computer since I posted my question and was surprised at all the responses. I'll have a good think about what appropriate tolerances to put on the drawing, I've got a lot of posts to go through and have a good read

I'll do a bit more on the drawing and then post a picture of it. I bought a set of castings for a Stuart Progress and you don't get a plan, but a facsimile of a construction booklet from 1913. I started to do some sketches of the components, but then decided to do drawings of the components instead. Being a Stuart engine I thought it appropriate to do them in imperial. One of the features of doing the drawings in CAD is that I can do a reprint with some of the dimensions in decimal, so it's easier to use with digital calipers or a micrometer.

Again thanks for your responses.

Regards,

Sarah

Should that be 15/16" being 1/2" and 3/8" and 1/16"

.

What's two hundred and fifty thou' between friends ?

MichaelG.

Sorry Neil. The ones I looked at had the silly markings.

The other useful markings are across the end of the rule. I have a rigid satin chrome one like that and 2 narrow flexibles that don't have this feature. I now have a 12" hardened rigid one on it's way to me. When I looked for one to link to I had a look around. The price of those vary a lot and I feel $26 for nos is pretty good. Shipping a bit expensive but .........

John

-

If you want to follow the Stuart style of drawings than just fractions and no tolerances are needed. The odd "ream ##" is all that is needed. Have you seen any other Stuart drawings to compare with?

J

Coming in late, without having read all the responses, for what they are worth, here are my comments:

If you call for a 3/16 BSF thread call it that rather than than 0.01875 BSF. 3/16 BSF is what everyone else will call it rather than the decimal size.

Are you intending to make just one machine? Or will do you want someone else to make lots, so that the parts are interchangeable?

if the first, then why bother with drawing tolerances, they are not necessary for a one off. You ream/bore the hole, and turn the shaft/piston/whatever to fit. (Depending upon the tooling, and the machine, the chosen dimension could turn out to be exact, undersize, oversize, or even tapered).

For production by someone else, you may well need them, even for a one off.

For mass production with interchangeability of parts, tolerances ARE needed.

Howard

Edited By Howard Lewis on 07/11/2016 11:50:57

That's important information Sarah.

Firstly, it's probably not a good idea to go metric if you are building an imperial model using imperial parts and imperial tooling. All you are doing is adding the risk of conversion errors. (But be aware that someone with an all metric workshop and a metric mind-set might disagree with this. For them it might well be easier to do all the conversions.)

Secondly, many of us work by producing a sketch or drawing of a component before starting work on it. Simple components may only need a sketch but in my opinion anything remotely complicated deserves a dimensioned drawing. Another advantage is that making the drawing often clarifies how to approach making the part. What you are doing is a very good idea.

Thirdly, as the drawing is for your own use, you don't have to adopt 'industry best practice' . In particular, as others have suggested, you won't need to put formal tolerances on the drawings.

Unless you're already comfortable with imperial fractions, I'd suggest decimal notation throughout. This is because digital calipers work that way and you can use one to check the size of drills etc before making a cut.

Another useful trick is to keep a pre-prepared table of conversions needed by the project handy in the workshop. For example, if a plan calls for many 1/8" holes, it's helpful read that 0.125" is 1/8" or 3.2mm rather than stopping work to do mental gymnastics.

By all means share some of your drawings with the forum. I'm sure you'll get lots of advice!

Good luck,

Dave

Mark P.

Bay 1 pipe shop?

Martin

When I was apprenticed 1958 to a small company making scientific instruments ,the drawings were the same as those I I had seen in model engineer magazines, boat and model aircraft books,so had no problems with simple fractional drawings with little tolerancing and holes specified by drill size or ream size, tapped holes were specified as so and so BA,whit or bsf,or brass, the only gauged threads were for eyepieces which had a internal standard.though their was little tolerancing,you were expected to turn fractional lengths to within 5 thou by rule and we did. The rule when producing dwgs was as dimensions get longer avoid using the smaller fractions,there was no dwg manual in the drawing office you just learnt company practice as there was no need for interchangeability except eyepieces and everyone got on with their jobs making good products at an economic price. I then moved to a company making very early automated typewriters and drawings used the decimal system with general tolerances specified on the drawing,never had any problems,and there again a small company with no drawing manual/standards. A further move was quite a culture shock,a big multinational still using imperial and as I was too find out ,a vast amount of specifications and standards covering every thing,and then it went metric. Plus there were volumes of specifications for our suppliers,no personal judgement allowed,just work to spec,plus a very good knowledge of the specs so you could pull a flanker on the inpection and quality personnel who got into a mode of looking for faults rather than, was the part acceptable for production,what that mode of working costs!!.At home I have some all imperial machines,one all metric and I find that I think quicker when working imperial and produce parts quicker.Though looking at the original post ,perhaps we should look at in a way like this the model we build mainly based on vintage items were made to a what is regarded a crude imperial system,the originals worked well some like locomotives were a bit crude ,but had vto work with steel weeks on a steel track and had a reciprocating engine with a sprung crankshaft,I think we get too tied up with measurements and take less interest in the geometry, the coupling rods on a loco have to be free and a bit sloppy to work,but no one seems to get into looking at the geometry , are the axles all parallei to each other,are the cylinder bores parallel to the frames,probably far more important than the actual spacing of the axles.And by the way the US imported rule I bought is crap, .it was divided into thirty seconds and the i/8 ,1/4 and 1/2 lines were all the same length as the 1/32 lines difficult to read,now at the back of the drawer.

Sarah, dependingon what drawing program you are using you may not need to do a reprint, as there is often an option to have a secondary form of dimensioning.

This one is decimal imperial and fractions, note that it has a tendancy to round to teh nearest fraction hence the 3mm (0.118" ) dimension is rounded to 1/8" , the drawing is set to use the nearest 64th.

This one has decimal imperial and metric, which is handy as although the engine is basically being built in imperial you can see I have used some metric thickness plate

J

Jason

Purely out of interest. Are these drawings just examples that you have dimensioned for illustration to Sarah?

It look like the original source data was imperial fractional measurements, but why does 1/8" not convert to 0.125" ?

Ian P

Ian, I just added the extra dims as an example for Sarah. I draw them for my own use and the original print just had decimal imperial as that is my prefered choice.

The horizontal part that is on my original drawing dimensioned as 0.118" is a piece of 3mm plate. If it was 1/8" flat bar then that would convert to 0.125. or vica versa. you can see this on the 5.125" dimension between the feet which has correctly converted to 5 1/8". When choosing fractional dimensions Alibre gives the option of what denominator to use, the list is 2, 4, 8, 16, 32, 64,128 and 256, I happened to choose 64 for this example.

So given that the 3mm plate is 0.118" thick Alibre will print the nearest fractional size which in this case is 1/8". Had I set it to a demominator of 4 then you get the interesting dimension of 0 as .118 is closer to 0.000" than it is to 0.250". alternatively if I set the denominator to 128 then the nearest fraction is 15/128"

Hi,

I have interest from a couple of friends for finished Progress engines and one or two people might like copies of the drawings when I finish them. Also I like to have good references/notes in case I need them at a later date, so I would like to do the drawings as best as I can.

I haven't changed anything on the example in the picture yet, but from the replies I need to show the 1/2" bore as a decimal to some higher degree of tolerance. Most of the other dimensions aren't too critical but the depth of the cover retaining screws and their PCD needs to be controlled so presumably I should do these in decimal? I'd appreciate your thoughts.

Regards,

Sarah

Edited By Sarah Frazer 1 on 08/11/2016 11:49:50

Sarah, this is a typical bit of a Stuart drawing, there is no tollerance on the cylinder that goes with it just 1 1/4 dia

For the holes in a PCD you are getting towards the area where geometrical tolerancing is required on the drawing. Have a look at this tutorial on this subject. **LINK**

For the bore of a cylinder you could use ISO fits such as H11, see **LINK**

As stated earlier though all this is based on the assumption that you are mass producing parts at different times or locations that you want to be able to assemble without additional fettling. For the average hobby machinist working on their own making a one off or two off assembly they will make parts that match what they made the day /week/month/year before.

Do remember though that sometimes words are all that are required. I was looking at a drawing produced this month for real world engineering production that had some holes for studs with the note "Tapping drill must not break through".

Martin

Here is a clip from the drawing I referred to above, slight variance on the wording but the meaning is the same. Also a mix of metric and imperial dimensioning.

Martin

Edited By Martin Connelly on 08/11/2016 13:45:12

Stuarts drawings are a nightmare, I would love to make several of their engines but the drawings are terrible to work to decimal inches no problem it is fractions I have a problem with could not be bothered to finish the SS50 I bought from them.

I know that lots of people on here seen it as a challenge and good luck to them I hope there are enough to keep Stuarts flourishing but model engineering is supposed to be fun I will stick to metric

Hi ,

Jason, I definitely think I'm capturing the Stuart sytle of drawing.

Martin, thanks for the links and the drawing with the notes. I will add some notes as I really hate it when I come back to a project after a big gap and I can't remember details or reasons for what I've already done. Which is all too common with me .

Also I've just started machining on the cylinder and realised that I've missed a dimension from my drawing for the spacing of the ports, ah well back to the drawing board

Regards,

Sarah

I think toleranced drawings can be the worst possible thing to give a beginner for several reasons.

Firstly they can be confusing and even intimidating.

More importantly, they lead the beginner to focus on exact dimensions in ways that slow them down and can even stop them progressing. An example could be a cylinder - how many people post worrying about reaming a cylinder for a steam engine because they need to get it an exact size? Or struggle to measure it accurately?

It's far more productive to tell someone to make the bore as close as they can to the specified dimension but to put their energy into achieving a good finish for the bore. The next step is to make the cylinder piston an appropriate fit using the bore as a gauge.

A good understanding of the whys and wherefores of fits is far more useful in the home workshop than an encyclopaedic knowledge of ISO tolerancing.

Very much my personal view and I know some will totally disagree.

Neil

Edited By JasonB on 08/11/2016 18:22:07