Can we get rid of silly dimensions, please?

[JST]

I run into silly and unprofessional dimensions that don't go away.

Things like 3' 12".

To any half intelligent semi-human, that is simply 4'. But Alibre insists that it is 3' 12" (which it technically is, of course, it is a sort of "semantics" issue). I have several of these in a drawing right now, in the materials list, and I can't get them to go away. I do not want to edit them, as then they will not be linked to the actual part anymore, and I will have to do any updates manually.

Naturally, in this actual part drawing, they are fine, 4'. It is just the materials list that is fouled up, reason for the difference is unknown. However they do show up in part dimensions also sometimes. I have ca dimension on many parts labeled as "length" and the BOM set up to take that named field and put it in a column of the BOM for each such part.

EDIT: I take it back, I also found an 11' 12" dimension in a part drawing Oddly, THAT one is listed as 12' in the BOM. Go figure.

Not only is this crazy dimension silly and unprofessional, but it is also potentially misleading. The person using the print may read it as 3' and some number of inches, and get it wrong. (Of course they should be careful. But Alibre should not make that dimension to begin with.

I assume this is a byproduct of rounding. If the real dimension was 3' 11 63/64", and you limit the resolution to 1/32" or 1/16" (which are considered "precision" by ironworkers), then the rounding up to the next inch could come out as 3' 12". I see that. I don't like it, but I understand the issue.

Checking that, I find that yes, it IS apparently rounding, as the part comes out as 11' 11 255/256", if I set the resolution to 256ths of an inch (an insane fractional value that nobody in the world uses, BTW). So Alibre rounded it up, came out to an even value, and used that without checking if it needed to increment the major units instead.

But it is so silly that it ought to just have a filter on it to put the rounded figure up as the even number of "units" (feet in this case) that the as-rounded number really is. That would be much better, more readable, less confusing, and far more "professional".

I have no idea if similar things occur in the Metric calculations. I have never seen it, but with ironwork in USA, I don't see much metric at the moment.

 

[bigseb]

I have no idea if similar things occur in the Metric calculations. I have never seen it, but with ironwork in USA, I don't see much metric at the moment.

You won't see this in metric because we don't work like that. By which I mean we don't dimension things as, say, 1m 250mm. We'll use only mm, m, etc.

 

[JST]

Builders are differnt, here, and elsewhere, but perhaps so. 1.25m would be fine for your example.

Although I am not sure if that extends out to all. If you want 25m over 5 km, does one really say 5000.025 m? The point is likely academic.

Regardless, it isn't going away soon in this area, whatever we may think about it. And the dimension really looks stupid on an otherwise standard drawing.

 

[jfleming]

Interesting.. I only dimension in inches so I have not seen that issue.

Years ago, I was an inspector for highway construction. One of the jobs was all calculated/planned out using the metric system, which I love. Plans would call for a 30.6m pipe run. The thing is though, as you stated JST, feet and inches are not going away, all of the building materials are in feet and inches... The pipe lengths, the guide rail panels, the signage, etc. so it became a bunch of horse hockey saying "well they laid 10 joints of pipe at 8 feet long, so that would be 24.38m , instead of simply "80 feet of pipe"

Same for dimensional lumber, etc.

Why the US ever felt we should use feet and inches is beyond me!!

 

[JST]

Metric has been legal for trade since about 1870 in the US.

That is now about 150 years ago that the US gave metric the green light.

If we didn't change in 150 years, it's 'cuz people did not want to very much.

The issue of the stupid round-off problem is, however both a current issue, and something that can likely be fixed fairly easily. Anyone else who has to work in the "feet and inch/fractional" realm must run into the same issue, so it is not "just me".

 

[NateLiquidGravity]

I've had this issue for years. I work in architectural structural steel in the US.

 

[JST]

Do you edit the affected dimensions?

I've done that, and I have just changed the part dimension to make the problem go away. I do not like either "workaround".

 

[NateLiquidGravity]

I have text edited them.

 

[DavidJ]

One option , though not perfect if this happens is to set units for that dimension to 'feet' and set precision to zero.

A ticket has been raised for Development to look into sorting out the 'carry' from inches to feet when rounding causes this issue.

 

[idslk]

they laid 10 joints of pipe at 8 feet long, so that would be 24.38m , instead of simply "80 feet of pipe"

...or 5 joints of 5m each, that would be 25m, instead of 82,021 feet...
But no matter from which side you're viewing should not allow rounding fails, so it's good from David to raise a ticket.

Regards
Stefan

 

[bolsover]

Dumb politicians in the UK have recently suggested reintroduction of imperial weights and measures. I'm old enough to remember when we had Pounds, Shillings and Pence. I even remember buying petrol (gas) when it was 2 shillings, six pence and three farthings for 1 gallon (imperial). Please use international system of units wherever possible - it is so much easier!

 

[JST]

Dumb politicians in the UK have recently suggested reintroduction of imperial weights and measures. I'm old enough to remember when we had Pounds, Shillings and Pence. I even remember buying petrol (gas) when it was 2 shillings, six pence and three farthings for 1 gallon (imperial). Please use international system of units wherever possible - it is so much easier!

Si is preferable. Not practical to use anything else for scientific or most engineering purposes. ("Kips", anyone?)

Ironworkers are stubborn folks, however, and normally are large enough animals that an argument is going to involve a good deal of work....... So we use what they use. Carpenters, and contractors the same, although they may be smaller. The woodshop superintendant at a prior employer once said "you can't measure that in millimeters"....... need I say more? (incidentally, we DID, the CNC router happily took mm).

Problem is that the US is so large that there is an absolutely unimaginable amount of "installed base" of everything from wood and steel buildings through plumbing, and on to machine tools etc etc that are all in the units that came over from the UK centuries ago. Unlike much of continental europe, everything was unharmed through the two wars, so it's all still here.

The usual system of "thou shalt use SI or suffer a public whipping" is impractical. The best that can be hoped for is that the same parts are referred to by a converted set of units. Standardizing on increments of 10, 25, 50 etc mm , while convenient, will not fill the bill.

Consequently, The use of feet, inches and fractions is likely to continue, and those of us dealing with it will be doing so for quite a while yet.

 

[bigseb]

The woodshop superintendant at a prior employer once said "you can't measure that in millimeters"....... need I say more? (incidentally, we DID, the CNC router happily took mm)

That's why we can pick our units, depending on the job. My cousin's a carpenter and according to her they work in cm. Same system, just bigger increments. We don't mix them though i.e. 155mm isn't ever referred to as 1dm 5cm 5mm. However its easy to cross-reference the one to another eg 15.5cm = 155mm. This makes metric a super easy system to use.

I worked with imperial during my time in the UK. A totally messed up system, imo. I'm fine with inches but then come the fractions and that's a totally illogical way to measure and dimension things. None of the machines or measuring equipment have fractions on them so its a case of constantly converting back forth between fraction and decimal. Then came the drills that have neither decimals nor fractions but rather a number eg #36. Unless you grow up with a system like this and know it well it is, as I said, completely illogical.

All this is academic though. JST is spot-on in pointing out that rounding off a dimension as 1'12" (or whatever it was) is not right and should be addressed.

 

[DavidJ]

This issue occurs when using fractional dimensions and the denominator value for the dimension in the 2D drawing is set to a lower value (coarser) than that used for the model design, or if model is designed in decimal, then drawn using fractional dimensions.

I'm puzzled by why people would do that, maybe there is some reason I've not thought of, or some indirect workflow that leads to this situation. Is perhaps the length of the part derived from combining other parts (e.g. diagonal of a rectangle in some structure)? - I can see that there the actual length might not match the fraction setting used for the original parts.

Understanding this might help to assign an appropriate priority for the ticket to sort out the 'carry on rounding'.

 

[JST]

This issue occurs when using fractional dimensions and the denominator value for the dimension in the 2D drawing is set to a lower value (coarser) than that used for the model design, or if model is designed in decimal, then drawn using fractional dimensions.

I'm puzzled by why people would do that, maybe there is some reason I've not thought of, or some indirect workflow that leads to this situation. Is perhaps the length of the part derived from combining other parts (e.g. diagonal of a rectangle in some structure)? - I can see that there the actual length might not match the fraction setting used for the original parts.

Understanding this might help to assign an appropriate priority for the ticket to sort out the 'carry on rounding'.

The reason I design in 64ths, and do the drawings in 32nds, is really to detect small errors, and be sure things are fitting.

By going one level further, I can see what the rounding is probably going to look like. And if I have a choice in dimension, then say the thing turns out to be 33/64", I can change to a plain 1/2" and know the maximum resulting error.

Finally, the 1/64" increment is enough smaller than the typical 1/16" oversize of fastener holes. The part will be made to the printed dimension. So if I know that the distance between holes of the part in the model is no more than 1/64" different from the drawing marked size, I can be reasonably confident that the oversize allowed will let it fit.

Ironwork is not high precision, but it is expected that the parts will fit.

Also, the program seems to "pre-round"... if you derive a dimension from another, sometimes it seems that the dimensions are not equal.... the original is at some dimension that is not on even fraction, but the copy seems sometimes to be on the fraction, and NOT on the original dimension. Not sure if that explanation is clear.

 

[NateLiquidGravity]

I design in decimals because I prefer it. I dimension drawings in fractions because that's what's on all the tape measures in the shop. It's silly to have to defend such a thing.

 

[DavidJ]

JST , Nate - thanks both for the detail, both were variations that hadn't occurred to me.

 

[JST]

Another example "just to grind it in"...........

I might add that this is not a case of the dimension style being changed, and obviously is also not in a drawing. It just popped up when measuring the assembly. Package included.

I don;t know if it is the same mechanism as in the other cases where the numbers turned up in drawings. Maybe this is useful, maybe not, as far as determining where/how/why it occurs.