Return to index: [Subject] [Thread] [Date] [Author]

Re: AISC's Metric equiv. of HSS

[Subject Prev][Subject Next][Thread Prev][Thread Next]
> From: "Sherman, William" <ShermanWC(--nospam--at)>

> Charlie, do you have a "quick trick" to convert hollow structural shapes
> (HSS) from US Customary to Metric? One of our engineers called AISC and was
> told to do a precise conversion and call out to decimal millimeters, eg HSS
> 8x6 becomes HSS 203.2x152.4 (using 25.4 mm/inch). This seems overly precise
> to me. 

> From: Charlie Carter <carter(--nospam--at)>

> > do you have a "quick trick" to convert hollow structural shapes (HSS) from
> US Customary to Metric?
> Unfortunately, I don't There's currently no metric equivalent specifications
> for A500 and other HSS material specifications, so the STI standardized
> their metric conventions as exact conversions as you've already discovered.

We (Canada) use rounded mm values for major dimensions and decimal mm
thicknesses for HSS designation (HSS 203x152x4.8). This still leaves us
with some bizarre numbers. Not nearly as nice as rounding to the nearest
cm (10 mm) for W sections, for instance. I hope that will improve as
North America moves toward hard metric mill sections.

Sounds like STI needs to go back to high school for a primer on
tolerance and significant digits:

A53 (and generally A500) tolerance on outside dimension is +/-1% for
sizes above 2". Very convenient since 0.02" = 0.508 mm. Therefore, a
conversion rounded and written to the nearest mm is within tolerance of
the specification. So, 2 +/-0.02" is neatly described 51 +/-0.5 mm
before we start discussing significant digits in measurement.

Not only is a description to decimal millimetres on the outside
dimension excessive but it becomes inappropriately accurate as the size
increases due to the increasing specific tolerance by percentage.

The situation is similar for wall thickness where A53 permits
+unlimited/-12.5% and A500 permits +/-10%. The minimum nominal wall
thickness for common structural sizes is 3/16" +/-3/160 (0.476 mm) for
A500. So, again, the tolerance starts at about +/-0.5 mm and increases.
Then, 3/16 +/-3/160" is described as 4.8 +/-0.5mm before significant
digits assessment.

So, STI implies that A500 tubes should be described, in metric, to a
degree that is AT LEAST 10 times more accurate than the imperial

Reality is, that if you round all dimensions to integer millimetres on
the drawings, you will probably get what you expected. However, there is
an advantage to using a decimal dimension for thickness to make it
distinguishable from the major dimensions.

That kind of heavy thinking has to be worth some money :-)

Paul Ransom, P. Eng.
Burlington, Ontario, Canada
<mailto:ad026(--nospam--at)> <>