Marking sheetmetal

RST Engineering wrote:
You are kidding, of course? You are marking a line 40 mils wide and
expecting a piece to come out to 20 mils or better accuracy?

Jim



"DonMorrisey" wrote in message
ups.com...
At the risk of beating this subject to death, use a sharpie, however a
Fine Point Sharpie is too thick. An Ultra-Fine Point Sharpie makes a
line approx 1 MM in width.

Don...




And if you could put the bend within 10 mils of where you intended every
time, would anyone care?

If the designer calls for that amount of accuracy in anything other than
bearings, it's time to find a new designer. I was told by a old-time
sheet metal worker that unless the plans state otherwise, standard
accuracy requirements are 1/32". That's plus or minus .03125 in either
direction for a .0625" window.

On parts that I do want to cut straight because I want them to be
pretty, I mark with a sharpie against a metal ruler, then aim for the
side of the line. The metal ruler limits where the sponge on the tip of
the marker can go.

"Ernest Christley" wrote in message
m...
RST Engineering wrote:
You are kidding, of course? You are marking a line 40 mils wide and
expecting a piece to come out to 20 mils or better accuracy?

Jim


And if you could put the bend within 10 mils of where you intended every
time, would anyone care?

I care very much. If you got a chassis from me and it had 20 mil gaps in
the corners, you'd think I was the sloppiest designer in the world. All my
freshman students bend within 10 mils on everything to pass the course.
Most of them can hold 5 if they try.




If the designer calls for that amount of accuracy in anything other than
bearings, it's time to find a new designer.

I *BEG* your pardon? In my part of the world, 0.0 is within 50 mils, 0.00
within 20 mils and 0.000 within 5 mils.

Jim

RST Engineering wrote:
"Ernest Christley" wrote in message
And if you could put the bend within 10 mils of where you intended every
time, would anyone care?

I care very much. If you got a chassis from me and it had 20 mil gaps in
the corners, you'd think I was the sloppiest designer in the world. All my
freshman students bend within 10 mils on everything to pass the course.
Most of them can hold 5 if they try.



Oh, no. Now you done gone and did it.

First, I need to make sure we're talking apples to apples. A mil is
1/1000th inch. Consulted http://www.onlineconversion.com/length_all.htm
just to make sure.

1 mil = 0.001 inch [international, U.S.]

I have here in my hot little hands (queue paper rattling for the ditto
heads) one gen-you-wine "RST-443 Panel Mount Intercom", the serial
number being 1014, which went together perfectly and works wonderfully.
I LIKE it, and you can't have it back. Starting at the front right
corner and proceeding counter-clockwise, the gaps in the bends on the
lower half of the chasis measure .020, .030, .032, .035. The bends in
the upper half required a different technique. The edge of an envelope
fit loosely in the gap of the rear corner bends, the envelope measuring
..010.

We must be talking different measurements; either that, or I have
extremely low expectations and you're headed out to chew some hides in
your quality control department. 'Cause, I think this box is tighter
than a cheerleaders butt.


If the designer calls for that amount of accuracy in anything other than
bearings, it's time to find a new designer.

I *BEG* your pardon? In my part of the world, 0.0 is within 50 mils, 0.00
within 20 mils and 0.000 within 5 mils.


I have only a couple measurements on my set of plans that call for a
specific tolerance. One is for the pins locking the gear leg halves
together. The other is for the wing lock pin. The rest are just
measurements with the smallest being with 1/16". Everything broken down
into feet, inches and fractional inches. But no "plus or minus"
anywhere. So I asked a recently retired machine shopman and sheetmetal
worker, because, heh, if you can't trust somebody that's been doing it
40yrs, who can you trust? He told me 1/32" if not specified. Does the
difference come down to how the numbers are specified? Fractional
inches resolve to 1/32, and decimal inches have the tolerances you state?

"Ernest Christley" wrote in message
m...
He told me 1/32" if not specified. Does the
difference come down to how the numbers are specified? Fractional inches
resolve to 1/32, and decimal inches have the tolerances you state?

I've watched this thread with some interest, I like being accurate but I'm
realistic. I have no doubt that Jim Knows what he's talking about. On the
other hand I'm fairly sure I can live with the accuracy that a Sharpie
affords almost any time. The most accurate project I've had recently was
replacing an engraved bezel. it had to inlay with no gaps and had four holes
for push studs. I took no measurements at all. I simply filed to shape, no
gaps, no slop.

"Ernest Christley" wrote in message m...
: RST Engineering wrote:
::
: I *BEG* your pardon? In my part of the world, 0.0 is within 50 mils, 0.00
: within 20 mils and 0.000 within 5 mils.
:
:
: I have only a couple measurements on my set of plans that call for a
: specific tolerance. One is for the pins locking the gear leg halves
: together. The other is for the wing lock pin. The rest are just
: measurements with the smallest being with 1/16". Everything broken down
: into feet, inches and fractional inches. But no "plus or minus"
: anywhere. So I asked a recently retired machine shopman and sheetmetal
: worker, because, heh, if you can't trust somebody that's been doing it
: 40yrs, who can you trust? He told me 1/32" if not specified. Does the
: difference come down to how the numbers are specified? Fractional
: inches resolve to 1/32, and decimal inches have the tolerances you state?

Where I work , unless otherwise stated, we tolerance to:
Fractions ±1/64
0.0 ± 0.010 (10 mils)
0.00 ± 0.005 (5 mils)
0.000 ± 0.001 (1 mil)
We do make assemblies where we hold clearances of 50 micro-inches (µin) (Oh no, not another one!)

Blueskies, you and I are in the same world. When something finished product
is only six inches long with 4 bends and four corner notches, if you can't
hold ten mils your product looks like it was built with Bondo.

As for Mr. CHristly's measurements (and I'm glad he likes the little box; we
sure sell a hell of a lot of them) when he doesn't have the original
dimensioned print in front of him, and is making measurements with
envelopes, how he can say a measurement is off by so many mils without
knowing what it is SUPPOSED to be is beyond me.

If he is referring to the corner gap, I just went out and looked at a sample
of 10 from the last batch of 100. THe corners are water-tight to the point
where I can't see sky through them. How he has a 30 mil gap in his is
somewhat of a mystery.

Jim




".Blueskies." wrote in message
...

"Ernest Christley" wrote in message
m...
: RST Engineering wrote:

Where I work , unless otherwise stated, we tolerance to:
Fractions ±1/64
0.0 ± 0.010 (10 mils)
0.00 ± 0.005 (5 mils)
0.000 ± 0.001 (1 mil)
We do make assemblies where we hold clearances of 50 micro-inches (µin)
(Oh no, not another one!)