Searching \ for '[EE] converting PCB copper thickness specs' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/pcbs.htm?key=pcb
Search entire site for: 'converting PCB copper thickness specs'.

Exact match. Not showing close matches.
'[EE] converting PCB copper thickness specs'
2007\06\22@155740 by

How does one convert oz/ft2 into mils as it relates to the thickness of
copper on a PCB?

The other day, a question was asked about calculating trace width for a
given current, etc... And one of the sites that came up as an answer belongs
to an ex-piclister who gave me permission to copy his trace with calculation
javascript. I wanted to make some minor changes to it (temperatures in F as
well as C) and feature creep led me to want to automatically convert the C
value to F and back as the units were changed which then led to me wanting
to add that feature for the other unit selectable fields, one of which is
oz/ft2 or mils for copper thickness.

But Google doesn't seem to know how to do that conversion. Any ideas? I seem
to be too stupid to figure it out today.

BTW the new page is at
http://www.piclist.com/techref/pcb_traces.htm

---
James Newton: PICList webmaster/Admin
jamesnewtonpiclist.com  1-619-652-0593 phone
http://www.piclist.com/member/JMN-EFP-786
PIC/PICList FAQ: http://www.piclist.com

>How does one convert oz/ft2 into mils as it relates to the thickness of
>copper on a PCB?

It should be converted thru copper density, which is 8 920
kilogram/cubic meter = 8 910 oz/ft3.

>to add that feature for the other unit selectable fields, one of which is
>oz/ft2 or mils for copper thickness.

Yet another way is thickness in micrometers.

Josef

Hi James and Josef,

My notes show 1.4 mils (thousandths of an inch) per oz/ft2. I think I
got that from one of the online PCB calculator sites.

Sean

On 6/22/07, Snail Instruments <snailsnailinstruments.com> wrote:
{Quote hidden}

> -
> >How does one convert oz/ft2 into mils as it relates to the
> thickness of
> >copper on a PCB?
>
> It should be converted thru copper density, which is 8 920
> kilogram/cubic meter = 8 910 oz/ft3.

Errr.. Huh? Yes I really am that dumb when it comes to this sort of thing.

Let me try again:

If I have copper thickness in mills, what mathematical formula yields
oz/ft2?

If I have copper thickness in oz/ft2, what mathematical formula yields
mills?

> >to add that feature for the other unit selectable fields,
> one of which
> >is
> >oz/ft2 or mils for copper thickness.
>
> Yet another way is thickness in micrometers.

K. Give me a set of formula and I'll add another option.

Thanks for trying to help.

---
James.

James Newton, Host <jamesnewton <at> piclist.com> writes:

> How does one convert oz/ft2 into mils as it relates to the thickness of
> copper on a PCB?

convert oz/ft^2 -> grams/cm^2

then insert copper density (from Wikipedia:copper): rho = 8.96 g/cm^3 @ 20C

so the thickness is: h [cm] = grams / ( rho * 1 cm^2 )

convert to mils or pounds or carats or whatever.

Peter P.

Hi James,

volume density=area density/thickness

therefore:

thickness=area density/volume density
area density=volume density*thickness

Using Josef's numbers, we have:

thickness (in feet)=area density (oz/ft2)/8910 (oz/ft3)

For 1 oz/ft2, this gives 1.12e-4 feet or 1.34 thousandths of an inch

To put the formula itself into thousandths, multiply by 12000:

thickness (mils)=1.35*area density (oz/ft2)

The reverse formula would then be:

area density (oz/ft2)=thickness(mils)/1.35

Sean

On 6/22/07, James Newton, Host <jamesnewtonpiclist.com> wrote:
{Quote hidden}

> -

>If I have copper thickness in mills, what mathematical formula yields
>oz/ft2?
>
>If I have copper thickness in oz/ft2, what mathematical formula yields
>mills?

oz * 1.4 = mils

mils / 1.4 = oz

James Newton, Host wrote:
> But Google doesn't seem to know how to do that conversion. Any ideas? I seem
> to be too stupid to figure it out today.

Sure it does. Here's the link to get the conversion constant - note the
lack of any explicit calculations (save for the inverse - take it out
and reverse the output units for the inverse conversion constant);
Google figured the units out itself.
http://tinyurl.com/2rklt7
(where 8.96g/cm^3 is the density of copper)

Here's how you go one way:
http://tinyurl.com/2ubxn8
And here's the other:
http://tinyurl.com/2udow5

Of course, you have to input the density of copper with proper units,
but you can't expect Google to assume you're dealing with copper!

As an aside, I've found Google to be a very good aid when doing
technical calculations like these, not only because it auto-handles
units (which is useful but can get strange sometimes), but because if
you mess up the units anywhere it will either not return an answer or
give out the wrong units.

--
Hector Martin (hectormarcansoft.com)
Public Key: http://www.marcansoft.com/marcan.asc

Thanks very much Sean, I've updated the program.

---
James.

{Quote hidden}

On 6/22/07, Snail Instruments <snailsnailinstruments.com> wrote:
>
> >How does one convert oz/ft2 into mils as it relates to the thickness of
> >copper on a PCB?
>
> It should be converted thru copper density, which is 8 920
> kilogram/cubic meter = 8 910 oz/ft3.
>
> >to add that feature for the other unit selectable fields, one of which is
> >oz/ft2 or mils for copper thickness.
>
> Yet another way is thickness in micrometers.

Which is the correct way used even on clever american PCB houses
(which are trying with efforts to change everything on metric system).
see:
http://www.pcblibraries.com/forum/
or
http://www.pcblibraries.com/resources/GEN-docs.asp

0.5 oz = 17.5 um = 0.7mil (rounded value)
1 oz = 35 um = 1.4mil (rounded value)

Vasile
Hector Martin wrote:

> As an aside, I've found Google to be a very good aid when doing
> technical calculations like these, not only because it auto-handles
> units (which is useful but can get strange sometimes), but because if
> you mess up the units anywhere it will either not return an answer or
> give out the wrong units.

Which is of course a general advantage of doing calculations with the
proper units throughout (including conversion factors). In many cases,
you'll see immediately when something is wrong. I'm glad that I was
introduced to physics in high school this way and have it done like this
ever since.

Gerhard

The units (grams, ounces, feet, meters, or whatever) should be carried
along with the numbers, and then crossed out when able to cancel. Then
remaining will be ONLY  the final units, which must be what you are
looking for. :)

Gerhard Fiedler wrote:
{Quote hidden}

Gerhard Fiedler wrote:
> Which is of course a general advantage of doing calculations with the
> proper units throughout (including conversion factors). In many cases,
> you'll see immediately when something is wrong. I'm glad that I was
> introduced to physics in high school this way and have it done like this
> ever since.

Of course - this is the way I do it. However, you can make mistakes with
units in much the same way you can mess up numbers when doing things on
paper. For example, you could inadvertently flip a unit around when
cancelling things out (essentially taking the reciprocal). Google is to
calculations with units as a calculator is to calculations without
units. Sometimes there just isn't enough time to do it carefully, too -
with today's education, sometimes there is way too much
formula-crunching. Google serves as a nice doublecheck on your formula
units vs. operations, as well as calculating the answer at the same time.

As for conversion factors, I try to use SI throughout most things to
avoid them - if you do the full copper thickness calculation by hand,
you'll see most of the work is due to the use of funky imperial units.

--
Hector Martin (hectormarcansoft.com)
Public Key: http://www.marcansoft.com/marcan.asc

Hector Martin wrote:

> As for conversion factors, I try to use SI throughout most things to
> avoid them

Unluckily it seems that Google Calculator doesn't understand a number of
proper SI units, at least not the standard abbreviation.

> if you do the full copper thickness calculation by hand, you'll see most
> of the work is due to the use of funky imperial units.

Yes, that's why I have somewhere stored in non-volatile memory that 1 oz of
copper is roughly equivalent to 35 um thickness :)

Gerhard

Carl Denk wrote:
> The units (grams, ounces, feet, meters, or whatever) should be carried
> along with the numbers, and then crossed out when able to cancel. Then
> remaining will be ONLY  the final units, which must be what you are
> looking for. :)
>
>
The problem I find with trying to do this is that in many cases the unit
symbols are the same bloody letters as the variable names. Maybe some
people have handwriting where upright and italic can be distinguished
but I sure as hell don't.

If need be just spell out long  hand, but the method was one of the
first things I learned as a freshman at Univ. of Mich. in 1960, and I
still do it. :)

peter green wrote:
{Quote hidden}

peter green wrote:

>> The units (grams, ounces, feet, meters, or whatever) should be carried
>> along with the numbers, and then crossed out when able to cancel. Then
>> remaining will be ONLY  the final units, which must be what you are
>> looking for. :)
>
> The problem I find with trying to do this is that in many cases the unit
> symbols are the same bloody letters as the variable names. Maybe some
> people have handwriting where upright and italic can be distinguished
> but I sure as hell don't.

I've never found this to be a problem. A unit never comes alone; it always
follows a number. A variable never follows a number without an operator in
between. (Ok, there's the math shortcut of "6 A" that could mean "six times
the variable A", but usually the context makes clear what is meant, and if
not, you can make the implied multiplication operator explicit.) Also
normally I either work in variables (on the right side) or in numbers;
rarely mixed.

Gerhard

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...