Detecting 2 oz copper on PCB

dl324

Joined Mar 30, 2015
16,846
I'll pay someone to change the oil and filter on our car, but won't spend on tools unless there is no choice. In the case of an oil change, occasionally I'll look in there right after to see if they actually put a new filter on, but not always.
I'd be more worried about whether they put the drain plug back on correctly or if they put in the correct type and amount of oil. No kidding. I've had several cases where my Wife got an oil change and oil was dripping from the oil pan because they didn't install the drain plug correctly. Fortunately, no one as stripped the threads (that I'm aware of).

A couple times too much oil was added. They didn't think it was a problem, but I made them remove the extra (they probably spit in it while they were doing it). Most times the oil is half a quart low.
 

joeyd999

Joined Jun 6, 2011
5,237
I'd be more worried about whether they put the drain plug back on correctly or if they put in the correct type and amount of oil. No kidding. I've had several cases where my Wife got an oil change and oil was dripping from the oil pan because they didn't install the drain plug correctly. Fortunately, no one as stripped the threads (that I'm aware of).

A couple times too much oil was added. They didn't think it was a problem, but I made them remove the extra (they probably spit in it while they were doing it). Most times the oil is half a quart low.
Worse is when they slice the CV boot and let you know you need front-end work -- in addition to the oil change. Oh, and we have a special deal today.
 

PeterCoxSmith

Joined Feb 23, 2015
148
It would be accurate to say I'm not suspicious enough, either. I actually don't have a 2 oz board in hand, at least not yet, and have no reason such as anecdotal information to doubt the supplier. I guess I'm a creature of contradictions. I'll pay someone to change the oil and filter on our car, but won't spend on tools unless there is no choice. In the case of an oil change, occasionally I'll look in there right after to see if they actually put a new filter on, but not always.
Ha ha, I get it. Years ago I bought a reasonable precision electronic gauge and have used it many times since. Not specially expensive item but invaluable when you want to know if something will fit (or if you've got the right thickness of copper).
 

dl324

Joined Mar 30, 2015
16,846
Worse is when they slice the CV boot and let you know you need front-end work -- in addition to the oil change. Oh, and we have a special deal today.
That reminds me of a gas station that offered to check your oil and cut your fan belt while they were at it. They had a huge number of belts hanging in their service area...
 

GopherT

Joined Nov 23, 2012
8,009
It would be accurate to say I'm not suspicious enough, either. I actually don't have a 2 oz board in hand, at least not yet, and have no reason such as anecdotal information to doubt the supplier. I guess I'm a creature of contradictions. I'll pay someone to change the oil and filter on our car, but won't spend on tools unless there is no choice. In the case of an oil change, occasionally I'll look in there right after to see if they actually put a new filter on, but not always.
- and only $12

http://sfbay.craigslist.org/nby/spo/5316989463.html
 

Thread Starter

Roderick Young

Joined Feb 22, 2015
408
Maybe this is too simple: Pick a trace, calculate the resistance you would expect if it were 2 oz copper. Measure the resistance.

If you don't have an ohmmeter that goes that low you can make a (for example) 1 amp current source from a heat-sinked LM317 and apply that to the trace. Using your DVM you can measure the voltage drop. All of my DVM's have resolution down to 100 microvolts.

Of course this might not be useful if your PCB doesn't have any traces that are long enough.
If I had been paranoid beforehand, I could have designed in an extra 10-mil trace running the length of the board for that purpose! Your approach is the most scientific, as it measures the parameter of interest (electrical resistance) directly. I don't have a meter with that precision, but it occurred to me that I might be able to put a controlled current through a trace sufficient to generate a temperature rise of X degrees, then I would be relatively independent of trace length. If I had a 1 oz and 2 oz board side-by-side, I might be able to just frost both boards with freeze spray, and count the seconds to defrost a powered trace.

Again, I'm not that curious to put a lot of effort into this, but if I really wanted to know, the most expedient way might be to just go to a friend who does have a good caliper, and take two measurements: one pinching the board near the edge where there is copper and solder mask on both sides, and another where there is only solder mask.
 

joeyd999

Joined Jun 6, 2011
5,237
...but it occurred to me that I might be able to put a controlled current through a trace sufficient to generate a temperature rise of X degrees, then I would be relatively independent of trace length.
That was going to be my suggestion. But I didn't have the time to post at that moment!
 

GopherT

Joined Nov 23, 2012
8,009
If I had been paranoid beforehand, I could have designed in an extra 10-mil trace running the length of the board for that purpose! Your approach is the most scientific, as it measures the parameter of interest (electrical resistance) directly. I don't have a meter with that precision, but it occurred to me that I might be able to put a controlled current through a trace sufficient to generate a temperature rise of X degrees, then I would be relatively independent of trace length. If I had a 1 oz and 2 oz board side-by-side, I might be able to just frost both boards with freeze spray, and count the seconds to defrost a powered trace.

Again, I'm not that curious to put a lot of effort into this, but if I really wanted to know, the most expedient way might be to just go to a friend who does have a good caliper, and take two measurements: one pinching the board near the edge where there is copper and solder mask on both sides, and another where there is only solder mask.

What other concerns do you have that are not worth investigating?
 

cmartinez

Joined Jan 17, 2007
8,220
WBahn's suggestion in post #4 is a good one. If you don't have a sheet of 20# paper you can use a "Typical" human hair, which will be near 3.9 mills in diameter.
:eek: you must have really thick hair!... mine's only 2 mills! unless you measure one of my gray hairs, which are about 3 mills...

Capture.JPG

now, from what other part of the human body can I get thicker hair?... oh oh... :confused::confused::confused:
 
Last edited:
Suppose I ordered a PCB with 2 oz copper from a fab, but wanted to verify that they really used the 2 oz copper that I paid for. Would I be able to tell the difference from 1 oz copper simply by sliding my finger over the traces? If anyone has direct experience with this, I'd love to hear it.
Both varies with their thickness as 1oz has 1.4mils & 2oz has 2.8 mils. In general a copper thickness of 35 µm (1oz) is used on outer and inner layers. If you need to increase the current capacity of a trace you can either increase the trace width or increase the trace height.
 

Thread Starter

Roderick Young

Joined Feb 22, 2015
408
Both varies with their thickness as 1oz has 1.4mils & 2oz has 2.8 mils. In general a copper thickness of 35 µm (1oz) is used on outer and inner layers. If you need to increase the current capacity of a trace you can either increase the trace width or increase the trace height.
Thanks! In fact, I'm going with 2 oz, which is the heaviest thing that doesn't involve special handling (and corresponding pricing).
 

SLK001

Joined Nov 29, 2011
1,549
My old PCB house would only stock 1 oz copper blanks. If 2 oz was specified, they would plate up to 2 oz. If you have plated-thru holes, getting 2 oz copper is not really a big deal (just longer in the plating tank).
 
When the thickness of a sheet of 20# paper is about 3.8 mils. Then can your traces are a minimum of half the thickness of a bit of typewriter paper, then it's probably 2 oz copper. If it's but that, it's probably 1 oz. If you've got regions of bare copper than doing a comparison under a microscope or maybe with a hand glass is pretty easy. If you've got a solder mask then it's tons harder.
 
Top