Steve,

 

I’m a big fan of XRF for material identification and plating thickness measurements.  We use our XRF to measure Sn or Sn and Cu or Sn and Ni over steel,  IAg or IAu and ENi or SnPb solder over copper.  The problem with XRF is you need to have calibration samples and you have thickness limitations. For SnPb solder you’re limited to less than 1500-2000 micro inches (1.5 to 2.0 mils) and in my mine the accuracy above 1 mil is questionable.  Because we primarily use the XRF for immersion silver (IAg), immersion gold (IAu), tin (Sn) and solder (SnPb) plating measurements we have many standards for calibration.  The vendors of XRF equipment will tell you their equipment is “good” to +/-5%.  However they usually don’t give you enough information to know +/-5% of what.  We’ve found that we can be within +/-5% of the values we calibrate with.  For example, if we calibrate our XRF for gold over nickel and our standards are 50 micro inches of gold over 400 micro inches of nickel our gold measurement accuracy is at best are +/- 2.5 micro inches and nickel is +/-20 micro inches.  That accuracy isn’t bad considering that that’s about the accuracy of cross sectioning unless you take special care in sample preparation and then use an SEM to measure.

 

Sorry for the long winded answer.  The short answer would have been no we don’t use our XRF to measure copper trace thickness.  We routinely use an optical microscope with X-Y-Z measurement capability at 400x magnification to measure the copper trace thickness.

 

 

-----Original Message-----
From: Steve Gregory [mailto:[log in to unmask]]
Sent: Friday, June 06, 2003 4:13 PM
To: [log in to unmask]
Subject: Re: [TN] Technical Question about Materials Labs

 

Hi George!

Me not being much of a lab person, I got a response to my trace thickness post recently, that XRF is not a good tool to measure copper trace thickness. Is this true?

Thanks!

-Steve Gregory-