I'd like to talk a little bit about gold plating thickness. First, a history, as I saw it.
One of the main purposes of gold plating is to protect the part with a non-corrosive coating. In the '30's and '40's the electronic companies were still experimenting with how much was needed to provide this protection. As a result, they overplated, just to make sure. Gold was cheap back then and the parts were expensive. Also, the gold plating technology was not very advanced. The old cyanide based gold coatings were very porous and it took more gold thickness to seal the pores. I haven't seen much scrap from that era but, one memorable case sticks out in my mind. We used to have a customer that had a buried cache of old stuff, mainly pins, somewhere near the Texas/New Mexico border. Once or twice a year, he would load his pickup with scrap and bring it to us for refining. The pins ran an ounce of gold per pound. They took forever to strip.
When I got into the business, in the mid '60's, new innovations in gold plating had already reduced the needed gold thicknesses and the manufacturers had settled on how much gold was necessary. The government had studied the problem and had written Mil-Specs, which spelled out gold thickness and the types of gold required for specific types of military parts. Although they didn't have to, most manufacturers followed the Mil-Specs and, still do today. And, even though the gold was controlled at $42/oz., it was still expensive and the manufacturers didn't waste it. Believe it or not, these thicknesses still basically hold true today. There are other reasons why gold scrap is worth less.
If I remember my dates, the government allowed the gold price to float in 1972. The price immediately started climbing and the manufacturers started thinking of ways to cut their gold costs. They did so using two main methods. First, they looked for gold substitutes on less vital parts. Sometimes, they redesigned the parts to fit a substitute. Second, the equipment makers came up with selective platers. Instead of plating the whole part, they only plated the portion of the part that required plating. For example, on backplanes, you'll often see pins that are plated only on the contact areas. Look around and you'll see a lot of examples of selective plating. The value of gold scrap started going down.
In 1980, the gold price skyrocketed and, again, the manufacturers started looking at the gold costs. This brought about even more innovations. The value of gold scrap suffered another hiccup.
Today, with the gold price high again, I'm sure the manufacturers are figuring out devious ways to reduce the values of our precious gold scrap. In one way, though, they have helped us. Today, there's a lot more platinum group used in electronics than ever before. The problem is that working with the platinum group is much more difficult than working with gold. And, it's harder to spot. Gold sticks out because it is yellow. There are only two metallic elements with color, copper and gold and their alloys. All others are white or some shade of gray. Silver is the whitest metal. German silver, also called nickel silver, is an alloy made to look like silver. It contains no silver.
Gold plating thickness is measured in microinches - millionths of an inch. You can't tell the gold thickness by looking at it. However, very thick gold is usually matte in appearance. Thin gold can reflect the brightness of the shiny nickel underneath it. The thickest normal electronic plating is usually found on parts that require heating further along in their assembly, such as those that use a gold/silicon braze for chip attachment. Examples are side braze, CPU, or hybrid circuit packages (Google). They run about 50-60 microinches of gold. There are parts that require thicker gold (as much as 500 microinches) but, most are military and are rare. The thinnest electronic gold you will probably find is about 12 microinches. Some cheap costume jewelry only has 5 microinches. Much electronics is intermediate between these two extremes, about 20-30 microinches. This intermediate level is usually found on parts that require multiple insertions, such as fingers or connector pins. These figures apply to products that the manufacturer wants to last a long time. On cheapo products, the thickness will vary but, always on the low side.
The above is only a average guide. You'll notice that I always use vague terms like: often, generally, sometimes, usually, etc. The only way to absolutely know values is to assay or refine the parts. I have seen fingers that run from $10/# to $150/#. The gold on packages, however, usually runs constant. Although not always reliable, due to the internal stress of some hard gold deposits, one way to estimate relative thicknesses is to put the part in a 25-50% nitric solution. Hard gold is used on insertion parts, such as fingers. As the copper and/or nickel, under the gold, dissolves, the gold flakes. Generally, the smaller the flakes, the thinner the gold. Also, thinner gold is more porous than thick gold. Since the acid penetrates through the pores, the metals under thin gold generally is attacked faster than that of thick gold. If you do play around with the nitric, it is best to run stuff side by side, to get a comparison. If you refine the items, make notes so everything is more meaningful.
Another way to estimate plating values is to calculate the surface area (number of square inches) of the plated part. This is the way that parts are plated. The plater calculates the area of the load and then sets the amperage based on the type plating bath that is being used. Gold bath current densities usually range from 3 to 10 amps per square foot (asf) of surface area. The thickness depends upon how long the parts were in the plating tank. It usually takes about 15 minutes to plate 100 microinches thick. For 60 microinches, it takes 9 minutes. The measurement of surface area takes a little math. Here are some surface area formulas:
Surface Area and Volume
Take the present gold market price, say $650, and move the decimal point three place to the left. This equals 0.65 or, 65 cents. This is very close to the value of one square inch of 100 microinches of gold. If you had 60 microinches of gold, the value per square inch would be .65 X .60 or, 39 cents per square inch. For 25 microinch gold, .65 X .25 or, about 16 cents. This is especially handy when you have a lot of identical parts, such as those boards with 1000, or so, 1" long pins. Determine the surface area of one pin and multiply by 1000 or, whatever. For more accuracy, use calipers, micrometer, or an optical comparator to make the measurements. Write down the measurements and make the calculations.
This isn't rocket science and nothing is exact. And, there are numerous exceptions. However, I know from experience that the above info can put you in the ballpark. Even with ballpark figures, you will probably know more than the seller. In this business, more knowledge of the material always puts you ahead in the buying/selling game. And, it is a game. No one really knows the true value except for the one who refines the material. That's why the refiner is called the "Last Liar".
One of the main purposes of gold plating is to protect the part with a non-corrosive coating. In the '30's and '40's the electronic companies were still experimenting with how much was needed to provide this protection. As a result, they overplated, just to make sure. Gold was cheap back then and the parts were expensive. Also, the gold plating technology was not very advanced. The old cyanide based gold coatings were very porous and it took more gold thickness to seal the pores. I haven't seen much scrap from that era but, one memorable case sticks out in my mind. We used to have a customer that had a buried cache of old stuff, mainly pins, somewhere near the Texas/New Mexico border. Once or twice a year, he would load his pickup with scrap and bring it to us for refining. The pins ran an ounce of gold per pound. They took forever to strip.
When I got into the business, in the mid '60's, new innovations in gold plating had already reduced the needed gold thicknesses and the manufacturers had settled on how much gold was necessary. The government had studied the problem and had written Mil-Specs, which spelled out gold thickness and the types of gold required for specific types of military parts. Although they didn't have to, most manufacturers followed the Mil-Specs and, still do today. And, even though the gold was controlled at $42/oz., it was still expensive and the manufacturers didn't waste it. Believe it or not, these thicknesses still basically hold true today. There are other reasons why gold scrap is worth less.
If I remember my dates, the government allowed the gold price to float in 1972. The price immediately started climbing and the manufacturers started thinking of ways to cut their gold costs. They did so using two main methods. First, they looked for gold substitutes on less vital parts. Sometimes, they redesigned the parts to fit a substitute. Second, the equipment makers came up with selective platers. Instead of plating the whole part, they only plated the portion of the part that required plating. For example, on backplanes, you'll often see pins that are plated only on the contact areas. Look around and you'll see a lot of examples of selective plating. The value of gold scrap started going down.
In 1980, the gold price skyrocketed and, again, the manufacturers started looking at the gold costs. This brought about even more innovations. The value of gold scrap suffered another hiccup.
Today, with the gold price high again, I'm sure the manufacturers are figuring out devious ways to reduce the values of our precious gold scrap. In one way, though, they have helped us. Today, there's a lot more platinum group used in electronics than ever before. The problem is that working with the platinum group is much more difficult than working with gold. And, it's harder to spot. Gold sticks out because it is yellow. There are only two metallic elements with color, copper and gold and their alloys. All others are white or some shade of gray. Silver is the whitest metal. German silver, also called nickel silver, is an alloy made to look like silver. It contains no silver.
Gold plating thickness is measured in microinches - millionths of an inch. You can't tell the gold thickness by looking at it. However, very thick gold is usually matte in appearance. Thin gold can reflect the brightness of the shiny nickel underneath it. The thickest normal electronic plating is usually found on parts that require heating further along in their assembly, such as those that use a gold/silicon braze for chip attachment. Examples are side braze, CPU, or hybrid circuit packages (Google). They run about 50-60 microinches of gold. There are parts that require thicker gold (as much as 500 microinches) but, most are military and are rare. The thinnest electronic gold you will probably find is about 12 microinches. Some cheap costume jewelry only has 5 microinches. Much electronics is intermediate between these two extremes, about 20-30 microinches. This intermediate level is usually found on parts that require multiple insertions, such as fingers or connector pins. These figures apply to products that the manufacturer wants to last a long time. On cheapo products, the thickness will vary but, always on the low side.
The above is only a average guide. You'll notice that I always use vague terms like: often, generally, sometimes, usually, etc. The only way to absolutely know values is to assay or refine the parts. I have seen fingers that run from $10/# to $150/#. The gold on packages, however, usually runs constant. Although not always reliable, due to the internal stress of some hard gold deposits, one way to estimate relative thicknesses is to put the part in a 25-50% nitric solution. Hard gold is used on insertion parts, such as fingers. As the copper and/or nickel, under the gold, dissolves, the gold flakes. Generally, the smaller the flakes, the thinner the gold. Also, thinner gold is more porous than thick gold. Since the acid penetrates through the pores, the metals under thin gold generally is attacked faster than that of thick gold. If you do play around with the nitric, it is best to run stuff side by side, to get a comparison. If you refine the items, make notes so everything is more meaningful.
Another way to estimate plating values is to calculate the surface area (number of square inches) of the plated part. This is the way that parts are plated. The plater calculates the area of the load and then sets the amperage based on the type plating bath that is being used. Gold bath current densities usually range from 3 to 10 amps per square foot (asf) of surface area. The thickness depends upon how long the parts were in the plating tank. It usually takes about 15 minutes to plate 100 microinches thick. For 60 microinches, it takes 9 minutes. The measurement of surface area takes a little math. Here are some surface area formulas:
Surface Area and Volume
Take the present gold market price, say $650, and move the decimal point three place to the left. This equals 0.65 or, 65 cents. This is very close to the value of one square inch of 100 microinches of gold. If you had 60 microinches of gold, the value per square inch would be .65 X .60 or, 39 cents per square inch. For 25 microinch gold, .65 X .25 or, about 16 cents. This is especially handy when you have a lot of identical parts, such as those boards with 1000, or so, 1" long pins. Determine the surface area of one pin and multiply by 1000 or, whatever. For more accuracy, use calipers, micrometer, or an optical comparator to make the measurements. Write down the measurements and make the calculations.
This isn't rocket science and nothing is exact. And, there are numerous exceptions. However, I know from experience that the above info can put you in the ballpark. Even with ballpark figures, you will probably know more than the seller. In this business, more knowledge of the material always puts you ahead in the buying/selling game. And, it is a game. No one really knows the true value except for the one who refines the material. That's why the refiner is called the "Last Liar".
