Sorry. Another long windy story.
Although this happened 50 years ago, this year, it posed a question or two that haven't been answered to my satisfaction.
One of my first jobs in LA was to head up the lab in a large plating facility. This place had about 100 active plating tanks, plus a ton of cleaners, acid dips, etc., ranging from 50 gallons to maybe 4000 gallons, of about every type of plating under the sun. My job was to analyze the solutions and keep them up to snuff. I also was the head troubleshooter and, in a plating shop, troubleshooting is a daily, if not hourly, thing.
Something that set this place apart was that they were a licensee for the Kanigen electroless nickel system developed by the General American Transportation Co. This was originally developed to plate the inside of railroad tank cars The round tank from a car was rotated on rollers while the 200F solution was pumped in constantly and drained out constantly and reheated. The solution plated catalytically and deposited about 1 mil, 25 microns, in an hour. We maintained 4200 gallons of the solution. The solution crapped out after about a month and had to be re-made from scratch. It was made up with 6 or 8 chemicals.
At night, it was pumped into 3 holding tanks and allowed to cool. During the day, it was pumped from the storage tanks to whatever tanks were needed to fit the particular shapes and sizes of the parts being plated that day. There were 20 or 30, 300 series SS tanks, of all shapes and sizes, all connected with about 2 or 3" pyrex piping. The solutions were heated with live steam from a big boiler. At night, when all the tanks were empty, a 20% nitric solution was pumped into all the SS tanks used that day. This dissolved any nickel plated on the SS and, at the same time, passivated the SS so that nickel wouldn't plate on the SS unless a part were dropped in the tank and that damaged the oxide layer on the SS, thus exposing nickel to the solution.
Patience, please. I'm getting there.
One big job we had was plating large SS screens, with holes in them, which were used to line centrifuges in large sugar plants. I think that the sugar was dissolved and then the solution was put into the centrifuges to remove undissolved crystals, etc. The flat SS sheet panels were about 18" X 24" and a number of them were used in a circle to line the centrifuge. They had many 100s, and probably, many 1000s, of these fabricated before discovering that the holes were too big. What we were doing was to plate nickel thick enough to fill in the holes until it met their spec. Since electroless nickel doesn't use a power supply, it plates extremely evenly over the entire part.
The holes were chemically machined and were hexagonal, about 1/32" (.8mm), or so, and were very close together. There was an immediate discussion on how QC should best measure the resulting hole size(s) after plating. Several methods were tried but found innacurate, tenuous, or time-consuming. I discovered that if a plated sheet were placed on top of an unplated sheet and the pair placed on a light box and the top plated sheet rotated, a moire pattern appeared. When the top sheet was rotated to a certain angle, the images of the holes were magnified from 1/32" to maybe 3 or 4", or more. I assumed that the magnified image was an average of all the holes in the area covered by the 4" hole image. The 4" hole image was measured along with the spacing between 4" holes and the hole diameter was calculated. Whether my thinking was correct, or not, in my assumptions, the hole size came out correct and the customer was pleased.
Questions: Was the large image truly an average of all the holes in that large area? If so, what is the math involved?
Every few years, I rehash this problem in my mind and lose sleep thinking about it. I just Googled again and could find no instant answer. I downloaded a file on Moire Patterns from Wolfram along with a viewer. Maybe it will help solve this problem.
NOTE: The Wolfram stuff was worthless - just drawings.
Although this happened 50 years ago, this year, it posed a question or two that haven't been answered to my satisfaction.
One of my first jobs in LA was to head up the lab in a large plating facility. This place had about 100 active plating tanks, plus a ton of cleaners, acid dips, etc., ranging from 50 gallons to maybe 4000 gallons, of about every type of plating under the sun. My job was to analyze the solutions and keep them up to snuff. I also was the head troubleshooter and, in a plating shop, troubleshooting is a daily, if not hourly, thing.
Something that set this place apart was that they were a licensee for the Kanigen electroless nickel system developed by the General American Transportation Co. This was originally developed to plate the inside of railroad tank cars The round tank from a car was rotated on rollers while the 200F solution was pumped in constantly and drained out constantly and reheated. The solution plated catalytically and deposited about 1 mil, 25 microns, in an hour. We maintained 4200 gallons of the solution. The solution crapped out after about a month and had to be re-made from scratch. It was made up with 6 or 8 chemicals.
At night, it was pumped into 3 holding tanks and allowed to cool. During the day, it was pumped from the storage tanks to whatever tanks were needed to fit the particular shapes and sizes of the parts being plated that day. There were 20 or 30, 300 series SS tanks, of all shapes and sizes, all connected with about 2 or 3" pyrex piping. The solutions were heated with live steam from a big boiler. At night, when all the tanks were empty, a 20% nitric solution was pumped into all the SS tanks used that day. This dissolved any nickel plated on the SS and, at the same time, passivated the SS so that nickel wouldn't plate on the SS unless a part were dropped in the tank and that damaged the oxide layer on the SS, thus exposing nickel to the solution.
Patience, please. I'm getting there.
One big job we had was plating large SS screens, with holes in them, which were used to line centrifuges in large sugar plants. I think that the sugar was dissolved and then the solution was put into the centrifuges to remove undissolved crystals, etc. The flat SS sheet panels were about 18" X 24" and a number of them were used in a circle to line the centrifuge. They had many 100s, and probably, many 1000s, of these fabricated before discovering that the holes were too big. What we were doing was to plate nickel thick enough to fill in the holes until it met their spec. Since electroless nickel doesn't use a power supply, it plates extremely evenly over the entire part.
The holes were chemically machined and were hexagonal, about 1/32" (.8mm), or so, and were very close together. There was an immediate discussion on how QC should best measure the resulting hole size(s) after plating. Several methods were tried but found innacurate, tenuous, or time-consuming. I discovered that if a plated sheet were placed on top of an unplated sheet and the pair placed on a light box and the top plated sheet rotated, a moire pattern appeared. When the top sheet was rotated to a certain angle, the images of the holes were magnified from 1/32" to maybe 3 or 4", or more. I assumed that the magnified image was an average of all the holes in the area covered by the 4" hole image. The 4" hole image was measured along with the spacing between 4" holes and the hole diameter was calculated. Whether my thinking was correct, or not, in my assumptions, the hole size came out correct and the customer was pleased.
Questions: Was the large image truly an average of all the holes in that large area? If so, what is the math involved?
Every few years, I rehash this problem in my mind and lose sleep thinking about it. I just Googled again and could find no instant answer. I downloaded a file on Moire Patterns from Wolfram along with a viewer. Maybe it will help solve this problem.
NOTE: The Wolfram stuff was worthless - just drawings.