Measuring glass thickness

Professional Engineer & PE Exam Forum

Help Support Professional Engineer & PE Exam Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

ClohKr

New member
Joined
May 21, 2011
Messages
1
Reaction score
0
I work for a Stained Glass manufacturer. I have been assigned a project seeking to be able to measure the thickness of the glass, flowing down a 300ft lehr, closer to our Hotend. Our glass flows from a furnace, down a forehearth, through a set of forming rolls, and down the lehr. Our glass thickness ranges from 0.500" down to 0.075". The hotend lehr zones are typically 990 deg F. Right now the way we measure the thickness of the glass is we wait for the glass to come out of the lehr, at the coldend, and cut a piece out and measure using a micrometer (the glass being around 30" wide and 300ft long, continous flow). If the glass is not in the target range then the coldend operators call it up to the hotend and they adjust. It takes around 15 min for the glass to get from the hotend to the coldend any adjustments made, that may or may not have been correct, will not be seen for another 15 min. Being able to measure the thickness of the glass closer to the hotend will greatly improve productivity and save the company time and money. Contact sensors don't seem to be the logical solution since roll wraps in our forming rolls or last glass from an end of a run will typically lose the flatness and could damage a contact sensor. I have been looking at displacement lasers to see if this is the avenue to go down. My concern is the temperture and reflectivity of the environment. I am asking if anyone out there has had any information I could use to find the best solution. Maybe some companies I could call to ask what they are using or if you know of a company marketing a device that I might wanna try. Thanks.

 
Well, if you can't find anything commercially you may have to roll your own system. The trick is finding a variable that changes in relation to the glass thickness and does so in a repeatable manner. Hall Effect sensors maybe? Once you've found a measurable quantity then you should be able to design a control loop that would automate the entire process using a PLC. You'll have to tune a PID, but there are lots of tricks out there for that. In fact, if you're clever enough, you may be able to come up with some auto-tuning algorithm. Just my two cents as I've seen this done successfully with gas analyzers.

 
Many people think that the actual thickness of the glass is a direct indicator of its safety. Avoid using rims that do not cover the entire circumference of the lens. Leaving the lens exposed will clearly show the actual thickness of the lens.

 
You may use sound detection - when you direct on the glass a pulse of a sound, you will receive 2 (and even more) responces - first (big) from the surface and all other shifted in time T=S/V, where S - thickness, and V - speed of sound inside of glass. And you can take such an info from the other side of glass - then the first response will be lower.

Even an usual oscilloscope will give you not bad information.

 
"...The trick is finding a variable that changes in relation to the glass thickness and does so in a repeatable manner..."
The forming rollers (with a set radius, and adjustable C-C) would provide a place to measure changes in C-C. Just find a position indicator, and attach it to the roller centers, calibrate, done.

 
Back
Top