Ive got a question about assigning a PDF to analog meter resolution. While perusing OIML R49-2 Test Methods for Cold Water Meters, the uncertainty analysis contained in Annex G indicates that the analog meter resolution is treated as a triangular distribution and they site the GUM section 4.4.6. I was under the impression digital resolution was handled as a uniform distribution and analog displays as Gaussian. The GUM 4.4.6 just describes the triangular distribution as being appropriate when you have less limited data as compared to the uniform distribution. Any thoughts on this??

Any calibration gurus here that can help with this?

I believe you are correct, Mike.

The explanation that I would have is that the manufacturer went on the side of caution since the two distributions approximate so closely.

The GUM 4.4.6 just describes the triangular distribution as being appropriate when you have less limited data as compared to the uniform distribution.

It has been a while since i last took a look into the GUM. I dont have a copy anymore. In my opinion taking the uniform/ rectangular distribution is a "safer" distribution compared to the triangular when you have little data, so i don't quite get it either. See if i can get my hands on a copy of the GUM.....