The math/ logic of the Tachometer

[CDL4582]

I am find the rotations per second (rps) of a tachometer, but I am having trouble figuring out the logic behind it/ how to calculate it. I am using a pulley with 10 spokes, and all that I can think of right now is have the tachometer count every 0.25 seconds then divide it by 10 to make it a full rotation. I tried simply dividing the count found by the tachometer by the total time to find rps, but it doesn't seem correct because with that logic the program won't be giving me the current rps. I don't really need a full vi, but any help with the logic behind would be appreciated. I am using a daq assistant that is detecing the voltage of the tachometer and using a threshold to detect the spokes to count the rotations.

Thanks.

[VItan]

Sounds like you have an incremental optical rotational encoder and are trying to make the system of the encoder + your software act as a tachometer.  A tachometer is a device that measures rotational speed, so the encoder on its own isn't actually a tachometer, although they are sometimes marketed that way if they are often used along with measurement hardware/software that makes the system behave as a tachometer.

Anyways, an encoder can only tell you the rotational position, not the rotational speed.  An incremental encoder can only be watched to see when it changes position.  Therefore, you won't ever be able to get a true current speed from it, but you can numerically differentiate the position with respect to time in order to get an estimation of the speed over the sampling period.  Most commonly one does what you have described - count each spoke over a sampling period, such as 0.25 s, and then divide the number of counts by the sampling period to get speed.  Shorter sampling times mean that you can sample much faster and have an estimated speed much closer to current speed; however, longer sampling allows better resolution.

[billko]

That was a great explanation!