# calculating speed of elevator...

Status
Not open for further replies.

#### kamani

##### New Member
Hello frds and experts!

I am a newbie to MIcros and a new comer in this forum too. and I am quite impressed with the number of responses even to some what *silly* (sorry to say that... but my intention is to sincerely appreciate the guys who are responding) postings.

And I am here to post one more stupid/silly thing...

My task is to measure the speed of an elevator using an optical sensor that gives out signals according to the speed of the elevator by reflecting light from an optical band placed on the guide rails of the elevator cab. The optical band is of alternative dark and white strips of say, 4mm width.

I was suggested to use a PIC Micro to do this task...

The rough idea in my mind at present is...

1) calculate the frequency of the signal emitted by Sensor(by counting the number of pulses in a time span),

2)calculate the velocity from the counted frequency (software module)

And I dont wanna go into the details of this project for now.

I request you to correct me if I am wrong and I await your valuable suggestions and your version of game plan for this task.

[email protected]@ni

#### stevez

##### Active Member
A pulse represents a specific distance and distance/time get you velocity or speed. It would seem that the simplest of circuits/programs would yeild coherent information once the elevator reached a steady state velocity.

In some arrangements the processor keeps track of pulses/time then performs a calculation that is eventually displayed. It's kind of a discreet process that is repeated over and over. If that takes place too quickly the display will be hard to read because the digits will be constantly changing as the elevator speeds up/slows down. Slow that process down and the display will be more readable but it may not reveal changes in velocity that matter. Creating some software to process the information might be another evolution to improve things.

It sounds like one of those things that could be easy or hard depending on how much you want to put in or get out of it.

#### kamani

##### New Member
Hi Stevez!

Thanks for being the earliest.

I dont need to display the results rather I should compare the velocity calculated with a threshold velocity preloaded into micro. Infact, my task is a sub-module of the project- Electronic Speed Governer(ESG).

The main aim of ESG is to keep the velocity of the elevator under surveilance and activate the emergency brake if it moves with an overspeed (in case of an accident or so).

So, for every pre-fixed span of times I need to i) count the Frequency, ii) calculate the Velocity, iii) should compare the velocity with the threshold value, and finally should give an appropriate signal to the braking unit.

Hope I have given a bit more clear information about my task.

Thank you once again.

Ciao - [email protected]@ni

stevez said:
A pulse represents a specific distance and distance/time get you velocity or speed. It would seem that the simplest of circuits/programs would yeild coherent information once the elevator reached a steady state velocity.

In some arrangements the processor keeps track of pulses/time then performs a calculation that is eventually displayed. It's kind of a discreet process that is repeated over and over. If that takes place too quickly the display will be hard to read because the digits will be constantly changing as the elevator speeds up/slows down. Slow that process down and the display will be more readable but it may not reveal changes in velocity that matter. Creating some software to process the information might be another evolution to improve things.

It sounds like one of those things that could be easy or hard depending on how much you want to put in or get out of it.

#### stevez

##### Active Member
Along with absolute velocity I'd look at the rate of change in velocity as well as compare the actual velocity with the velocity you expect.

Rate of change - if the elevator is at rest and the cable breaks you can either wait until the car is going faster than it should to apply the brakes or you can, by the rate of change in velocity, predict that it will soon be going faster and apply the brakes sooner - for a safer stop.

Comparison - if your controls tell the elevator to be going up and you know it should go up at 0.5 meters/second then you ought to measure that value within a reasonable window. If you don't for whatever reason you could go into emergency mode - apply brakes, shut down. All you'll know is that something isn't right. Depending on the failure mode, you might actually "see" this sooner than the rate of change.

I'd do both if I could. Consideration for false triggering should be given. All of this would seem quite manageable with a capable controller.

Status
Not open for further replies.