A DIY Velocity Transducer

Alec_t

Joined Sep 17, 2013
14,280
I'd describe that as an acceleration/displacement transducer? To get velocity would require a fair bit of signal processing, unless the motion sensed is oscillatory.
 
Last edited:

nsaspook

Joined Aug 27, 2009
13,079
I'd describe that as an acceleration/displacement transducer? To get velocity would require a fair bit of signal processing.

The article seems to be assuming all scalar values and results because of the single transducer instead of something like a strapdown inertial navigation system need to track the 3D time position of an object.

https://byjus.com/physics/velocity/

Let’s summarize what we have learned:
  • (a) This is a function of displacement.
  • (b) We require both magnitude and direction to define velocity.
  • (c) Instantaneous velocity is the velocity at any given moment of time, whereas average velocity is the total displacement divided by total time.
  • (d) A changing velocity indicates acceleration.
https://rotations.berkeley.edu/strapdown-inertial-navigation/

The modern 9DOF MEMS sensor has revolutionized what's possible at low-cost.
https://forum.allaboutcircuits.com/...avoidance-cars-turn-angle.185116/post-1710816
https://forum.allaboutcircuits.com/...re-less-then-1mm-movmenet.186616/post-1729529

https://aerospace.honeywell.com/us/.../four-facts-about-inertial-navigation-systems
 
Last edited:

Thread Starter

Jon Chandler

Joined Jun 12, 2008
1,029
The intended purpose of this DIY velocity sensor is to measure machinery vibration, not linear velocity (i.e., it's not a speedometer).

Similar velocity sensors have been used for years for machinery condition analysis, although the usual approach today is to measure acceleration and integrate the signal to velocity. From Istec.com:

SmartSelect_20220622-050515_Edge.jpg
 
Last edited by a moderator:

strantor

Joined Oct 3, 2010
6,782
@jayTechStuff did you have anything to do with this article? It's uncanny that it was posted to EDN 2 weeks after this thread, and the concept is similar to the voice coil concept that I described here and in post # 8 of the first link. I suspect this is a vast improvement over desktop computer speaker coupled to vibrations through a pipe. I might give this a go and compare results to the accelerometers I bought.
 

nsaspook

Joined Aug 27, 2009
13,079
https://power-mi.com/content/seismic-velocity-transducers
This type of device was developed for its first industrial use in the late 1940s by Arthur Crawford, and its application was extended during the 60s and 70s.
https://instrumentationtools.com/moving-magnet-type-velocity-transducer/

Principle of Moving Magnet type Velocity Transducer
When a permanent magnet moves inside a coil, the change in the length of the air gap varies the reluctance. Hence the output voltage is directly proportional to the rate of change of the length of the air gap (change in length produced by velocity). Thus the output voltage becomes a measure of the velocity when calibrated.
 

ApacheKid

Joined Jan 12, 2015
1,533
https://power-mi.com/content/seismic-velocity-transducers


https://instrumentationtools.com/moving-magnet-type-velocity-transducer/

Principle of Moving Magnet type Velocity Transducer
When a permanent magnet moves inside a coil, the change in the length of the air gap varies the reluctance. Hence the output voltage is directly proportional to the rate of change of the length of the air gap (change in length produced by velocity). Thus the output voltage becomes a measure of the velocity when calibrated.
But strictly speaking this measures the relative velocity of the coil to the magnet not the transducer relative to some other reference frame.
 

nsaspook

Joined Aug 27, 2009
13,079
But strictly speaking this measures the relative velocity of the coil to the magnet not the transducer relative to some other reference frame.
Sure, it's accurate, when calibrated (compared to a standard) and used properly.

https://transtekinc.com/product-categories/linear-velocity/
Unlike LVDTs, these transducers do not produce a voltage output unless the magnet is moving. This makes it impossible to electrically identify the correct location of the magnet for linear operation. Users must mechanically position the magnet in the LVT at a known reference point.
https://transtekinc.com/teknotes-linear-velocity-transducer-calibration-report/
The Trans-Tek Series 100 Linear Velocity Transducers produce a DC output voltage proportional to the velocity of the magnet relative to the coil. These units are designed and tested to assure that they conform to prescribed specifications. Each unit is calibrated by our QC Dept., and a strip chart recording of the results are packaged with each unit.
 

ApacheKid

Joined Jan 12, 2015
1,533

ApacheKid

Joined Jan 12, 2015
1,533
I can envisage a velocity measuring device that is based on acceleration (force) and time. But it would need to be at "rest" initially. Then every time we accelerate we measure that and the time at that acceleration and calculate the instantaneous velocity. Of course rotation would require additional work...

But if I accelerate from "rest" at 5 MPH/sec for ten seconds then stop accelerating, I can be sure I'm travelling at 50 MPH.
 

nsaspook

Joined Aug 27, 2009
13,079

Thread Starter

Jon Chandler

Joined Jun 12, 2008
1,029
So to be clear, if I had one of these transducers in a system sitting on the passenger seat and switched off. If I drive along at 20 MPH and switch the device on, would it tell me I was driving at 20 MPH?
No. The signal is generated by the motion between the coil and magnet. No motion, no output. Its intended use to to measure vibration.

Piezoelectric accelerometers behave the same way. A signal is generated when changing force on the seismic mass deforms the piezo crystal. A piezoelectric accelerometer does not respond to static acceleration (i.e., flipping it over with respect to the earth's gravitational field does not result in a – 1 g signal).
 
Top