Stepper Motor problems...

[Mr John]

I am using stepper motors as indicator instruments in an anemometer project I have made.

There is a problem in that the stepper motor driving the wind speed dial tends to loose steps, but only in one direction. Under ordinary circumstances the wind may vary from say 2 to 10 knots, when the wind increases I step the motor to move the indicator (needle) and when it decreases I step the motor back the same number of steps. However after a period of time there is an increasing error in which the needles slowly moves down in the range and eventually past the zero point.

It is as if it takes more steps to move up than to move down the same distance on the scale.

I count the up steps and the down steps and all seems in order as regards to the arithmetic of my program.

This is very puzzling and somewhat frustrating, any comments greatly appreciated.

John

 

[Harald Kapp]

Do you properly ramp up and down the stepper drive? Simply applying a number of steps at full speed may lead to the motor loosing steps.

A stepper motor may not be the best solution for your application. Wind speed can rapidly change and having the motor follow these ups and downs will require it to brake and reverse direction continuously. This is, afaik, not what stepper motors are designed for.
Alternatively, use an analog meter (like e.g. this one or this one with a vintage look) and drive it with an analog voltage proportional to wind speed. You may bea bele to drive such a meter directly from a pwm source (µC output) without the need to smooth the output as the mechanical inertia of the meter will do the low pass filtering.

 

[Mr John]

The stepping is very slow so I doubt ramping is an issue.

You are quite right that the stepper is maybe not ideal for the job and I would rather have selsyns to work with but I dont so steppers it is but thanks for the suggestions anyway.

I have been thinking since posting that micro stepping may be the problem so that is something to experiment with.

 

[davenn]

I am using stepper motors as indicator instruments in an anemometer project I have made.

There is a problem in that the stepper motor driving the wind speed dial tends to loose steps, but only in one direction. Under ordinary circumstances the wind may vary from say 2 to 10 knots, when the wind increases I step the motor to move the indicator (needle) and when it decreases I step the motor back the same number of steps. However after a period of time there is an increasing error in which the needles slowly moves down in the range and eventually past the zero point.

hi there

years ago us ham radio guys used to use Selsyn motors for indicators of where our antennas were pointing.
one motor on the mast on the bottom of the pole that rotates connected to the other one in the shack that had
a needle indicator on its shaft to indicate compass bearing

https://www.allaboutcircuits.com/textbook/alternating-current/chpt-13/selsyn-synchro-motors/

 

[Mr John]

I have been out and checked on how my driver is set and I set it is set for 32nd microsteppping

 

[Mr John]

Hi Dave

I have a number of selsyn devices but not enough to make the full 'set' that I want for our airport museum project. I would use selsyns if I had enough of them!

 

[Bluejets]

This is very puzzling and somewhat frustrating, any comments greatly appreciated.

Found one unit that uses an Arduino to drive a stepper for a speedo readout.
If you are familiar with Arduino it shouldn't be too much to modify for your wind speed as it's simply a pulse input.
Pulse would I imagine just come from a hall effect and magnet on the anemometer.

http://www.hillclimb.org/forum/viewtopic.php?f=16&t=1133&sid=260e267564c31b08855a0e21ea57cbac

Files are here ............. https://github.com/Walterclark1/Stepper_Speedometer

 

[kellys_eye]

Can you determine if the step loss is consistent? I doubt this is a motor drive issue.

It could be as simple as forgetting that steps count from 'zero' as opposed to 'one' and adding an extra step in the reverse direction may be all that's needed? I.e this is a programming issue.

Which stepper device are you using?

 

[Mr John]

Hi Bluejets, yes, I am using an Arduino and pulses from the head unit, I am using an opto device instead of a Hall device but I sure will look at those files for inspiration.

K_E, The loss does not appear to be consistent and I count the steps up and the steps down which appear correct.

The stepper driver is a 'generic' "Microstep Driver" DC9-42VDC. The motors are an unknown type and one of a dozen I obtained NOS in Philips packaging. They are possibly quite old and only 48 steps per 360 degrees.

Thanks for all the comments.
john

 

[Minder]

I agree with the others a stepper motor is far from an ideal for a anemometer. especially a low res odd ball that is only 48steps/rev!
I have made a couple over the years, the first for my kids school, a later one using a Picmicro.
But there are IC's designed for it such as the National LM2907 the app notes show a simple linear output to supply a meter of some sort, I used a slot opto and slotted disc to read the pulses.
M.

 

[Mr John]

This is for a museum exhibit and the aim is to replicate the Munro style of instrument..

This is their wind speed indicator and the wind direction is the same style.

Thanks for the suggestions of alternatives but I am somewhat committed to what I am doing especially as I need to make six instruments (i.e. 3 pairs). However of course if I can not get these Philips steppers to operate satisfactorily I will be able to invest in other steppers.

But I would really like to arrive at a conclusion as to why these steppers do not operate as I expected.

Brgds
John

 

[Minder]

Stepper/synchro's are OK for say wind direction, but not speed, you need friction-less devices.
There is usually too much indent torque to overcome in a stepper.
Those appear as moving coil meters so they could be ideal to be calibrated to a linear output of the LM2907 etc.
M.

 

[kellys_eye]

Many car speedometers are stepper diven and there is a range of stepper controller chips (plus the controlling mechanisms to which a pointer can be attached) available on t'internet.

Adafruit does one and has all the adetails you need

https://www.adafruit.com/product/2424

 

[Mr John]

Thanks for the comments.

If had the Munro instruments I would be using them, but I do not.

I made two changes today, I converted to single step operation and I increased the time delay after selecting speed direction. It appears to be operating perfectly but I will give it a couple of days before relaxing.

 

[Mr John]

K_E , that is a very interesting link, thanks.