electromagnet gauge makes buzzing noises?

hi, i've been working on a project that involves reusing the electromagnetic needles commonly used in 90's era ford gauges. i can get it to work fine, but it seems to cause a buzzing noise as it is doing the work.

there are three pins, ground, power, and signal. right now i just have it working off of 5v.

when i have it going, and you hold the needle with your finger, you can feel it vibrating. i'm using an arduino nano, a npn transistor, and the gauge with a PWM output. do i just need to put a capacitor in? or how do i keep it from buzzing like it does? its not a loud buzz, but you'd notice it if you were in a quiet room.

sort by: active | newest | oldest
If you run it on DC, derived from a pot, say, does it vibrate then ?
zack247 (author)  steveastrouk4 years ago
no, it doesn't.
zack247 (author)  zack2474 years ago
but it also doesnt seem to work the same when i do, when i have it set at around 3v it is below the actual printing of the gauge. if i pull the signal pin low it goes to the high end of the gauge, so ive been using PWM to control the contact to ground to control it.
iceng zack2474 years ago
Drawn to change PWM to a smooth voltage for the meter as described...
pwmZACK.gif
In that case try Iceng's circuit
zack247 (author)  steveastrouk4 years ago
i tried the circuit and it only caused the needle to stay at the upper limit, the needle doesnt move according to the signal.
iceng4 years ago
Here is a starting point.
FilterPWM.gif
zack247 (author)  iceng4 years ago
i tried this circuit sans diode, and it did not work. my gauge has 3 pins though; power, ground and signal. which would i put the diode across?
iceng zack2474 years ago
One has to assume the gauge design being made for a car
has that inductive kick protection built-in....
I'm trying to find out what kind of instrument this really is.
zack247 (author)  steveastrouk4 years ago
this is the gauge i'm talking about.
hopefully this should help. i have the pins marked on the back.
IMG_20130908_235113.jpgIMG_20130908_235133.jpg
Aren't these just ordinary moving coil meters ?
zack247 (author)  steveastrouk4 years ago
im not sure, you tell me. the only way i can get the needle to move properly is through PWM.
What gets me is this "3rd pin" - it looks for all the world like a simple traditional meter. I don't know what this extra pin is for !
That "3rd middle pin" in Zack's excellent and well focused picture
could be an electrical zero centering adjust since there is no mechanical
adjusting screw visible.

And why or how is this gas gauge showing almost 2/3 full tank ??!!



zack247 (author)  iceng4 years ago
there is the power and ground pins, marked "G" and "+", and then i control the needle with the signal pin, marked "s". the middle pin is ground, the rightmost one is +ve, and the lefmost is signal.

and the fuel gauge has more friction to it so it stays where it last was when you turn the car off. the car also tells every other gauge to go to zero when it is shut down.

the zero is permanently set on the gauge, when you apply power to the +ve and ground pins, it goes to zero until you start controlling the signal pin.
You COULD increase the arduino's PWM frequency to stop being able to feel it vibrate....

http://playground.arduino.cc/Code/PwmFrequency
zack247 (author)  steveastrouk4 years ago
the problem isnt that it can be felt, its that it can be heard.

the page you linked to said that timers wont work properly after changing the frequency, so what do i do about it then?
What use are you making of the timers in the program ? If its waits, rescale them in proportion to the increase in PWM frequency. Make the PWM frequency above 20kHz, and no-one will hear it.

Did you try a capacitor across the load ?
zack247 (author)  steveastrouk4 years ago
capacitor didnt do anything, and attaching anything besides the signal to the signal pin causes it to display improperly.
Hmmm. could it be ratiometric then ??
Good thought, I agree, but doubt there are more then 5 people here who
would know what we are talking about
iceng zack2474 years ago
There are some processors that cam deliver a true analog voltage on an output pin..

PWM is the only way a digital device using a resistor, and capacitor
can deliver a smooth voltage.

Unless you want to add a serial DAC IC to the mix...
Put a diode backwards across the coil, before you blow up your arduino by accident.

Put a nice fat capacitor across the coil too.
zack247 (author)  steveastrouk4 years ago
my arduino has been running it for a while, and it hasnt even gotten warm. this is a 3 pin setup for the coil, designed to be controlled by a microcontroller elsewhere. the position of the needle depends on how much voltage is present on the signal pin compared to the +ve pin.
What's the part number of the thing ?
zack247 (author)  steveastrouk4 years ago
f4zf-10e853-ea, but google doesnt bring up anything relevant. it's the fuel/temperature gauge out of a 1995 mustang. ive been using the temperature gauge for all of this.
Kiteman4 years ago
If the magnet being powered from the mains, then what you are hearing is probably just the natural AC hum as the magnet switches on and off 50 or 60 times a second.

Without knowing exactly you have done, the best I can suggest is to suggest you look at smoothing and/or rectifying an AC voltage.
zack247 (author)  Kiteman4 years ago
actually, i'm using DC voltage for this one.
zack247 (author)  zack2474 years ago
it is kind of like a hum though, i'm assuming the noise is coming from the needle as it moves back and forth ever so much as the coil is turned on and off.
You will need to create a low pass filter. This article should help.

http://provideyourown.com/2011/analogwrite-convert-pwm-to-voltage/