graemeftv Posted May 9, 2011 Share Posted May 9, 2011 just been trying to service some moving lights for a customer, and the stepper motors all seem to make a high pitch noise when they are in a static position. apart from mounting them on rubber washers to aviod the plate of metal they are attached to vibrating anymore. does anyone have any suggestions or experience? many thanks Graeme Link to comment Share on other sites More sharing options...
maxjones2000 Posted May 10, 2011 Share Posted May 10, 2011 I personally dont know about this kind of thing, but for other BR members that do, there is info needed that will help.. What make and models are the moving lights? Link to comment Share on other sites More sharing options...
TallMike Posted May 10, 2011 Share Posted May 10, 2011 I remember every Martin Roboscan 812 I've ever used doing this. I think it's to do with when the motor is stationary 'between' steps, but this is speculation. Link to comment Share on other sites More sharing options...
fliggygeek Posted May 10, 2011 Share Posted May 10, 2011 I remember every Martin Roboscan 812 I've ever used doing this. I think it's to do with when the motor is stationary 'between' steps, but this is speculation. Whine on stepper motors is generally due to the stepper trying to hold position halfway through a step, so it bounces between the closest steps. You might be able to see the mirror shivering or something as a symptom of this. Im not sure what the RoboScans use for sensing rotation, but if its an optical encoder, moving the encoder a tiny bit on the shaft or the sensors a bit might fix this. This is speculation as I havent tried it, but in theory, the issue is just that the conversion between DMX 8 or 16 bit values is giving values halfway between steps on the stepper, and moving the encoder might line up more of the values than were previously aligned and in doing so lessening the amount of steps that are being held halfway through steps. Link to comment Share on other sites More sharing options...
Brian Posted May 10, 2011 Share Posted May 10, 2011 A sort of HF hissy fizzy sound sometimes with a detectable note amongst it? Stepper motor controllers often use PWM to control the current flow. It's this that you hear; or to be accurate it's sometimes a subharmonic that you hear. As the waveform fed to the motor is usually a rectangular wave it contains loads of harmonics. The noise comes either from small movements in the motor or from sources like magnetostriction of the windings/rotor/stator. It is possible to design quiet drives which once the motor is no longer moving switch over to using DC to hold position. The trouble is it adds cost to the driver circuitry. Another option, although one for the design stage, is to use low-pass filters on the motor feeds to get rid of the HF content. Google a chip like the L297 for a bit more info. If it's more of a fixed frequency tone... If the lantern in question uses feedback sensing, even something as simple as an end-stop switch, you might find that slightly adjusting that might quieten things down. The software has determined that the position isn't quite right and is trying to correct but is ending up between two adjacent steps. Good software control should eliminate this by applying a bit of filtering. Lots don't. Link to comment Share on other sites More sharing options...
fliggygeek Posted May 10, 2011 Share Posted May 10, 2011 A sort of HF hissy fizzy sound sometimes with a detectable note amongst it? Stepper motor controllers often use PWM to control the current flow. It's this that you hear; or to be accurate it's sometimes a subharmonic that you hear. As the waveform fed to the motor is usually a rectangular wave it contains loads of harmonics. The noise comes either from small movements in the motor or from sources like magnetostriction of the windings/rotor/stator. It is possible to design quiet drives which once the motor is no longer moving switch over to using DC to hold position. The trouble is it adds cost to the driver circuitry. Another option, although one for the design stage, is to use low-pass filters on the motor feeds to get rid of the HF content. Google a chip like the L297 for a bit more info. If it's more of a fixed frequency tone... If the lantern in question uses feedback sensing, even something as simple as an end-stop switch, you might find that slightly adjusting that might quieten things down. The software has determined that the position isn't quite right and is trying to correct but is ending up between two adjacent steps. Good software control should eliminate this by applying a bit of filtering. Lots don't. I think you mean Servo motors use PWM, the idea behind steppers is they use lots of poles and feed dc to a set of poles which drags the shaft around, theres no voltage or current changing besides current usage generated by load in the magnetic field. Good idea to check whether you are using steppers or servos though, its a valid point. Link to comment Share on other sites More sharing options...
timsabre Posted May 10, 2011 Share Posted May 10, 2011 I think you mean Servo motors use PWM, the idea behind steppers is they use lots of poles and feed dc to a set of poles which drags the shaft around, theres no voltage or current changing besides current usage generated by load in the magnetic field. Good idea to check whether you are using steppers or servos though, its a valid point. No, Brian is correct, steppers use PWM as well, to do microstepping - this allows you to obtain positions in between the fixed poles, which are normally 1.8 degrees apart. This would be very jerky in a moving light fixture. Older fixtures often use a low PWM frequency which is audible as a whine, and there's not much you can do about it. Newer designs have got faster PWM above audible frequencies, so it's still there, you just can't hear it. Link to comment Share on other sites More sharing options...
Brian Posted May 10, 2011 Share Posted May 10, 2011 ...there's no voltage or current changing besides current usage generated by load in the magnetic field.A stepper controller chip set, like the L297 and L298, use PWM to control the load current. The load current is fed through a pair of sense resistors which feedback to the controller to shut off the PWM drive once peak current has been reached. This gives you a PWM signal, which may well be ultrasonic, which is being modulated by the current signal. A nice recipe for audible effects. As external forces act to move the stepper away from where it wants to be (ie magnetically centred) the current will go up, increasing drive and noise. L297 and L298 data. Also it's quite a good explanation on how steppers work. @OP - is it obvious which driver chips they are using? A google on them might through up a clue as to what technology they use. Link to comment Share on other sites More sharing options...
fliggygeek Posted May 10, 2011 Share Posted May 10, 2011 ...there's no voltage or current changing besides current usage generated by load in the magnetic field.A stepper controller chip set, like the L297 and L298, use PWM to control the load current. The load current is fed through a pair of sense resistors which feedback to the controller to shut off the PWM drive once peak current has been reached. This gives you a PWM signal, which may well be ultrasonic, which is being modulated by the current signal. A nice recipe for audible effects. As external forces act to move the stepper away from where it wants to be (ie magnetically centred) the current will go up, increasing drive and noise. L297 and L298 data. Also it's quite a good explanation on how steppers work. @OP - is it obvious which driver chips they are using? A google on them might through up a clue as to what technology they use.https://www.lightparts.com/martin/Roboscan-812_list.php has the first two driver chips listed, My guess was that one is a dual stepper driver and one was some form of lamp control, the 3773 Driver chip appears to be an NJM3773 Dual stepper driver, and the 2803 seems to be a ULN2083 8 way Darlington chip for some sort of control. Datasheet for NJM3773 Suggests use with microstepping so that might be the noise. L297 seems to be the very similar, I CBF going through pin for pin and comparing but it *MAY* be a pin for pin replacement. both drivers run on 7vdc and have at least 750mA max current output. both are dual H-Bridge units too. Link to comment Share on other sites More sharing options...
Brian Posted May 10, 2011 Share Posted May 10, 2011 Datasheet for NJM3773 suggests use with microstepping so that might be the noise. It also says... Constant current control is achieved by switching the output current to the windings. ...The current decreases until the clock oscillator triggers the flip-flops of both channels simultaneously,which turns on the output transistors again, and the cycle is repeated. So like the L297/298 combination there is an HF motor drive component modulated by the current feedback. Link to comment Share on other sites More sharing options...
fliggygeek Posted May 10, 2011 Share Posted May 10, 2011 So, the answer is, from what I can see, No, you cant get rid of that noise unless you want to seriously modify the drive circuitry. You can decrease it by fine tuning your cues that need silence so they are magnetically centred. My other thought on the technical side of this, is that given most PWM is generated at microcontrollers, you may have to just recode the controller not the driver chip. Redoing the ROM in these to utilize 25khz PWM would help, but thats way too many man hours. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.