EE's in the house?

[kevinthenerd]

I'm trying to switch a stepper motor. I think I messed up on ebay and bought "sand resistors" that are really wire-wound. (10 watt, 12 ohm, "J" which I think means 5%) Would using these in series with the stepper motor still improve my time constant? The stepper motor is a very inductive load, so if the resistors are cross-wound, I guess they'd be less than the stepper motors, right?

 

[kevinthenerd]

I guess one way to answer the question would be if you could provide typical inductance values for a 1.2 ohm 2.5A unipolar stepper motor and the aforementioned resistor.

 

[drinkmorejava]

I don't understand what you mean by improving your time constant, if you want it to be more accurate, you'd want more accurate resistors. Depending on the design of your circuit though, unless you're running huge amount of power though them, inductance interfering with the timing circuit should be minimal.

 

[PottedMeat]

Time constant? You're trying to improve the response time of the coils i guess?

If you want to reduce the inductance you should have bought some metal film / carbon film / 'low inductance' resistors. However the stepper coil inductance should be >>> wirewound sand resistor inductance.

I think raising the supply voltage on the motor will improve response. The steppers in Epson printers are driven by 40V supplies while being switched with +3.3 or +5V logic - maybe that helps.

 

[MoMeanMugs]

Yes, they should decrease the time constant, since it is calculated as L/R. However, the resistors aren't perfect and are going to carry some capacitance, so it's really an LCR circuit. Do you have the power rating for the motor? If it's given in volt-amps instead of watts, I can tell you what the inductance value is. It's a simple complex power problem. If the power rating isn't in volt-amps, you'd really need an oscilliscope to test it.

 

[kevinthenerd]

Originally posted by: MoMeanMugs
Yes, they should decrease the time constant, since it is calculated as L/R. However, the resistors aren't perfect and are going to carry some capacitance. Do you have the power rating for the motor? If it's given in volt-amps instead of watts, I can tell you what the inductance value is. It's a simple complex power problem. If the power rating isn't in volt-amps, you'd really need an oscilliscope to test it.

Good call. I can check for that, and I think I might remember where to find that in an old textbook. (I'll let you know if I can't.)

 

[kevinthenerd]

Originally posted by: PottedMeat
Time constant? You're trying to improve the response time of the coils i guess?

If you want to reduce the inductance you should have bought some metal film / carbon film / 'low inductance' resistors. However the stepper coil inductance should be >>> wirewound sand resistor inductance.

I think raising the supply voltage on the motor will improve response. The steppers in Epson printers are driven by 40V supplies while being switched with +3.3 or +5V logic - maybe that helps.

The X-Y stepper motors are rated for something like 2.5 amps, but since I have no clue how to replace them (they look expensive), I want to run them somewhere around 1 amp. The Z-axis stepper motor came out of a dot-matrix printer, and I'm not as worried about the time constant since it won't be moving as much. (Its DC resistance is around 40 ohms.) The DC resistance for each coil of the XY is 1.2 ohms, so with a 12 ohm resistor in series, I should get an L/R improvement of 11:1 (assuming non-inductive, non-capacitive resistors). The supply voltage had to be raised accordingly. I'm going to use the 12V supply out of an old computer power supply if I can get my hands on one. (I gave all mine away not too long ago.)

I'll be running well under an amp when you consider the collector-emitter voltage of the Darlington I'm using to control them... one ULN2064B for each axis. Luckily they already have protection diodes.