Just got new MINI steppers. What is the smallest way to do PWM?

[ROBOT420]

I just got in a shipment of MINI steppers, and I can not wait to try them out, the possibilities are endless! :-) Here is a pic of them and a mini solar cell I have a few of. What should I use to step them that will also be very small? Thanks!

http://s1362.photobucket.com/user/420photo420/media/20130610_152133_zps7b6ec755.jpg.html?sort=3&o=4

P.S. If anyone wants one (or several) of these just PM me.

[waltr]

Wow, they are small.
I see four terminals so these steppers are bi-polar which needs a dual H-bridge.
There are a number of stepper drivers available in SM chip.
Here is one selection:
http://www.allegromicro.com/en/Products/Motor-Driver-And-Interface-ICs/Bipolar-Stepper-Motor-Drivers.aspx
Other companies make similar chips.

How much current do they draw? At what Voltage?
I'll guess not much so if lower than typical micro-controller outputs (~25mA) then four pins can drive these as an H-bridge would (examples: ATiny or PIC10F or PIC12f processors in SSOP, MSSOP, DFN or QFN packages). If they need a little more current then there are some driver chip (for MOSFETs or data busses) that would work.

[ROBOT420]

Cool, thanks. I have zero info on them other than the resistance per set of poles is 38 Ohms. Other than that I will just have to get some working and push a few past their limits to find out.  Thanks for the links!

[waltr]

Ok 38 Ohm:
at 5V is 131mA which is too much for any u-controller output pin
at 3.3V is 86mA still too much.

so you probably will need a higher current driver.

Looking forward to your first powered tests and measurements.

[waltr]

Thanks.
I got motivated last night and dug out the PIC/Dual H-bridge board and a small (large compared to the ones you have) stepper motor. Soldered wires to the stepper, powered the board the its still working. Yea.

So I'm ready to test out those tiny steppers.

[Sisco]

Mini? Damn.. Those things are Micro Steppers

[ROBOT420]

waltr; Did they get there yet?

Sisco; Want some?

[waltr]

Not yet. Maybe tomorrow.

[jwatte]

Where'd you get them? They are adorable!

[ROBOT420]

Just got this info from a fellow who ordered some from me off ebay.....

Hi Stephen,
I received the stepper yesterday and I have done done some preliminary
testing on then. I have a manual step controller. It's nothing but a
bunch or switches that switch in polarities to simulate stepping. I
started using 5 v and then pushed it up to 7 v. With continuous
current on the coils the stepper started to heat up noticably. I found
at this  voltage the current was 153 ma @ 7v. The steppers have an 8
step per revolution. I think this can be refined using half steps and
a dedicated controller chip. I'll be looking into that. Soldering a
lead to the post is something that you have to be quick and use a very
small iron. The plastic case for the post will melt in a heart beat! I
do not have anything definite on the torgue yet as the controller chip
when I find it and get it hooked up will answer this more directly. So
far the steppers seem well made and work consistently with the
primitive setup that I used to make the steps.
Again the controller chip is what I need to look into next. I will
keep you informed. I would be interested in ordering more in the near
future after I get a controller chip connected.

[ROBOT420]

@J, where ALL electronics come from.....China!
@ Sisco, Want some of them?

Problem is, I just got these MICRO motor driver (from Mouser) chips that make these steppers look big. I have no clue how to even go about soldering these new chips..... Any tips!?

http://s1362.photobucket.com/user/420photo420/media/20130619_164633_zps5ba3ca16.jpg.html?sort=3&o=0

http://s1362.photobucket.com/user/420photo420/media/20130619_164751_zps55c84154.jpg.html?sort=3&o=2

I have out MICROed myself on this one....

[waltr]

Thanks for the info you received on some tests.
I've been wondering what is the lowest Voltage for these steppers that still have reasonable torque.
Still didn't get the ones you sent me.

What are the part numbers (Mouser) of these driver chips?

They look like QFN packages that require a properly made PCB due to the small pads sizes and clearances.
The chip is soldered using solder paste put down with a mask and then re-flowed in an oven or with a hot air re-flow tool. Also note the large (sic) pad on the bottom of the chip. This pad also needs to be solder to the pcb and is the main transport of heat out of the chip.

I do designs with chips like that but no longer try to solder them. We have board assembly vendors and re-work people that are very good at doing these. A stereo-microscope is also required just to see these well enough to place them correctly on the pcb pads.

For a commercially produced pcb and assembly these chips are not a problem.

Now wait until you see the 0201 resistors and caps needed.

[ROBOT420]

Thanks for the info you received on some tests.
I've been wondering what is the lowest Voltage for these steppers that still have reasonable torque.
Still didn't get the ones you sent me.

What are the part numbers (Mouser) of these driver chips?

They look like QFN packages that require a properly made PCB due to the small pads sizes and clearances.
The chip is soldered using solder paste put down with a mask and then re-flowed in an oven or with a hot air re-flow tool. Also note the large (sic) pad on the bottom of the chip. This pad also needs to be solder to the pcb and is the main transport of heat out of the chip.

I do designs with chips like that but no longer try to solder them. We have board assembly vendors and re-work people that are very good at doing these. A stereo-microscope is also required just to see these well enough to place them correctly on the pcb pads.

For a commercially produced pcb and assembly these chips are not a problem.

Now wait until you see the 0201 resistors and caps needed.

Solder the CENTER bottom of those too!!? I am FU :-XED on them!  Well, at least I bought 26 of them to get a good price! 

Here is the Mouser part number for them is 595_DRV8837DSGR

As for the Steppers I told them to put them on the fastest Mule they had!lol If they are not there soon I will just send you some more.

[waltr]

With hot air reflow techniques soldering all the pins plus the center pad is not hard. There is also the 'frying pan' reflow method for at home DIY'ers (google will find info).

The chip you bought is a single H-bridge to drive one coil of the stepper so you need two chips per stepper.
TI (at Mouser) also has a dual H-bridge chip so only one chip per stepper is needed.
http://www.mouser.com/ProductDetail/Texas-Instruments/DRV8836DSSR/?qs=sGAEpiMZZMsLzCkzIJTB6yodKjxFpDyT

[ROBOT420]

Thanks
!

[waltr]

Steve, received the stepper you sent. Should have some time over the weekend to run tests on them.

[ROBOT420]

Steve, received the stepper you sent. Should have some time over the weekend to run tests on them.

Good deal, keep us posted!

[waltr]

Got to play with the stepper last night. I did break off the terminals of one after I soldered on 26ga wire. This wire was just too stiff for those tiny terminals. I was successful using 34ga enamel wire.

Some physical measurements first (note that these may not be exact, I used a dial caliber then assumed a nominal metric dimension).
The screw mounting hole has an M1.4 x 0.2 thread that is 0.030in (0.76mm) deep.
The gear is 8 teeth, Mod 0.25
Shaft diameter is 0.030in (0.762mm).

Now for the preliminary electrical.
I used my PIC16F84 controlling an Allego A3966 driver chip.
Vbb to the H-bridge FETs is 10V.

Now, a motor runs by the magnetic field and the field generated by a coil is dependent only on the current through the coil (B = m * i *n where i = current, n = number of turns, m = physics constant).
So, at DC (stepper holding a position) Ohm's Law holds for the current verse Voltage. These steppers coil R = ~38Ohm. Using this if the Voltage is larger than the current desired in the coil a series resistor can be used to decrease the current. The series resistor is one part of a resistive Voltage divider with the Coil's R as the other part. The calculation is: Vcoil = Idesired * Rcoil, Rseries = Rcoil *(Vsource - Vcoil)/Vcoil.

Since my H-bridge is powered at 10V and an earlier post said the steppers got hot at 7V and applying 5V directly to the stepper seemed to produce reasonable torque I calculated what the current would be at different Voltages and then what series resistor would be needed for a 10V source. These are:
@10V, 254mA, 0 Ohm
@7V, 182mA, 13.5 Ohm
@6.6V, 166mA, 22 Ohm
@5V, 131mA, 38 Ohm
@3V, 80mA, 87.5 Ohm

I have a bag of 22 Ohm resistors so place one in series with each coil and connected to the A3966 driver outputs. Applied powered to the PIC and driver. The PIC code starts sequencing the driver with half-steps (8 in the sequence) at a slow speed then gradually increases the stepping speed. The tiny stepper motor ran fine. After a minute of running the stepper's case was warm but not hot to the touch. By applying pressure the the gear with my finger the stepper could be stopped but the torque seemed to be rather strong for such a small motor.

Next is to build an apparatus to measure the torque output. This is difficult with a DC motor because to always has a continuous rotation. With a stepper one can measure the torque of one step.
My idea is to attach an arm to the output gear this is centered to have equal length, thus equal mass, of each side. With the stepper's shaft horizontal this arm is balanced and does not apply any torque to the shaft. Now add a weight (mass) at a know distance from the canter of the shaft. The torque this applies is easily calculated. Now apply power to a coil an see if it can lift the weight. If it can then increase the weight until it can not. The greatest weight it can lift is the Maximum torque.

Repeat but apply power to two coils.
Repeat but keep power applied to measure the holding torque.

Another test is to see how fast this motor will run (this may be driver and not motor dependent). I will also take a stepper into work and measure the Inductance of the coils.

Stay tuned for more.

[jwatte]

Thanks for the detailed report! I'd love to see a picture, too.
One thing about your current limiting resistor:
166 mA through 22 Ohms will dissipate about 0.6 Watts. If that "bag of 22 Ohm resistors" is the typical 0.25 Watt variety, you will burn out those resistors over time.
A better way to limit overheating is to use PWM. A 25% duty cycle at 10V generates roughly the same amount of torque and heat as a 100% duty cycle at 5V, as power through a fixed load is proportional to the square of the voltage.
The nice thing with a higher voltage and using PWM is also that you "force" the motor to the right position a little faster, as you have more voltage available to overcome the inductance of the windings.

[waltr]

Yep, the resistors are a bit under rated but this is just for testing the stepper motors. The one spec I am trying to determine is the least Voltage (current) that still provides adequate torque.

The Stepper motor driver chip (A3966) I'm using has built-in high frequency PWM based on an external current sense resistor. Check the A3966 data sheet to learn the details on how this works. I would set the current limit this way in a permanent circuit.

I'm not using this PWM limiting at the moment but thanks for the reminder. I may try it on these micro-steppers.

[waltr]

Ok, Got some Torque measurements done. This were measured with just one coil energized. The first picture shows the test setup. The motor is held by a parallel clamp. A small piece of wood was glued (CA) to the gear with holes drilled (0.02in) 35mm from the shaft center. This distance is the 'arm' for the torque calc. Pieces of resistor leads were put through the hole and form a loop to hang weight.
The weight is small pieces of solder.
The green board to the left is a stop to not allow the weight to hit the table when the motor is not energized. The picture show the motor energized, lifting the weight.

Small pieces of solder were added one at a time and the motor energized until it would not lift the weight. Then the last piece of solder removed the and remainder weighted on a precision balance.
The weight times the arm length equals the torque in gram-centimeter then converted to milliNewton-meter.

Coil Volt  coil I (Amp)     Torque (gm-cm)  (mN-m)
2.7          0.067               1.02                 0.10               
3.0          0.073               1.25                 0.12               
3.6          0.088               1.47                 0.14
4.7          0.110               2.04                 0.20       
6.3          0.145               2.4                   0.24

At 4.7V the motor does get very warm if not mounted but only slightly warm if mounted. The mounting does help in heat sinking the motor and if higher torque is required this should be a consideration.
At 3V the motor only get slightly warm.
As expected the Torque to linearly proportional to the coil current. Also, this steppers would work well at 2.7 to 3.6V and the H-bridge drive could be powered from a single LiPo battery.

The second picture is just a close up of the wire leads soldered onto the motor terminals.

[jwatte]

Is that unipolar or bipolar? If unipolar, it could be driven simply by four small transistors (like BS170 or 2N2222 or whatever.)

[waltr]

Bi-polar (only four leads to two isolated coils). So a dual H-bridge is required.

[ROBOT420]

Well I have to say that I am damn impressed Keep it up fellas!
I have plenty more if anyone else wants any just PM me.

[ROBOT420]

Waltr, gona send you some more for free tomorrow just for your fine work!

[waltr]

Thanks Steve.
I have two left that are not modified or broken.
Now to think of some thing to use these for and to design and built a small driver/controller circuit.

Any one have ideas?

[ROBOT420]

Waltr, 10 more headed your way!

For sure they would work to make a gauge for real-time monitoring of projects. I was thinking of an RC that is essentially just the RX with stuff mounted under it. What total weight would 4 of em do? Is moving an RX around an option?

[waltr]

Thanks Steve.

Sorry but I have no idea what you are asking about.
RC = resistor-capacitor? radio Control?
RX = receiver?

What total weight would 4 of em do?

4 of your micro steppers?
How much weight will they lift? Is this the question?
This will depend on the mechanics connected to the stepper.
Four steppers together should lift four times what one stepper will lift (minus losses due to several factors).

[Arduino02]

impressed!

[ROBOT420]

Thanks Steve.

Sorry but I have no idea what you are asking about.
RC = resistor-capacitor? radio Control?
RX = receiver?

What total weight would 4 of em do?

4 of your micro steppers?
How much weight will they lift? Is this the question?
This will depend on the mechanics connected to the stepper.
Four steppers together should lift four times what one stepper will lift (minus losses due to several factors).

Sorry,  . Radio Control. I was asking what would be a max weight to stay under for a small Remote Control car powered by 4 of these steppers. P.S. your steppers are on the way!