stepper motor question (again...)

Salam mes Amis,

I got hold of some stepper motors the other day, and now I'm wondering if I could give them a try.

Those are sanyo denki 103g771-0611 unipolar motors (6-wire)
rated 2,46 V and 1,82 A

Am I right to assume that if treated as bipolar motors they have a current of about 1.4A which could be handled by the L298?

thank alot for your help

'sid

I'm a bit confused where the 1.4A comes from. You can either drive them 'half winding' using the center taps and one end of each coil or you can run them 'full winding' using the end connections of each coil.

See page 7 & 8 for diagrams
[www.geckodrive.com]

If you are running them 'half winding' then the full current can be applied to them, 1.82A in this case.

If you are running them 'full winding' then half the unipolar max current can be used, 0.91A in this case.

The relatively low voltage of the steppers might be a challenge.
On the Stepper Motor Driver for the Arduino The 12v connection to the board is only used for the stepper and so could be connected to a 2.5V 2A power supply instead

The PIC's 'Stepper Motor Controller Card' uses the 12V both for the motor and to generate the 5v supply used by the other components. To use an alternative power supply for the motor the 5v regulator will need to be removed and replaced with a separate 5v power supply.

The
With the 297 it might be possible to add a voltage cut off to work in parallel with the current limiter.

Power is I^2 * R so you can actually use about 1.3A (1.82 / root2) in the full winding mode. This may give you more torque than 0.91A assuming the magnetics are not fully saturated.

Half winding is lower inductance so better for high speed running though.

The Arduino stepper motor driver board is a constant current chopper so no need to reduce the power supply voltage, 12V is fine. You just adjust the preset to give the current you want. I would recommend using half coil and setting the current to 1.82A maximum. You may find you can reduce it to run the motors cooler.

Edited 1 time(s). Last edit at 07/07/2008 02:34PM by nophead.

---

[www.hydraraptor.blogspot.com]

peteredworthy,
the 1.4 came from the formula nophead just posted, with a nice error in my headcalculation or let's say guestimating to simplify I took 2Amp instead of 1.8 to come up with nothing more than the sqrt2 itself .. there I forgot the last step reducing 10%

Well as far as google told me the motor can be run with up to 5V (at least that's what I think -language issues-)
And yes I've received the arduino set last thursday, so that's what I'm going to use.


nophead, I'm terribly sorry, but I don't get it.
It's fine to just connect the motors (in serial bipolar manner) and drop the amps within the arduino?
I thought I could only reduce the voltage...
And, isn't the halfcoil a pretty bad idea, doesn't max current mean that there's MORE heat and LESS torque? And isn't 1.7amps the upper limit for the L298 (as mentioned in the stepper-page [dev.www.reprap.org] ?

Lot's of questions here, but I don't like blowing things up with just the first test

'sid

The motor is rated for unipolar 1.82A. If you use half coil bipolar mode @ 1.82A you will get exactly the same heat dissipation and torque as the motor is designed for.

If you use use bipolar serial mode you have to reduce the current to 1.82 / root2 to get the same heat dissipation. The torque can be root2 times higher but if the magnetic materials are saturated at the rated current you may not get any more torque. But you have increased the inductance by a factor of 4 so your maximum speed will be less than unipolar mode.

With a chopper circuit the supply voltage is higher than the motor voltage. The current is controlled by switching it on an off very quickly. It is sensed by a small resistor and compared to a reference voltage set by a potentiometer.

The max current for the L298 is 2A per coil if you have a big enough heatsink.

---

[www.hydraraptor.blogspot.com]

Hi, Just trying to follow where my logic went wrong.

I get with the half vs full winding current I was trying to keep the voltage the same where power and hence heat build up is the actual limiting factor.

The current limiter also limiting voltage I'm having more trouble with, as I thought the effective resistance of a motor changed according to it's load. So under a light load/no load the motor might not be drawing the full current even at 12v.

My thoughts are:
- The resistance doesn't change that much
- Higher voltages are not a significant problem if the current isn't too high
- I get it even less than I thought ;-}>

The chopper does not limit voltage. Some of the time it applies 12V, the rest of the time it applies 0V.

The current in an inductor builds up slowly, at a rate equal to V/L initially. The chopper applies as much voltage as it can to build up the current as fast as possible. When it reaches the desired value it turns it off and it begins to fall again.

It is the current that determines the torque and also the heating. The voltage across the coil only affects how fast the current changes in a chopper circuit.

If you calculate the average voltage by multiplying the supply voltage by the duty cycle then that will equal the current times the resistance.

The load doesn't have a significant effect on the current drawn by a stepper because it does not slow it down. It is only DC motors where the load causes the speed to fall, which reduces the back EMF, hence increasing the effective voltage across the coil resistance and hence the current drawn.

---

[www.hydraraptor.blogspot.com]

Thanks nophead

That is almost chinese to me.
Okay what I do understand is:
with the configuration planned
I will not have to worry about the 12V thanks to the chopper
I can limit the current to the desired ~1.3 amps
I should check whether or not to use half or full coils regarding speed and torque

But where I get lost is the temperature.
I did not care about the motor heat dissipation, in fact I never thought of it;

nophead Wrote:
-------------------------------------------------------
> If you use half coil bipolar mode @ 1.82A you will get
> exactly the same heat dissipation and torque as
> the motor is designed for.

Now I'm curious to know what happens to the motorheat in fullcoil mode
Do I have to worry?

nophead Wrote:
-------------------------------------------------------
> The max current for the L298 is 2A per coil if you
> have a big enough heatsink.

And again; what is "big enough" ?
Must I use any kind of active fan cooling or is a midsized coppercooler (based on chip-dimension) enough?

What do you consider sufficient cooling for not blowing the L298 @2amps running several hours ?

Is there any operating experience you can share?

'sid

>But where I get lost is the temperature.
>I did not care about the motor heat dissipation, in fact I never thought of it;

It is the heat dissipation that determines the rating of the motor. At the rated power they get hot, often around 80C. If you exceed the power rating they will get hotter, risking melting the insulation.

>Now I'm curious to know what happens to the motorheat in fullcoil mode
>Do I have to worry?

With half coil the power is I^2 * R. With full coil it is I^2 * 2R. So the maximum current which does not exceed the power rating of the motor is root 2 times less. It doesn't really matter which part of the coil the heat is generated in. It all goes to heat up the whole motor.

>And again; what is "big enough" ?

To calculate the size of a heatsink required you look up the maximum junction temperature in the datasheet for the chip. You also get the thermal resistance from junction to case. Heatsinks also have a thermal resistance stated in degrees/W. You also need to know how much voltage the chip drops at 1.8A. It will be close to 4V IIRC.

You calculate the total power dissipation in the chip, which will be ~ 2 * 4 * 1.82 = 14.6W. The junction temperature is then ambient plus the wattage times the sum of the heatsink thermal resistance plus the junction to case thermal resistance. This temperature should be below the maximum stated for reliable operation. The lower it is the longer the chip will last.

I personally like to run my heastinks less than about 60C so I don't get burnt when I touch them but that is very conservative for a chip like this. The heatsink temperature is ambient plus power times heatsink thermal resistance.

>Is there any operating experience you can share?
I haven't actually used this chip myself but I do design electronics for a living.

You may find you do not need the full torque of the motor in which case it is better to wind down the current a bit to keep things cool. I.e. find the point where the motor starts skipping and wind the current up a bit until it is reliable.

---

[www.hydraraptor.blogspot.com]

Wow..
thank you nophead, although I do try to follow all you're saying, I must admit I'm unable.
You know I don't do this for a living, so I'm not that well informed.

Maybe it's a lack of english that I cannot understand every word you say (or is written in the datasheets)
In the datasheet there's a junction-ambient thermal resitance of 35

Sid,
The junction-ambient thermal resistance is what you get if you don't use a heatsink. With a heatsink it is the junction-case resistance plus the heatsink resistance.

So we have 14.6 * (Rh + 3) + 20 < 130 so yes you do need a heatsink 4.5K/W or better.

The junction temperature is the actual temperature of the transistors in the chip. The case and the heatsink will be 3 * 14.6 degrees lower, i.e. about 86C. Or another way to look at is the heatsink will be 14.6 * 4.5 + 20 = ~86.

---

[www.hydraraptor.blogspot.com]

Sid,
Actually thinking about this again: the chopper will only be on for about 2.46/(12 - 4) of the time (the motor voltage over the supply voltage minus what is lost in the chip). I.e. 31% of the time. That means you only need a heatsink that is 15C/W.

---

[www.hydraraptor.blogspot.com]

Now that's possible I guess..
best heatsink I've found (that seems to be mountable of course) has an 8.4K/W
Thanks alot now I know what I'll have to look for at the electronics store

You've been an enormous help

'sid

Actually I got is slightly wrong again (I don't call myself nop head for nothing!).

Power is 31% of 14.6W so 0.31 * 14.6 * (Rh + 3) + 20 < 130

So Rh > 21K/W, so the original heatsink is just big enough, with no margin at all. It will however get to 0.31 * 14.6 * 21 + 20 which is 115C, so very hot.

Your 8.6K/W heatsink will only get to 58C which is much safer and leaves a nice margin.

---

[www.hydraraptor.blogspot.com]

hehe..
I just came to say that I was wrong too...

The heatsink I've mentioned above fits the Multiwatt chip, but it does NOT fit on the pcb unfortunately.
Now I've ordered some other that hopefully fits, but has a thermal resistance of 11K/W if it doesn't, I'll have to take some closer looks at different electronic stores.

And I've stumbled over some tiny little fan (25x25 mm) maybe it fits somewhere,
I don't think it can drop the temperature much, but maybe a little is enough

'sid

Fans make an enormous difference to heatsink efficiency, so even a small one will work.

I prefer not to use them though because of noise, dust and the fact that they wear out.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> I prefer not to use them though because of noise,
> dust and the fact that they wear out.

LOL! They're also bloody expensive!

I tend to cheat and point the exhaust fans for my power supply at my heat sinks.

Edited 1 time(s). Last edit at 07/09/2008 09:48AM by Forrest Higgs.

I use power supplies without fans for the reasons above.

---

[www.hydraraptor.blogspot.com]

ATX power supplies, the cheapest around here, come with them.

You get what you pay for!

---

[www.hydraraptor.blogspot.com]

I guess I stayed in Africa too long. "Cheap" in Afrikaans translated as "goedkoop", which, literally translated is rendered "good buy".

LOL..

well yes, those tiny fans are expensive (in comparance to a big sized fan), the one I have laying around is just not enough to cool three stepper boards
but after all they're cheaper than buying a new l298

I'll give it a testrun if I manage to hook it up to the cooler.

Next step is to watercool the whole system I guess, according to pc-modding fanatics

'sid

I ran a L298N at two amps for days on end with a nice big aluminum heat sink. That chip runs very happily at temperatures suitable to fry bacon at.

Now that's an idea..
bacon and eggs made by the RepRap's stepper drivers whilst printing the plate and cutlery.

Full service here I come

'sid

Dear sir
Could you please provide me the formula to calculate turn/volt for EI typr carbone core (chopper), I want to make a IPS on my own. I need the output of 220v/50hz and 1000W capacity for my home. And it would be very nice if I could get a good circute diagram for it. Many thanks in advance.
Kallol

Sid, did those Sanyo motors work out for you? I see some on ebay.de (or are those yours that you're trying to get rid of? and wanted to hear if they're worth going for.

Thanks in advance,
-Lars Clausen

Old Processor fans and Case or PC PSU fans work great for £0.

I have a small CPU fan on the Gen2 Z axis driver with 1/3 of a 4" x 2" heat sink originaly bought for an LM317 or a TN3055 power transistor.

With the Gen2 X & Y axis drivers mounted 10mm or 1/2" apart each with their 1/3 of the heatsink with an Old PC case fan spanning the heatsinks see pics on my Blog.

The bigger Problem I had was the Stepper Motors sitting at 70c so I had to start using the power enable pin on the Gen 2 system & Arduino Diecimla this is coverd in the Blog also.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]