Running the steppers from 24v

[Brent Crosby]

So I experimented a bit with running my Gen 4 stepper boards (
http://wiki.makerbot.com/smd3 )from 24 v.

The A3877 and 7805 are both rated to more than 24 v, so I took the
chance!

I made a harness from three power supply splitters (good ones :) ) and
connected that to a 24v desktop power supply I had.

I tied the ground of the 24v to steppers harness back too the PC power
supply. Kinda scary having two supplies tied together, but it seems to
be working OK.

The 7805 still seems cool even at 24v, so I guess there is not much
current used from it.

My first tests indicate that we can step at about twice the speed as
expected. To test this I simply switched from 8 to 4 micro steep.
Instant 2x speed boost and no stalling! (BTW, the MS1 and MS2 switches
appear to be switched compared to the online documentation.)

I also have some 24v LED lights connected to the same 24v supply. They
flicker during certain stepping patterns, so the supply I used must
not be as stable as I had hoped. Well, now that I think about it, it
is only rated at 2.5A, so maybe the steppers can gang up and draw too
much current. I guess I'll have to measure the current and get a bench
supply or a beefier desktop supply.

Now to re-calibrate everything again tomorrow. Whoo Hoo!

[dremd]

Wow awesome!
Just to check, you just cut the 12 volt power wire and replaced it
with 24 colts?

On Feb 7, 11:59 pm, Brent Crosby <br...@crystalfontz.com> wrote:
> So I experimented a bit with running my Gen 4 stepper boards (http://wiki.makerbot.com/smd3)from 24 v.

[Andrew Plumb]

Hey Brent,

What stepper motors are you driving with these?  (The MakerBot machines' steppers are only rated to 14V.)

How are you managing heat generated in the A3977 chip?  That 2.5A max rating only holds if you have it well heat-sinked and possibly fan-cooled to keep the internal temperature within spec.

Andrew.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

--

Makerbot Number Nine... #9... 0x09... 0o11... 0b1001... 
http://clothbot.com/wiki/MakerBotNumberNine

[Brent Crosby]

> Wow awesome!
> Just to check, you just cut the 12 volt power wire
> and replaced it with 24 colts?

No young horses were involved. LOL.

I basically made a harness of three HDD power supply connectors,
removed the red wires (the 5v is provided by the 24v through the on-
board 7805). Soldered the three yellow ones to the +24v supply lead.
Soldered the six black ones to the supply ground lead, plus an
additional black wire that I ran over to a black wire near big
motherboard connector. That is an attempt to makes sure the power
supply grounds are equal. There are probably also grounds in the data
lines. This does make multiple grounding paths, which is generally not
a great idea but it seems to work, and is no worse than the original
setup in that manner.

I'll get some photos up today of how I did it. I am not sure the 24v
2.5A desktop supply is up to the task, I bought it for LED lighting
and never really checked the specs against what the 3x steppers would
need.

> What stepper motors are you driving with
> these? (The MakerBot machines' steppers
> are only rated to 14V.)

To my understanding of steppers, (which may be flawed) the limiting
issue is heat. Heat comes from putting in too much current at any
voltage. Since in theory the A3977 limits the current to the same
value at the motor for both 12v and 24v, the heat in the motor should
not change. If you put a constant 14v on a coil of the stepper I think
it would get very hot.

The electrically the motor is a coil. You want to make a magnetic
field with the coil. The magnetic field is created by the current in
the coil (which is why an ideal stepper motor driver would be an ideal
current source).

At rest, the average voltage that the A3977 has to apply to the motor
may be quite low for the motor to draw the current set by the A3977.
This state is governed by the motor's DC resistance and Ohm's law.

V = I * DCR

Since motors DCR is probably small, the A3977 will reduce the voltage
to a small value -- just enough to maintain I. This limits the total
power in the motor:

P = V * I

When a coil is switched on, the transient is governed by:

V = L di/dt

(the voltage seen on the motors coil is equal to the inductance of the
coil, times how fast the current is changing)

The L we can't really change. It is a function of how the motor is
wound.

We want a big di/dt (get the current going quickly so the magnetic
filed and thus the torque happens quickly, so the motor can keep
torque at higher speeds).

If you increase the V you can increase the di/dt and step the motor
more quickly.

Whether the motor can handle this is really about how well the A3977
does its job of limiting the current.

> How are you managing heat generated in the
> A3977 chip? That 2.5A max rating only holds
> if you have it well heat-sinked and possibly
> fan-cooled to keep the internal temperature
> within spec.

I have not changed the vref pot setting on the stepper, so hopefully
MBI set it reasonably and therefore current is reasonable.

I should measure that to makes sure it is an OK value.

The heat in the A3977 is largely dependent on the current and the DCR
of its internal FETs. It has an internal charge pump to drive the
gates of the FETs really high so they turn on hard. Cleverly, this
charge pump allows them to use the better N FETs to pull up and pull
down.

In any case, I'll give the A3977 a "rule of thumb" test today and see
if they have a reasonable package temperature.

The A3977 is a pretty cool chip: http://www.allegromicro.com/en/Products/Part_Numbers/3977/3977.pdf

Do you have a handy link to the Cupcake stepper motor data sheets? I
poked around on the wiki a bit but did not find it right away.

[Andrew Plumb]

On 2011-02-08, at 10:20 AM, Brent Crosby wrote:

[deletia]

Do you have a handy link to the Cupcake stepper motor data sheets? I
poked around on the wiki a bit but did not find it right away.

I followed the link from the NEMA-17 store entry (http://store.makerbot.com/nema-17-stepper-motor.html) to http://svn.makerbot.com/assets/datasheets/kysan-1123029.pdf

Andrew.

[Andrew Plumb]

Also w.r.t motor ratings, bare in mind that these are Kysan motors, the same folks who've had difficulty delivering consistent, dependable quality in their DC geared motors.  YMMV; proceed with caution. ;-)

Here's their full set of Kysan NEMA-17 stepper motor offerings: http://www.kysanelectronics.com/Products/SubClass.php?recordID=123

Here's the MakerBot one: http://www.kysanelectronics.com/Products/Detail.php?recordID=7435

Andrew.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

[Brent Crosby]

Andrew: Thanks for the stepper links. I'll try to read through them
tonight.

Here is a photo of the 24v supply I used. PLEASE KEEP IN MIND that
this is all "quickie proto" stuff. There may be trouble -- I still
need to analyze it.

http://sandbox.cfontz.net/~Brent/MiseryBot/MiseryBot_24v_Stepper_Proto.jpg

[Brent Crosby]

Great, the data sheet lists 14v at 0.4A, then their torque curves are
for 24v!

LOL

http://www.kysanelectronics.com/Products/datasheet_display.php?recordID=7435

I think the 14v comes from the DC resistance of 35 ohms:

0.4A * 35 ohms = 14v

0.4A * 14v = 5.6 watts

So if you were using a voltage source to drive the stepper, then yes
you need to limit it to 14v. Luckily the Allegro A3977 controls the
current.

I need to measure the currents, but my guess is that the thing we need
to worry about is the 0.4A max current. If the A3977 is set properly
to 0.4A average current the voltage at the motor should average 14v in
the DC case.

But when the inductance is dominating, the voltage will increase to
get better dI/dt

Still need more data . . . .

[Brent Crosby]

(Math is delicious!)

Looking at the A3977 data sheet:

Itrip_max = Vref / (8 * Rs)

Rs = 0.25 ohms (slightly more with traces)

Itrip_max should be 0.4A for these motors

So Vref should be Itrip_max * (8 * RS) = 0.4 * 8 * 0.25 = 0.8v

Can someone from MBI (Charles? Zach?) verify what Vref is set to at
the factory?

What current is your stepper motor running at? Measure REF to GND as
shown here:

http://www.flickr.com/photos/makerbot/5241567357/sizes/l/in/photostream/

Then divide by 2, that should be the stepper motor set current for
your bot.

It looks like with R1 turned down, you should be able to get REF close
to 0v. With R1 turned up you should be able to get it to:

(5v / (12K+47K)) * 47K = 4v

If the pot is cranked so Vref is 4v, then Itrip_max for 0,25 ohm Rs
is:

Itrip_max = 4v / (8 * Rs) = A

If the R1 pot were perfectly centered:

(5v / (12K+47K)) * 23.5K = 2v

which would make Itrip for the stepper 1A, or over twice the rated
current of the stepper.

Maximum voltage of SENSE should be below 0.5v:

0.25ohms * 0.4A = 0.1v

So that checks out for 0.4A

The recommended value of Rs is give by:

Rd = 0.5 / Itrip_max = 0.5/0.4 = 1.25 ohms

So the 0.25 ohm used by MBI may be a bit low for use with this chip on
this motor. On the other hand, taking it the other direction:

Itrip_max = 0.5 / Rs = 0.5 / 0.25 = 2A

So it looks like MBI set the smd3 to work up to 2A to give some
headroom for higher current motors.

Now I want to go measure the REF settings for my stepper controllers.

[Brent Crosby]

Maybe it should be 35 volts ?

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=285-1924-ND

(I guess you would have to use a couple of zeners to drop that 36 v to
35v)

[Brent Crosby]

OK, by using the DMM to measure Vref on all three of my Gen 4 stepper
motor driver boards, each of them was set very close to 2v. So I
assume that MBI is tweaking them to 2v at the factory.

2v on Vref with a 0.25 Rs should give 1 Amp in the motors.

The motors are rated at 0.4 amps, so that is 250% of rated current.

No wonder the Z motor gets hot as hell when it is enabled for the
whole print.

[dremd]

LOL!!
I should be more careful when posting from Iphone.

Thank you very much for experimenting + posting !

[Brent Crosby]

Considering that the Digi-Key supply was $84, I just ordered a 50v
bench supply:

http://shop.vendio.com/evan2002/item/770921907/index.html

That way we can monitor current and adjust the voltage.

> How are you managing heat generated in the A3977 chip?
> That 2.5A max rating only holds if you have it well
> heat-sinked and possibly fan-cooled to keep the
> internal temperature within spec.

Using the A3977 at 24v and 1A, both at idle and in the middle of a
large build, they are just barely warm. I can leave my finger pressed
against them for any length of time with no discomfort.