Innovative A4988 G3D Stepper Drivers

Hello!

We want to present you a new version of A4988 G3D stepper drivers for your 3D printers



4 pcs of professionally manufactured A4988 G3D Drivers complete soldered with mounted heatsinks - tested before shipping, just plug and play. Original Allegro A4988 inside.

Professional design and manufacture ensure stable operation of our driver - it contain double copper PCB for better heat dissipation and all the parts are from brands such as TDK, Murata, etc. - sense resistors are true sense resistors and all capacitors are Class 2 selected to work properly up to 25V!

This is the new innovative stepper motor driver for your 3D printer or any task involved stepper motors! It is a new generation as replacement for Pololu/StepStick it is fully compatible with Pololu and StepStick.


Q: Probably you'll ask why it is better and innovative ?

A: Using standard A4988 drivers you face the problem of vibration / shaking stepper motors and frame when stepper motors are moving (most noticeable on Z stage). These drivers solves this problem by using an additional 30k trimpot to correctly select the off-time settings. Is it complicated to use? No, you only need to turn second trimpot (OSC) fully counterclokwise and that's all - you can enjoy smoother and quieter stepper motors operation and no more lost motor steps because of non-optimal off-time settings.

Set includes:
- 4x A4988 G3D Drivers rev1
- 4x Heatsink (already mounted on the driver)
- 1x Safety plastic box

A4988 G3D Drivers are available here:
GADGETS3D Online Shop: [gadgets3d.com]
eBay auctions: [stores.ebay.com]











Best Regards,
Rafal
GADGETS3D

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

You sure have a lot of opensource logos but no access to the source files for these drivers.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Sublime Wrote:
-------------------------------------------------------
> You sure have a lot of opensource logos but no
> access to the source files for these drivers.

The files will be available when we finish the sale of the first batch.

Best Regards,
Rafal
GADGETS3D

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

RAF Wrote:
-------------------------------------------------------
> Sublime Wrote:
> --------------------------------------------------
> -----
> > You sure have a lot of opensource logos but no
> > access to the source files for these drivers.
>
> The files will be available when we finish the
> sale of the first batch.

In that case, you should not use the Open Source logos until after you have finished sale of the first batch. The logo means it is Open Source *now*, not some unspecified time in the future.

There is an obvious violation here, I think it is worth contacting the relevant organisations to see what they think.

bobc Wrote:
-------------------------------------------------------
> RAF Wrote:
> --------------------------------------------------
> -----
> > Sublime Wrote:
> >
> --------------------------------------------------
>
> > -----
> > > You sure have a lot of opensource logos but
> no
> > > access to the source files for these drivers.
> >
> > The files will be available when we finish the
> > sale of the first batch.
>
> In that case, you should not use the Open Source
> logos until after you have finished sale of the
> first batch. The logo means it is Open Source
> *now*, not some unspecified time in the future.
>
> There is an obvious violation here, I think it is
> worth contacting the relevant organisations to see
> what they think.

Hello,

Special for you I'll prepare the new photos

BTW: I thought this is Sales section

Best Regards,
Rafal
GADGETS3D

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

If all one has to do is turn the 30KΩ pot fully CCW then why isn't that a fixed 30kΩ or 0Ω resistor?
It doesn't make sense.

Nophead has published a number of articles on this issue, I think, the essence of which is to replace R4 on a Pololu/Stepstick with a short or a 0Ω resistor. Is this what you are doing with the 30kΩ pot?

Second trimpot turned fully counterclockwise is just a 0Ω

For you it doesn't make sense but allegro datasheet explaining this clearly:


"By pulling the ROSC pin to ground, mixed decay is set to
be active 100% of the time, for both rising and falling currents, and
prevents missed steps as shown in figure 3. If this is not an issue, it
is recommended that automatically-selected mixed decay be used,
because it will produce reduced ripple currents. Refer to the Fixed
Off-Time section for details."

It is just in case - not all motors are same and also not all applications are same this is why Allegro have few modes:

"ROSC tied directly to ground — off-time internally set to
30 μs, current decay is set to Mixed decay for both increasing
and decreasing currents for all step modes.

ROSC through a resistor to ground — off-time is determined
by the following formula, the decay mode is automatic Mixed
decay for all step modes.
tOFF ≈ ROSC ⁄ 825
Where tOFF is in μs."

And also this:

"Typically, (fully = ROSC tied to ground) mixed decay is only necessary when the current in the
winding is going from a higher value to a lower value as determined
by the state of the translator. For most loads automatically-selected
mixed decay is convenient because it minimizes ripple when the
current is rising and prevents missed steps when the current is falling.
For some applications where microstepping at very low speeds is
necessary, the lack of back EMF in the winding causes the current to
increase in the load quickly, resulting in missed steps."

We not just put there second trimpot but redesign PCB to traces matches 2A, VMOT capacitors placed near to IC as datasheet also suggest etc. so why to shorting ROSC into ground if we can give 30K trimpot for advanced users to choose off-time as they want instead of messing with soldering iron and 0402 resistors which are painfull to solder by hands.

Regards.

Edited 2 time(s). Last edit at 10/19/2012 02:53AM by RAF.

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

Early version of page is ready:

[reprap.org]

Edited 1 time(s). Last edit at 10/19/2012 02:57AM by RAF.

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

What didn't make sense was the "turn the pot fully CCW" that did not explain the need a 30kΩ pot. Your explanations here remedy that.

Good to see the schematic and Gerber files published under a CC license.

If motor (or you) is not happy with this (fully CCW) then turing trimpot CW you can go back to the default Poplolu/StepStick 10K or even 30K as you wants to - it is fun, flexibility and more control over the driver without messing with resistor soldering.

Regards.

Edited 2 time(s). Last edit at 10/19/2012 08:59AM by RAF.

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

@ RAF,

Where can i get eagle files (.brd and .sch) for this board?

Hello,
I just recently found out about the problems of the A4988... My main concern is the driver losing steps when the motor is forced away from where the driver tells it to be.
What I am looking for is something that can act like the drivers presented here, on the right:
http://www.youtube.com/watch?v=2SozZ7af3wg
Is this new driver capable of this kind of "memory"?
Thank you,
Eros

The motor on the right has an encoder on the back, you can see the extra wires. It is driven with a closed loop control system. So they are comparing a stepper driver with a driver plus a control loop.

---

[www.hydraraptor.blogspot.com]

I see. Well, any ideas where to get such encoders? I could really use that kind of reliability...
Thank you,
Eros

Here's a pretty smart IC: http://www.onsemi.com/PowerSolutions/product.do?id=AMIS-30623
If you could devise a Pololu-comptible driver like your G3D, but based on a chip like this, you'd have a winner! What do you think?
Eros

@Eros - thanks for info. We will check it soon

---

www.gadgets3d.com - parts for your 3D printer
GADGETS3D on eBay - parts for your 3D printer

You welcome, RAF!

In the meanwhile, I discovered some guys who already started on this path, but incomplete: they took the Trinamics chip and transplpanted it on an Arduino shield. It's not far from there to where I'm aiming...

http://www.motioncontrol-community.org/?page_id=353

I'm using these on my Prusa i3 build, and they work great. I keep the trimpot all the way CCW and the motors are quiet as can be. For some reason the Z motor driver is prone to overheating, maybe because it's attached to two motors, or because the Z axis is a little busier with retract_lift. But maybe it would help if I use "turn off motors when not in use" in the firmware. The next thing I plan to do is of course to add a fan to the electronics. But wouldn't it be great if there was a stepper motor driver that was immune to overheating? That would be my wish for your next revision, total heat immunity!

Following up, my Z max feedrate was set way too high in Marlin, so it wasn't overheating after all, just losing steps from trying to move too fast. Meanwhile I haven't seen any problems with these overheating after eliminating that false flag. Nevertheless I put a fan into my RAMPS box to cool things down. Now, of course, after pushing the speeds upward I have the new challenge of an overheating extruder motor, which is attached to a well-calibrated stepper driver. I think I'm just trying to run it too fast without any fan on the motor. So it looks like a 20mm fan on the motor would be the next thing to add...

I'm trying my Shinano Kenshi motors (STP-42D3018) in bipolar mode and getting unusual results. They had been working very well as Unipolar but I learned they might use less current and run cooler in bipolar mode, so I am trying that out. The problem I am seeing is that printing suddenly stops, which made me think possibly the G3D failsafe was tripping. I want to make sure these motors are compatible in this mode with the G3D drivers, and if all is good I can look into other causes for my woes.

The spec sheet for these motors shows the following requirements:

Unipolar: Set to 0.9A/phase, BSD-016, Vref=2.4V | Tested resistance of leads across half coils: ~6 Ohms
Bipolar: Set to 0.64A/phase, BSD-016, Vref=1.7V | Tested resistance of leads across full coils: ~12 Ohms

I'm running my RAMPS off a 30A PSU tuned up to 12V. I can turn up the voltage on this PSU if that is necessary.

I went through tuning the G3D REF voltage to a point where the motors exhibited enough power to do their work, plus 1/8 turn more. To make sure the problem is actually related to the switch to bipolar mode, tomorrow I will be trying the motors again in Unipolar mode to see if it eliminates the problem. And I may also try switching the wires (ABCD -> BADC) to see what effect that has. If there's anything interesting to report I'll mention it here.

I hope I can use these motors in Bipolar mode, as it seems to offer a good balance of torque and speed with reduced heat and current, especially important for my E motor which I want to keep as cool as possible.

==== LATER ====

According to the RepRap wiki "If the resistance is too high (i.e. 24V steppers) the current doesn't raise fast enough for reliable microstepping," and this seems to be the case with these 3 stepper motors. In bipolar mode the resistance is 12 Ohms. While this seemed to work okay for the XY motors, which don't need much torque, on the Extruder the motor can't maintain timing and/or torque, the motor runs choppy, and steps come out incorrect. This is most obvious after a retract move, where the motor moves too far backward, and/or not far enough forward on the move, leaving large gaps in the extrusion. And when running this motor in Unipolar mode it heats up past the failure point on longer prints (over 2 hours) even at average speeds. For longer prints I have to under-run this motor (at ~40% of the default speed).

So for best results with the G3D drivers, it looks like the best solution is to use motors with lower current, as recommended by the RepRap wiki. Although 24V motors like these can work in Unipolar mode, they simply can't perform as well as motors that are rated more closely to what the G3D steppers (and A4988 drivers generally) are designed for.

Edited 4 time(s). Last edit at 03/31/2013 06:08PM by Thinkyhead.

---

|
| Lead Developer of Marlin Firmware
| Help support my work at Patreon and Elsewhere.
|

Coming back after some recent experience with the A4988 from Pololu (testing config: Megatronics + A4988 + 42BYG011-25, powered up from a 12V PC case power supply)

1. The motor can hardly achieve 200-210rpm, before starting to buzz like crazy. Even this is not always true - sometimes it starts jittering just under 150rpm
2. Between 200-300rpm, the motor acts like a drunkard - moving spasmodically - and has near zero torque
3. over 300-350rpm, the motor simply stays still or buzzes still, with no rotation whatsoever
Note: Neither the motor nor the driver (get a chance to) heat up

Questions:
1. Am I asking too much in terms of RPM? What's the normal speed regime I should be expecting from a pretty small format NEMA stepper motor?
2. If I should however expect more RPM than this, does it mean the A4988 is to blame? I was initially tempted to purchase and try a G3D, but they only sell in 4-pieces packages and they're also based on the same A4988 IC, not inspiring much confidence to me after the Pololu driver failure
3. Besides the DRV8825, would you have any other suggestions for a good driver?
Note: I'm using motors rated between 0.5-1.5A and between 8-14V

Eros

The reason your top speed is limited is because 12V is not enough for motors rated 8-14V. You need a lower voltage motor or a higher voltage supply. We normally use 2.8V motors on 12V.

---

[www.hydraraptor.blogspot.com]

Hi, @nophead, and thank you for your reply!
I seem to be at a loss here, in terms of what the best thing should be for me, and in terms of recommended voltages/amperages. I thought if you have a 12V motor you should run it at or around 12V... Your reply leaves me at a brand new crossroad. Pumping 12V into a 2.8V motor won't burn it?

Needing to run on batteries, I am limited to overall smaller voltages and/or amperages. Megatronics works with a 12V supply, so I need to find out what are the best suited motors that could deliver both a decent torque (1Kg-cm) and more speed than I can achieve right now...

1. I already ordered the 18 motors, 35BYG302, rated 10.4V/0.4A. If you say I should run these at much higher voltages, for now I'll call it a loss and try to re-sell them when they arrive from China. For testing purposes, just tell me, please, how many Volts/Amps should these 10.4V/0.4A motors receive, to be able to run at higher speeds?

2. I already ordered and received the Megatronics boards - it's the only 6-motor solution I could find at a reasonable price. Could you please take a look at its specs PDF and tell me, in light of what you just told me about supercharging the motors or using lower power motors, what voltage/amperage should I look in the next motors to buy? I mean, I can't even tell if I can run a 2.8V motor from the Megatronics board... please help me!

Thank you so much,
Eros

Later Edit 1: looking at the same motor listing from Wantmotor, I can see a potential candidate: 35BYG304 (3.5V, 1A, 3.5Ω) - do you reckon it would work with the Megatronics? (they state something like 8-48V for motors, but I might have misread or misunderstood)

Edited 1 time(s). Last edit at 05/18/2013 04:36AM by erosnicolau.

If you connected 12V directly to a 2.8V motor then yes it would burn out but we use constant current drivers which limit the current to a target value. The supply voltage determines how fast it reaches that target value, so the higher the voltage the faster you can go. In practice the maximum supply voltage is determined by the driver, not the motor.

The Megatronics looks like it has a maximum of 24V for the motor supply. Yes it will drive 2.8V motors. It will drive any motor as long as the current is not too high and the motor voltage is less than the supply voltage. However, for fast operation you want a low inductance motor, low voltage motor.

If you look at the speed / torque curves you see they are for 24V operation and 35BYG204 is the flattest because it has the lowest inductance.

---

[www.hydraraptor.blogspot.com]

Thank you so much for your explanation, @nophead! You really shed some light on this one!

It means that we made a mistake buying those motors initially... Oh well, I'll wait for them to arrive and will try selling them online, to recover at least part of the "damage", and re-order other motors instead...

As for the 35BYG204: it has a lower torque (800g-cm) than what we need (at least 1000g-cm).

Comparing the 35BYG204 to the 35BYG304:
35BYG204: 6.84V, 0.76A, 4.8Mh, 800g-cm
35BYG304: 3.50V, 1.00A, 3.5Mh, 1250g-cm

35BYG304 seems to fit more into your advice: stronger, lower phase inductance, lower voltage. Am I wrong? Please advise me again

Thank you so much,
Eros

Sorry I meant to say 35BYG203 was the fastest. It has the lowest voltage and inductance. If you need more torque then use 35BYG304. It has more toque all the way up to ~4000pps but it falls off quicker due to having more inductance so at higher speeds the 203 will actually have more torque. It is a shame the graphs don't go up to 10000pps as they commonly do.

---

[www.hydraraptor.blogspot.com]

Thank you so much, man!

(I'm guessing pps = pulses per second, and therefore 4000pps = 20rps = 1200rpm)

The graphs are in half steps i.e. 400 per rev, so only 600 rpm.

---

[www.hydraraptor.blogspot.com]

Roger that!

I've been getting educated on similar topics in "Lower Voltage But Louder Motors?" and maybe some of the information there will also be useful for you.

---

|
| Lead Developer of Marlin Firmware
| Help support my work at Patreon and Elsewhere.
|