Welcome! Log In Create A New Profile

Advanced

Introducing the ramps 10 - "Ramps10" an open source multi-configurable stepper controller shield based on the Ramps 1.4 design.

Posted by Simba 
We are in the final stages of developing "Ramps10" an open source multi-configurable stepper controller shield based on the Ramps 1.4 design. It can control 10 motors up to 2A each, or 5 motors for 4A each, etc.

We do not know how it will be received by the DIY (reprap, cnczone, and robotics communities), but we think there is a need here. Personally, I can't believe people still use EMC and parallel ports with overpriced motor drivers and that some of the great reprap developments could work for CNC if the stepper controllers were more powerful. Also, we realized many boards would be needed to control more 10 steppers at once so we set out to build RAMPS10 to solve both needs. This isn't directly focused on 3D printing, more so it is for CNC controllers and robotics. However, with 10 axes (XYZWVABCDE) reprap could benefit by making more complex systems, such as printheads that rotate or many extruders, combined systems, etc.

Better yet, the output boards are made to be linkable, so we can have 5X 4 amp stepper outputs for beefy CNC applications and other permutations (2x 6 amp controllers, 2x 4 amp controllers), In all, I would love to have the reprap communities' opinion as this is a derivative work. As a first pass, we decided to greatly simplify the board just to focus on the stepper motor control aspect, so please remember this when considering features as it is not primarily intended for 3DP:

First, our specification:
Specification:
  1. Ramps10 is a link between Arduino mega and 10 compatible stepsticks
  2. The board is intended to handle 20A at up to 24V (30V theoretical limit) ~= 500W capacity
  3. Axes are easily defined as having 4 pins (Enable, step, dir, min) in firmware
  4. We have increased the self resetting fuses to handle 25A @ 30V. We feel above 30VDC is not safe for typical user application.
  5. There are no additional features such as thermistor, mosfets, servos, etc
  6. Instead, we provide an expansion header where the above features, plus other typical boards could be used (SDramps, LCD, etc) for the unused pins of arduino

Questions for you:
  1. Do you think anyone would use it?
  2. Is there any other non-chained board that could do the same thing?
  3. Can you buy a 250W 12V power supply or 500W 24V power supply without a problem?
  4. Are the boards really linkable (Would two or three linked stepstick outputs be able to drive a more powerful motor or would they clash?
  5. How many stepper motors would you want to control? 5? 6 ? 10? 20? 100?
  6. Are extra features necessary for you (like spindle control, dc motor driver, 120VAC control, heated bed, etc)
  7. Is there a better name than RAMPS10?
  8. Do you need more than 2-pins for an endstop (both min and max can be linked to this, but we feel optoendstops are going out of style due to costs and reliability)
  9. Do you need micro stepping? The boards claim 1/16th microstepping, which I can believe actually happens with no/low load. For direct drive from a stepper, I doubt more than 1/2 stepping is practical. We provide the same jumpers as RAMPS 1.4 for microstepping if desired.
  10. What is a good retail price for Arduino mega + ramps10 + 10 2A capacity stepsticks?
  11. Do you think we made the right choice in keep with Arduino Mega? We feel that the average non-tech heavy developer does not want to deal with the low-level code of ARM processors, so we chose not to use DUE, pi, etc. We also think PIC controllers are not ideal for this task. However, we could be near the limit of arduino processing power, depending on the firmware.

Edited 3 time(s). Last edit at 03/20/2013 02:20PM by Simba.

Measure once, Cut twice, Print 3 times.
Eye candy/loose screenshots of early development

I'm looking forward to it =)
I want to control 100 stepper motors DUH!, nah probably 10-12 is enough (Build a 5-axis 3d printer, with multiple extruders)

I'd say a 12V power supply would be easier to get than a 24V
Would need heated bed and thermisistor
probably is a better name but idc, name it w.e
microstepping would be useful to have
Try to make it into 1PCB?
I think you should go with the best for the job, don't have to to stick to arduino mega
I can easily get a lot of different power supplies here in Germany from an online electronics supplier, so this shouldn't be a problem in continental Europe.

As more and more people seem to be interested in some kind of arm designs like SCARAs, I can imagine the MEGA slowly becoming too limiting. It would be great if you can find another easy to use processor. I wonder if the MEGA will still be able to drive a 5 or 6 DOF robot arm? Alternatively, one could calculate the steps for these kinds of designs with the host computer and use the Arduino solely as stepper controller that reads axis steps instead of GCode.

Why would you do away with mosfets completely? I understand that you want a "pure" universal stepper board, but I would appreciate at least one general purpose mosfet. The idea to make the board extensible is pretty neat, though.

I would leave the end stop pins as they are now. Some people use hall effect end stops (which are supposed to be very precise and thus well suited for CNC applications). It is better to be more inclusive for alternative hardware, I think.

Have you considered alternative fuses? I have read some criticism about how the PTC fuses are not the best choice, so I am wondering whether others might also want to try different solutions. The Azteeg board uses fuses for automobile purposes for example.

What about external stepper drivers like in CNC applications? Would be neat to have easily accessible connectors to plug these things in for people who already have a working mill.

UGH. I sound extremely critical and negative, but actually like your idea.
Simba Wrote:
-------------------------------------------------------
> [*] Do you think anyone would use it?

For 3DP, possibly very few without support for 3DP features (heaters, thermistors etc).

> [*] Can you buy a 250W 12V power supply or 500W
> 24V power supply without a problem?

Sure, ebay. Ex-server PSUs are beefy and cheap.

> [*] Are the boards really linkable (Would two or
> three linked stepstick outputs be able to drive a
> more powerful motor or would they clash?

If you are hoping to have two stepper chips drive one motor, that will not work. sad smiley

> [*] How many stepper motors would you want to
> control? 5? 6 ? 10? 20? 100?

Probably 8+.

> [*] Are extra features necessary for you (like
> spindle control, dc motor driver, 120VAC control,
> heated bed, etc)

3DP features would be pretty essential for me.

> [*] Do you need more than 2-pins for an endstop
> (both min and max can be linked to this, but we
> feel optoendstops are going out of style due to
> costs and reliability)

Personally I use microswitches, but I like to leave design open to opto or hall.

> [*] Do you think we made the right choice in keep
> with Arduino Mega? We feel that the average
> non-tech heavy developer does not want to deal
> with the low-level code of ARM processors, so we
> chose not to use DUE, pi, etc. We also think PIC
> controllers are not ideal for this task.
> However, we could be near the limit of arduino
> processing power, depending on the firmware.

If you are designing a new Arduino shield, then it would be a shame not to make it compatible with Arduino Due. All the Allegro stepper chips will run fine with 3.3V logic, and also TI DRVxxx chips I believe. It would not take much I think to make it compatible with all Arduino boards, at least for 3.3V compatible stepper drivers. I think it is just a question of routing 3.3V/5V which works for all combinations of board, but I would not bother putting level converters in.

ETA: checked data sheet for DRV8825, it appears to be 3.3V logic compatible.

Edited 1 time(s). Last edit at 03/21/2013 01:32PM by bobc.
Good comments guys, thanks for the input.

We are designing RAMPS10 to be a beefy motor controller. therefore we agree with the need for a powerful output so we are installing a FET capable of up to 30 amps in the current version. I may try to include a thermistor, but i think it should go off chip.

It is important that linking two chips works. I can't see why it wouldn't - sharing the load, perhaps it could be noisy, that's why we have to build it and try : ). It certainly works in the opposite direction (one stepstick, two motors). I saw a post once by Nophead explaining how you could link multiple stepsticks with success but I can't verify that right now.
Also, the board has a first ever - overpowered regulator. It can take up to 24VDC and convert to 5VDC for arudino mega, so it can run independent of USB or act as USB host, rather than draining the source computer. This also keeps the arduino from being overheated (above 12V before). IF the DUE takes the 5VDC input and converts to 3.3V, it may work without modification?

Edited 1 time(s). Last edit at 03/21/2013 01:41PM by Simba.
Simba Wrote:
-------------------------------------------------------
> It is important that linking two chips works. I
> can't see why it wouldn't - sharing the load,
> perhaps it could be noisy, that's why we have to
> build it and try : ). It certainly works in the
> opposite direction (one stepstick, two motors). I
> saw a post once by Nophead explaining how you
> could link multiple stepsticks with success but I
> can't verify that right now.

How would the current sensing work, unless the drivers communicate with each other?

My day job is writing software for motor drives, the company I work for makes a 22kW drive made from two 11kW single drive units. They communicate over CAN, one is the master and does all the control, the other is a slave and just does what the master tells it.

If Nophead has figured out a better way to do it, we would be very interested winking smiley
What does current sensing do? The whole system is passive, based on assumptions.
yes i think u missed the point about current sense Mr Simba.

also ... each stepstick as it is ... is not designed to be in "bridged mode" like a stereo amplifier to get 2x the power.

for your "bridge mode" stepstick to work, you need to re-design the stepstick into a new type of "pairing" capable stepstick. it will prolly involve a daughter board that sense the drive frequency of 2 stepsticks. like kind of a frequency auto balancing circuitry (like in ham radio auto tuning circuit). it already sounds complex to me, am i right? or maybe someone can see something i didnt?

alternatively, you can design a new kind of super stepstick based on external H drive mosfets, now with this, i am very sure even 10A stepper can be beaten or eaten by the driver spinning smiley sticking its tongue out (as long your new power re-design is going to handle in excess of 1kw feeds). at this rate, i think it will def be a cartesian type gantry hybrid where i would mount a really big milling motor when i am not printing.

anyway, my yet to be finished machine uses 2 Z, 2 Y, 1 X, 1 extruder (maybe 2?).

my current gripe abt the ramps is that the stepsticks are so near to each other.
and current stepsticks are so tiny. if u are to do any mods to them like crazy me do, its a very tight squeeze.
btw are there any KiCad compatible files on ramps and stepsticks?


______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
redreprap Wrote:
-------------------------------------------------------
> my current gripe abt the ramps is that the
> stepsticks are so near to each other.
> and current stepsticks are so tiny. if u are to do
> any mods to them like crazy me do, its a very
> tight squeeze.
> btw are there any KiCad compatible files on ramps
> and stepsticks?

I have put the Kicad files for my RAMPS/Due project on github [github.com]. I used the Eagle schematic for RAMPS as a starting point, so if you strip out the stuff I added for Due compatibility then it should be very similar to RAMPS and you could use that how you like.

I also did a Kicad version of stepsticks for A4988 and A4982/5/7 these are at [github.com] and [github.com].

I have built and tested some prototype A4982 stepsticks from the above project, there are a few mods to the layout I haven't pushed to github yet, I've got the latest rev boards in front of me that need building and testing. The other projects are untried, so use at own risk etc, but it's all GPL so use as you like.

I was thinking of creating a new "stepstick" form factor based on a card edge design, I would call them "blades" but someone probably has a trademark on that. PCI edge connectors are very cheap (cheapest per pin), and the blade needs no connector pins so reduces the cost. By using the seeedstudio 5cm x 5cm option, a larger board can be used which allows better heat spreading (use copper as heatsink), more space for layout, more options for driver chips. (I have some Panasonic chips which are cheap but too big for current stepstick).

Downside of card edge system is that they are no longer compatible with 0.1 footprint, and they will need a physical latch to stop them working out of the connector.
arrrgh error error error ... lol cant open it


______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
The more I think about the name of this board, the more it seems to be misleading. If I got it right, the RAMPS10 is not a direct successor or replacement for any RAMPS board, but more a heavily modified fork or even an entirely different product. Naming it anything with RAMPS in the name might be confusing people who are new to RepRap.
Unfortunately, I cannot think of any decent alternatives now, but one element the name or acronym should contain in my opinion is "universal". "Universal Stepper Board" would be nice, but the acronym for that would be USB...bummer.
yea true ... this "ramps" is not a full ramps...

since its 10 stepsticks ... decastepsticker? confused smiley


______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
Using Latin enumeration is quite the nice idea I think...
Gradus would be the Latin word for step, but "decagrade" sound too much like some measurement unit like centigrade.
Using pes (foot) instead of step yields a nicer name: decapede. Sounds like some kind of biological denominator which might fit into the whole evolving, organic nature of RepRap.
decapede is cool
that in fact we realized the entire board is completely reworked and has practically no ramps element in it, we did use their circuit design for thermistor, so I don't consider this a true RAMPS board anymore, a new name is fitting.

For those who asked, we just added the expansion board design. It has
:EIGHT thermistor outputs!!!
:EIGHT addition FETs capable of 1A each @ 24V and 4 of them are PWM controllable
:4 D I/O expansion pins are left (for example, control of powerswitch tail devices).

what do you think now reprap grinning smiley
Attachments:
open | download - New board.pdf (39.3 KB)
well i really dont know simba, i have been putting my hands in and out of experimenting stuffs related to this for a few months now. TBH, i have yet to get to the parts that command the whole printer, i have been unable to put my whole brains into this as something else distracts me.

atm, my mods are simply, take a ramps, multiply it by using ramps as a master board, extract un-used pins, pull over to a new daughter board, use back all the same logics and firmware.

example, those machines that uses 2 steppers, instead of doing a series or parrallel wiring, i am adding another stepstick with signal feeds duplicated instead.

but realistically, if i were to do what you are doing, i will have to look at re-doing the whole firmware, or i dont? that what i do not know.

spinning smiley sticking its tongue out

Simba Wrote:
-------------------------------------------------------
> decapede is cool
> that in fact we realized the entire board is
> completely reworked and has practically no ramps
> element in it, we did use their circuit design for
> thermistor, so I don't consider this a true RAMPS
> board anymore, a new name is fitting.
>
> For those who asked, we just added the expansion
> board design. It has
> :EIGHT thermistor outputs!!!
> :EIGHT addition FETs capable of 1A each @ 24V and
> 4 of them are PWM controllable
> :4 D I/O expansion pins are left (for example,
> control of powerswitch tail devices).
>
> what do you think now reprap grinning smiley

Edited 1 time(s). Last edit at 03/29/2013 02:20PM by redreprap.

______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
Thanks for the input (and esp the name, uGen!)

Please see the full details here:
[www.reprap.org]

and updates here:
[armanicreations.com]

I propose around $250 for the board + 10 stepsticks + the reprap expansion for those referred to us by this site.

Edited 2 time(s). Last edit at 03/31/2013 05:36PM by Simba.
This is exactly what I need. Hopefully you can bring the price down a little in the future, though. $200 is about the cost of a RAMPS kit + 5 more stepper drivers, so thats the target you want to hit.

Simba Wrote:
-------------------------------------------------------
> Also, the board has a first ever - overpowered
> regulator. It can take up to 24VDC and convert to
> 5VDC for arudino mega, so it can run independent
> of USB or act as USB host, rather than draining
> the source computer. This also keeps the arduino
> from being overheated (above 12V before). IF the
> DUE takes the 5VDC input and converts to 3.3V, it
> may work without modification?

The Due has a 5V output but it does not appear to accept a 5V input. According to the Arduino documentation, it accepts 7-12V to the barrel plug and/or the VIN pin. The only way to power it with 5V is over USB. Couldn't you just pass the unregulated voltage from the power supply to VIN? This may also be helpful:

"The IOREF pin which allows an attached shield with the proper configuration to adapt to the voltage provided by the board. This enables shield compatibility with a 3.3V board like the Due and AVR-based boards which operate at 5V."
Okay, seems fair. Thanks for your support!!!
PS, we recently revamped the expansion board. IT is now up to the user to detemine the safe output levels. Each of the 8 fets on the reprap expansion can individually put out 5 amps - this was only intended to run 2 extra extruders, and some fans (but not 8 extruders, 5A each).
In fact, the entire board is versatile in that you as the user have to control the current so the total is <40A. You can be clever and time everything right so that you could have 20A of stepper motors, 20 A fet output, and 8x 5A outputs, just not all signaling high at the same time : )
What do you guys think of SPI interface control with 3A output and 1/128 microstepping?
"L6470dSPIN: Fully integrated microstepping motor driver with motion engine and SPI"
[www.st.com]

I think the microstepping is kinda overkill, unless you have several amps and lots of torque. Even for the purpose of smoothing movement, wouldn't the motor inertia dramatically counteract the "springiness" caused by motor coils beyond a few microsteps (1/4, 1/8th, etc)?
are you already trying out the driver?

128th step is quite a feat. cos even now it seems at 1/16th step, not all steppers are getting it right. ill be happy to try some 1/128s :p. speaking of which, @ 128x ... the Gcode load will be 8x more yes? in that case, the STL file will be like really huge eh?

Edited 2 time(s). Last edit at 04/07/2013 10:15PM by redreprap.

______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
These are the kinds of things to discuss.

Technically the processing is done on the chip entirely, the arduino just needs to tell it where to go (it has on board position tracking PER stepper motor). In fact, the same board does the acceleration coding locally too! So I think it is possible to even lower gcode size. However, I don't understand SPI yet to extrapolate more than this.

Can you tell me more about how 1/16th stepping doesnt "Get it right" currently? Whats going wrong?
like you said, the motor has its own inertia to overcome, and like nophead has explained in his own findings, each model of motor needs a specific drive to get it to work optimally. otherwise, you get miss steps or steps skipping. (you prolly need to see my blog #006th post)

nophead suggested to use a laser device to ascertain the real accuracy of the perstep movement, unfortunately, that device is impounded at my local customs. they claim it is a dangerous irradiation device! ... buggers ! however, if you saw my video, i did it the mechanical method with a pointer and a scale of sorts. not all steps can be 100% equal. this may translate to movements of say ... 10.0um, 10.5um, 9.3um, etc. i will be doing more of this once my micrometer dial gets here :p. apparently on another video i did see another guy having problems in the region of +/- 5-7um accuracy. not much, but it all adds up per mm as you get to 200mm by 200mm by 200mm

*edit* something really interesting about the china chinese way of driving steppers, everything that is a stepper seem to drive at 30volts ... even if it is rated at 12v. so to continue my "findings" like a frankenstein experiment on steppers, i have order some more steppers from china to see how other steppers will fare.

Edited 2 time(s). Last edit at 04/07/2013 10:44PM by redreprap.

______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
redreprap Wrote:
-------------------------------------------------------
> like you said, the motor has its own inertia to
> overcome, and like nophead has explained in his
> own findings, each model of motor needs a specific
> drive to get it to work optimally. otherwise, you
> get miss steps or steps skipping. (you prolly need
> to see my blog #006th post)
>
> nophead suggested to use a laser device to
> ascertain the real accuracy of the perstep
> movement, unfortunately, that device is impounded
> at my local customs. they claim it is a dangerous
> irradiation device! ... buggers ! however, if you
> saw my video, i did it the mechanical method with
> a pointer and a scale of sorts. not all steps can
> be 100% equal. this may translate to movements of
> say ... 10.0um, 10.5um, 9.3um, etc. i will be
> doing more of this once my micrometer dial gets
> here :p. apparently on another video i did see
> another guy having problems in the region of +/-
> 5-7um accuracy. not much, but it all adds up per
> mm as you get to 200mm by 200mm by 200mm

I would love to see that video - where can I find it?

I completely get your point about not all steps being created equal. It may imply something either you meant or I extrapolated from your statement. I totally forgot about this too. The accuracy of the stepper motor fabrication is far below the accuracy of even 1/16th steps. For that reason, trying to microstep in many cases would be deleterious - actually resisting inertia and non-linear ways, that would make even more vibration and resistance and feedback current.

I don't understand your point about missed steps. If a 3D printer makes a perfect printout state to finish, there may have only been an accumulation of a few steps of error maximum. My assumption has always been that no steps are missed.

Also, how does error add up - because within one rotation, it should be 100% consistent and accurate. Within the 360 degrees rotation, there would be some error, yes. This is specified in most motors, such as 8% step angle accuracy (that means if 3.6 degrees is specified, it could be ~3.3-3.9 degrees in reality).
For a reprap, microstepping is probably overkill since the current positioning resolution is already so much smaller than the detail that can be achieved with our current extruders. That being said, it certainly doesn't hurt to have it.

Having the stepper drivers on an SPI bus sounds nice, especially with some of the processing offloaded to them. It would require some major reworking of the firmwares, though. I also wonder if the bus could keep up with the small and rapid movements a printer makes.
aha, i think mechanically, every stepper is made in a std way with std grooves (i assume from pictures i have seen). how identical are those grooves i guess would be the answer to how "linear" is the result of each step (i assume again) as it move and stop and move and stop.

say for a certain motor to achieve perfect movement, it has to have 1A and 10volts drive. for a stepper driver to "know" it is supplying 1 A and 10 volts exactly is another matter. in a stepstick we know that it uses a Rsense n vref to measure this standard of "knowledge" internally. however, it is also true that every electronic circuit, there is noise (not from the TV or WiFi). and it is this spurious noise that will occasionally add some extra bits.

see this [3roomlabs.blogspot.sg]
and this [3roomlabs.blogspot.sg]

for a calibrated drive, i think the sense voltage is sitting well above the noise, and reads the feedback perfectly well to ensure that blocks of power are sent and stopped where it should be. but then again, i might be wrong, maybe the A49xx series does have a better signal to noise rejection. maybe i need a better scope LOL !!!

*edit* i forgot to say that, while mechanically, there are surely some inconsistencies in forces that may add to inaccuracies, electronically, you always have noise, which will always affect sense signals. say Vref of 1.5v, a 50mV noise (0.05v) = 31/3 % ... if you read the findings, there can be occasional noise spikes 100-200mV = 13 % !!! because these are cheap parts that i am working on, i will not know if popolus suffer any of these.

Edited 2 time(s). Last edit at 04/07/2013 11:19PM by redreprap.

______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
side tracked ... the bloody RIGOL DS2072 scope is asking me to buy it ... damn electronics toys ... WHY DO THEY BULLY ME ! spinning smiley sticking its tongue out


______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
microstepping is not overkill my friend, it is 3d printing porn ! lol (my gut feel)

when done right, we all do not need to "smoke" our ABS models in acetone to get shiny finish :p (but then again, having a fine finish AND smoking acetone ... even better ! haha)

finer and finer accuracies i believe is what will make it better ... or at least i assume. while bigger NEMA23 can get you brute force, fine detail is an entirely different beast (again i assume)

look at this rant about some un-fun (how do you say opposite of fun?) i have with numbers with many zeroes hahaha [3roomlabs.blogspot.sg]

smileys with beer

unlimitedbacon Wrote:
-------------------------------------------------------
> For a reprap, microstepping is probably overkill
> since the current positioning resolution is
> already so much smaller than the detail that can
> be achieved with our current extruders. That being
> said, it certainly doesn't hurt to have it.
>
> Having the stepper drivers on an SPI bus sounds
> nice, especially with some of the processing
> offloaded to them. It would require some major
> reworking of the firmwares, though. I also wonder
> if the bus could keep up with the small and rapid
> movements a printer makes.

Edited 1 time(s). Last edit at 04/07/2013 11:22PM by redreprap.

______________________________________
__my mixed bag blog || aka --> [http] || ___ so 3D printing is everywhere ... dont worry, hospitals can now 3Dprint body parts, they will charge you $1million excluding surgical fees ... you will die paying your debts. thats their aim ___ if every patent expires tomorrow, everybody will surely get a 3dprinter and make EVERYTHING ! ____ there is a "DIY-DTG" t shirt printing forum, you can mod an EPSON printer to PRINT like a pro. ___ CNCzone? overly commercialized it seems ___ my country? they will be taxing you for every cm of road you use and track you to your grave using GPS and its government authorized, now they will fire all the traffic wardens instead.___ EEVBLOG? there is only 1 way to do things --> take it apart like a pro
... superhigh resolution microstepping will slow down the speed and blow up the transferred data, as you have to 'tick' every single step.

I have 1/256 microstepping drivers for 1.8deg-steppers, that will perform 51200 steps per rev at highest resolution. The max. clocking speed with this setting can go up to 10MHz ... but which controller can run/calculate with such speeds confused smiley

The other problem is the absolute accuracy with changing loads - the micropositioning between the 1.8deg 'pole'-positions is depending on load or friction ... this can be off until maybe 1/3 of the full-step range eye rolling smiley

***
Other high-res microstepping drivers+motors I have do this with higher mechanical accuracy, but with much more expensive motors+electronics -- this are 3-phase stepers, that are driven with 325Volts and reconfigurable settings, so the 'fullstep' range can be switched between 200, 400, 500, 800 and 1000 steps per rev. simply by changing the coil activation. And additionally the drivers can be set to 1/10 microstepping, what gives 2000 to 10000 steps per rev.

The max. clocking speeds here are 20kHz with 1000 fullstep and 200kHz with 10000 steps/rev microstepping, what's the same max. speed, but with 10 times the amount of transferred data ...


Viktor
Sorry, only registered users may post in this forum.

Click here to login