Smoothie Board V2 Prime

[DogBoy]

Hello, 

I know these topics have been discussed many time but....

I have been thinking about starting a Open PnP build for a few years now and have decided that I am ready (Time, money, space etc...)

At the moment I think that a new Smoothie Board V2 Prime would meet all my needs but before I pre order one, I would like to see what this forums thinks?

Of course every build is different and everyone's requirements are different, but in general, what is considered the most suited controller for Openpnp. Is it still a smoothie board?

What is the ultimate controller for Openpnp that is currently available? (Or close to being available). 

My build spec would be pretty standard: XY stage, 2 pnp placement heads, two camera (Up and down), and later multiple feeders.

I am also happy to have dedicated external drivers for each motor. 

My previous CNC build used Leadshines Easy Servos closed loop Nema 23 steppers, which are excellent. Maybe a V2 mini with external stepper or servo drivers?

Any general thoughts on this? 

Thanks

Paul 

[mark maker]

Hi Paul,

Welcome!

The Smoothie Board V2 Prime sounds very nice, but it is completely unproven. And I'm not sure about real world availability.

Note that even for Smoothieboard V1, I had to make a special firmware to make it work with advanced features of OpenPnP.

https://makr.zone/smoothieware-new-firmware-for-pnp/500/

So at least some of these changes would have to be integrated into V2 too. In the past I had no luck for them to accept my contributions.

If you're feeling adventurous and don't mind spending a lot of time and energy being the "pioneer" Smoothie v2 user, that would be cool. I would support this effort on the OpenPnP side, best I can. But like I said, the firmware needs to be fixed too, and I don't see myself doing this (again), so somehow we would have to convince them to fix remaining bugs and add needed features.

If you just want a proven, powerful controller available today, that has a firmware that is still very actively being improved, I would personally recommend a Duet3D model, which can go to 48V too, which is important for speed.

Otherwise there are some Smoothie V1 clones available, made by OpenPnP users, like Bing's Mircosmt PNP (which I'm currently testing), and Shai's Rapid Star Board  (hope I did not forget others...).

Overview:
https://github.com/openpnp/openpnp/wiki/Motion-Controller-Firmwares

_Mark

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/930af3cf-40c9-4f37-9fcf-343d3c4440bfn%40googlegroups.com.

[Arthur Wolf]

On Wed, Aug 30, 2023 at 11:39 AM 'mark maker' via OpenPnP <ope...@googlegroups.com> wrote:

Hi Paul,

Welcome!

The Smoothie Board V2 Prime sounds very nice, but it is completely unproven.

Both the hardware and firmware are evolutions of the v1 project, which has been extremely stable for years now.

It's been running in several machines for months now with no issues so far.

And boards have been shipping to Kickstarter backers for about two months now (~150 boards).

 

And I'm not sure about real world availability.

We have components secured (purchased) for hundreds of boards, and are producing in-house in the US at a steady pace.

Note that even for Smoothieboard V1, I had to make a special firmware to make it work with advanced features of OpenPnP.

https://makr.zone/smoothieware-new-firmware-for-pnp/500/

So at least some of these changes would have to be integrated into V2 too. In the past I had no luck for them to accept my contributions.

v1 was extremely memory and flash-constrained, we were getting at the limits of what could fit into the board.

That's one limitation that's gone with v2, where we have a lot more room to grow.

We're still planning on being very strict on the contribution process like we were on v1 though, it's worked very well to ensure the quality of the product, and contributions do make it through with some work/patience.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/cced2c4c-0f53-964b-e222-cea451898a54%40makr.zone.

--

勇気とユーモア

[bert shivaan]

Direct email sent

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CAMHZkm4q7pEz6zq6hRdc8yMibwYq4TnzTGa0ECHzrzH6ewiqBg%40mail.gmail.com.

[huynht...@gmail.com]

This is the board I designed and used for my AFARCO SMT project.  It is designed with 6 axes, Opto isolated home sensor and mosfet output used for LED, valve, pump.  In addition, I also designed the 0816 feeder control circuit right on the board, making the connection simpler and easier.  I revised Smoothie's USB communication library

Vào lúc 23:44:27 UTC+7 ngày Thứ Tư, 30 tháng 8, 2023, cncmachineguy đã viết:

[Teemo Vaas]

Hi.
Just found this info here. Although this is already few months old discussion the topic is still quite relevant.
I try to bring out some observations about the topic.

As I understand Smoothieware have been mostly standard firmware for Openpnp.

But it is not the main branch of the Smoothieware but fork at: https://makr.zone/smoothieware-new-firmware-for-pnp/500/

Main Smoothieware documentation server have been down already for some months. So that it seems it is in trouble. It is still accessible via the web archive. We would need to make the local copy of the documentation preferably to the our( makr.zone) github fork, to avoid losing it alltogether.

Smoothieware main is now moving towards V2 that has more memory, so the maintaining and accepting any improvements or fixes(pull requests towards) to the old V1 is most likely not happening.
It would render all old boards not usable/updatable from the original developer side, but we actually do not need the new "high memory consuming" features for openPnP so the old boards are actually quite fine for our use.

When comparing the possible firmware options there are three at the moment: Smoothieware, Marlin, Reprap(duet3D).

1.Smoothieware- in first glimpse good coding style and, we already have the "our own official" known good fork. .
Code review:
Pros:

• Supports LPC1768 processor (my personal preference, original smoothieboard V1 processor, lot of various clones and remakes exists, good proven platform)
  
•  We already have the "official" known good fork.
• Easy and very nice configuration via the config.txt file on the SD card. Can easily convert the same machine for different toolheads by changing only SD card.
• Code style seems nice at first glimpse. Clean, without developing ones own programming language via "macro magic".
•  Uses standard C++ classes and inheritance, so should be quite readable without deep learning curve.
• 
  

Cons: 

• Configuration via .txt file nessesitates that all features are compiled and stored in the flash of the microcontroller. So it tends to run out of memory for development, because all stuff is always in the memory.
• Possible RAM issues because RAM is allocated at runtime depending on the configuration in the text file, so some configurations might cause it to run out of RAM and into the undefined behaviour, not sure how this is managed.
• Nothing is selectable at the compile time to save some flash and ram.
• Not exactly Arduino style, main loop is not easily distinguishable, hard to profile the code and follow what runs with the main loop and what in the interrupts.

2.Marlin
 reviewed the code:

Pros:

•  Supports lot of different boards and microcontrollers. (including  LPC1768, so can be used on Smoothieboard V1 and compatibles)
• Good support for various LCDs and capable of driving them also with software SPI.
• Compile time conditional compilation depending on the configuration saves the flash ans ram on the chip. Unnedded parts for cpecific configuration are not compiled. 
  This also makes memory usage observable at the compile time, so there is no possibility of running out of RAM at runtime. Good practice for embedded.
• Can be compiled using Arduino IDE or PlatformIO. Setting up build environment was very smooth and painless.

Cons:

• Code is not good style, lot of hard to follow "macro magic", often unnessesary. Seems like almost trying to invent some own proprietary high level language.
• Some instantly noticeable rookie mistakes in the code (buggy millis() usage for example).
• Noticeable aim for "extra short code" at the expense of most times sacrificing readability(and possibly reliability).
• PWM outputs poorly implemented. Code regarding PWM stuff seems to be a mess. Not easy to add any hardware or software PWM channels. All PWMs seem to be running only at one single fixed frequency(bad when you need some fast and some slow PWMs)
• Configuration changes generally mean recompiling the firmware, though some things can be done via Gcode commands.

3. Reprap (duet3D)

General:

• Seems to be the base from where all the others are derived originally. 

Cons:

• Do not support  LPC1768 or LPC boards in general officially. Though a port for the LPC exists, but seems to be discouraged by the main branch...
• Would need to maintain own fork if wanting to use the LPC based existing boards. Seems like some extra work...
  
  

[Clemens Koller]

Hi!

http://smoothieware.org/ seems to be down. I hope not a permanent thing.

Is there an overview somewhere about the different most used motion controllers
used with OpenPnP especially with a feature-list of what's implemented/possible
and what's not... like a side-by-side comparison?

I've started with Synthetos/TinyG which came with my Liteplacer and now I am migrating
the whole thing to a Duet3D/Duet 3 Main Board 6HC which seems to be way more fun
and an improvement in reliability. The steppers are way more silent and the connectivity
and performance is way better / worth the elevated price.

Regards,

Clemens

On 18/12/2023 19.11, Teemo Vaas wrote:
> Hi.
> Just found this info here. Although this is already few months old discussion the topic is still quite relevant.
> I try to bring out some observations about the topic.
>
> As I understand Smoothieware have been mostly standard firmware for Openpnp.
>

> But it is not the main branch of the Smoothieware but fork at: https://makr.zone/smoothieware-new-firmware-for-pnp/500/ <https://makr.zone/smoothieware-new-firmware-for-pnp/500/>

>
> Main Smoothieware documentation server have been down already for some months. So that it seems it is in trouble. It is still accessible via the web archive. We would need to make the local copy of the documentation preferably to the our( makr.zone) github fork, to avoid losing it alltogether.
>
> Smoothieware main is now moving towards V2 that has more memory, so the maintaining and accepting any improvements or fixes(pull requests towards) to the old V1 is most likely not happening.
> It would render all old boards not usable/updatable from the original developer side, but we actually do not need the new "high memory consuming" features for openPnP so the old boards are actually quite fine for our use.
>
> When comparing the possible firmware options there are three at the moment: Smoothieware, Marlin, Reprap(duet3D).
>
> 1.Smoothieware- in first glimpse good coding style and, we already have the "our own official" known good fork. .
> Code review:
> Pros:
>

> * Supports LPC1768 processor (my personal preference, original smoothieboard V1 processor, lot of various clones and remakes exists, good proven platform)
> *  We already have the "official" known good fork.
> * Easy and very nice configuration via the config.txt file on the SD card. Can easily convert the same machine for different toolheads by changing only SD card.
> * Code style seems nice at first glimpse. Clean, without developing ones own programming language via "macro magic".
> *  Uses standard C++ classes and inheritance, so should be quite readable without deep learning curve.
> *
>
> Cons:
>
> * Configuration via .txt file nessesitates that all features are compiled and stored in the flash of the microcontroller. So it tends to run out of memory for development, because all stuff is always in the memory.
> * Possible RAM issues because RAM is allocated at runtime depending on the configuration in the text file, so some configurations might cause it to run out of RAM and into the undefined behaviour, not sure how this is managed.
> * Nothing is selectable at the compile time to save some flash and ram.
> * Not exactly Arduino style, main loop is not easily distinguishable, hard to profile the code and follow what runs with the main loop and what in the interrupts.

>
>
> 2.Marlin
>  reviewed the code:
> Pros:
>

> *  Supports lot of different boards and microcontrollers. (including LPC1768, so can be used on Smoothieboard V1 and compatibles)
> * Good support for various LCDs and capable of driving them also with software SPI.
> * Compile time conditional compilation depending on the configuration saves the flash ans ram on the chip. Unnedded parts for cpecific configuration are not compiled.

> This also makes memory usage observable at the compile time, so there is no possibility of running out of RAM at runtime. Good practice for embedded.

> * Can be compiled using Arduino IDE or PlatformIO. Setting up build environment was very smooth and painless.
>
> Cons:
>
> * Code is not good style, lot of hard to follow "macro magic", often unnessesary. Seems like almost trying to invent some own proprietary high level language.
> * Some instantly noticeable rookie mistakes in the code (buggy millis() usage for example).
> * Noticeable aim for "extra short code" at the expense of most times sacrificing readability(and possibly reliability).
> * PWM outputs poorly implemented. Code regarding PWM stuff seems to be a mess. Not easy to add any hardware or software PWM channels. All PWMs seem to be running only at one single fixed frequency(bad when you need some fast and some slow PWMs)
> * Configuration changes generally mean recompiling the firmware, though some things can be done via Gcode commands.
>
> 3. Reprap (duet3D)
> General:
>
> * Seems to be the base from where all the others are derived originally.
>
> Cons:
>
> * Do not support LPC1768 or LPC boards in general officially. Though a port for the LPC exists, but seems to be discouraged by the main branch...
> * Would need to maintain own fork if wanting to use the LPC based existing boards. Seems like some extra work...

>
>
> On Thursday, August 31, 2023 at 11:15:53 AM UTC+3 huynht...@gmail.com wrote:
>

> This is the board I designed and used for my AFARCO SMT project.  It is designed with 6 axes, Opto isolated home sensor and mosfet output used for LED, valve, pump.  In addition, I also designed the 0816 feeder control circuit right on the board, making the connection simpler and easier.  I revised Smoothie's USB communication library20230823_104058.jpg20230823_122418.jpg

> Vào lúc 23:44:27 UTC+7 ngày Thứ Tư, 30 tháng 8, 2023, cncmachineguy đã viết:
>
> Direct email sent
>
>
> On Wed, Aug 30, 2023 at 6:08 AM Arthur Wolf <wolf....@gmail.com> wrote:
>
>
>
> On Wed, Aug 30, 2023 at 11:39 AM 'mark maker' via OpenPnP <ope...@googlegroups.com> wrote:
>
> Hi Paul,
>
> Welcome!
>

> The Smoothie Board V2 Prime sounds very nice, but it is /completely/ unproven.

>
> Both the hardware and firmware are evolutions of the v1 project, which has been extremely stable for years now.
> It's been running in several machines for months now with no issues so far.
> And boards have been shipping to Kickstarter backers for about two months now (~150 boards).
>

> And I'm not sure about /real world /availability.

>
>
> We have components secured (purchased) for hundreds of boards, and are producing in-house in the US at a steady pace.
>
> Note that even for Smoothieboard V1, I had to make a special firmware to make it work with advanced features of OpenPnP.
>

> https://makr.zone/smoothieware-new-firmware-for-pnp/500/ <https://makr.zone/smoothieware-new-firmware-for-pnp/500/>

>
> So at least some of these changes would have to be integrated into V2 too. In the past I had no luck for them to accept my contributions.
>
>
> v1 was extremely memory and flash-constrained, we were getting at the limits of what could fit into the board.
> That's one limitation that's gone with v2, where we have a lot more room to grow.
> We're still planning on being very strict on the contribution process like we were on v1 though, it's worked very well to ensure the quality of the product, and contributions do make it through with some work/patience.
>

> If you're feeling adventurous and don't mind spending a lot of time and energy being the "pioneer" Smoothie v2 user, that would be cool. I would support this effort on the OpenPnP side, best I can. But like I said, the firmware needs to be fixed too, and I don't see myself doing this (again), so somehow we would have to convince /them /to fix remaining bugs and add needed features.
>
> If you just want a proven, powerful controller available today, that has a firmware that is still very actively being improved, I would personally recommend a Duet3D model <https://docs.duet3d.com/en/Duet3D_hardware/Hardware_overview>, which can go to 48V too, which is important for speed.
>
> attainable speed at required torque <https://makr.zone/choosing-a-motion-controller-the-panucatt-azteeg-x5-gt-32bit/455/>
>
>
> Otherwise there are some Smoothie V1 clones available, made by OpenPnP users, like Bing's Mircosmt PNP <https://www.microsmt.com.cn/products/mainboard-for-openpnp> (which I'm currently testing), and Shai's Rapid Star Board <https://www.deltaprintr.com/product/rapid-star-board/> (hope I did not forget others...).
>
> Overview:
> https://github.com/openpnp/openpnp/wiki/Motion-Controller-Firmwares <https://github.com/openpnp/openpnp/wiki/Motion-Controller-Firmwares>

>
> _Mark
>
> On 30.08.2023 09:34, DogBoy wrote:
>> Hello,
>>
>> I know these topics have been discussed many time but....
>>
>> I have been thinking about starting a Open PnP build for a few years now and have decided that I am ready (Time, money, space etc...)
>>
>> At the moment I think that a new Smoothie Board V2 Prime would meet all my needs but before I pre order one, I would like to see what this forums thinks?
>>
>> Of course every build is different and everyone's requirements are different, but in general, what is considered the most suited controller for Openpnp. Is it still a smoothie board?
>>
>> What is the ultimate controller for Openpnp that is currently available? (Or close to being available).
>>
>> My build spec would be pretty standard: XY stage, 2 pnp placement heads, two camera (Up and down), and later multiple feeders.
>>
>> I am also happy to have dedicated external drivers for each motor.
>> My previous CNC build used Leadshines Easy Servos closed loop Nema 23 steppers, which are excellent. Maybe a V2 mini with external stepper or servo drivers?
>>
>> Any general thoughts on this?
>>
>> Thanks
>> Paul
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups "OpenPnP" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.

>> To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/930af3cf-40c9-4f37-9fcf-343d3c4440bfn%40googlegroups.com <https://groups.google.com/d/msgid/openpnp/930af3cf-40c9-4f37-9fcf-343d3c4440bfn%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/cced2c4c-0f53-964b-e222-cea451898a54%40makr.zone <https://groups.google.com/d/msgid/openpnp/cced2c4c-0f53-964b-e222-cea451898a54%40makr.zone?utm_medium=email&utm_source=footer>.

>
>
>
> --
>
> 勇気とユーモア
>
> --
> You received this message because you are subscribed to the Google Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
>

> To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CAMHZkm4q7pEz6zq6hRdc8yMibwYq4TnzTGa0ECHzrzH6ewiqBg%40mail.gmail.com <https://groups.google.com/d/msgid/openpnp/CAMHZkm4q7pEz6zq6hRdc8yMibwYq4TnzTGa0ECHzrzH6ewiqBg%40mail.gmail.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google Groups "OpenPnP" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com <mailto:openpnp+u...@googlegroups.com>.
> To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/afab5729-7ac0-461a-b6ed-1e16b1a6cc56n%40googlegroups.com <https://groups.google.com/d/msgid/openpnp/afab5729-7ac0-461a-b6ed-1e16b1a6cc56n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Arthur Wolf]

On Fri, Dec 22, 2023 at 10:01 PM 'Clemens Koller' via OpenPnP <ope...@googlegroups.com> wrote:

Hi!

http://smoothieware.org/ seems to be down. I hope not a permanent thing.

Fighting with Hetzner to get it back up. Hopefully this month.

Web archive has it in the meantime.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/f91acc8b-93b9-4d82-bcf5-001fece47b52%40gmx.net.

--

勇気とユーモア