New MKS SBASE Smoothieware-compatible board from China

Quote
arthurwolf

Quote
Dark Alchemist

Quote
w3drk
When switching from Marlin to Smoothie my only gripe -and this is with Smoothieware and not the SBASE- is Smoothieware doesn't support software end-stops. So if you don't have limit switches at both ends and not careful, you can easily jog your axis's past their mechanical limits and break stuff. Other than that it's been great.

Alright that has to be retarded

Well thank you for that ... That's nice.

I'm going to try to explain why, I hope I'm not so retarded that you won't be able to understand.

There is a process to adding things to Smoothie, which involves making sure we implement things *the right way*, rather than implementing it whichever way we think *might* work, and then fixing problems when they appear.

Smoothie is more abstracted than other firmwares. It turns out implementing soft endstops is more complex than it seems in that context.

I think soft endstops is the only significant basic feature Marlin has that Smoothie doesn't. We'll implement it at some point, but not before we figure out how to do it *right*, which we haven't yet. It doesn't really help that we only get asked for the feature very rarely : some things have a higher priority than others.

I'll go back to porting Smoothie to a whole new microcontroller familiy, with my retarded friends, now. Thanks.

I call them as I see them and if this feature isn't implemented then that was a retarded decision. When I turned Marlin's soft stops off I have many times hit the wrong button, or when trying to level a bed or find the max bed area, and had the damn X, and/or Y, sitting there slamming against nothing so I had to hit the disconnect and connect buttons on Pronterface to stop it. It might be a seldom asked for addition but it should have been part of the basic set. Mimic that which works FIRST from Marlin then implement the really cool stuff next. I know your eventually Arthur and eventually is a Euphemism from you meaning never but whatever you can go play with your friends now and I didn't say you were retarded just the decision that if it wasn't in it THAT was [a] retarded [decision]/

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
w3drk
I'm 100% positive it's not supported.

I posted a question directly related to this on the Smoothie forum a week or two ago and Arthur responded that it's not supported because it's apparently difficult to do correctly. I don't understand why it would be any more difficult to evaluate the current axis positions in addition to the limit switch states, but I'm not a firmware developer so I didn't push the issue.

Sometime soon I plan to look at the code to try and get familiar with it so I can understand what might be involved in getting software limits supported.

As we see Arthur chimed in and it isn't implemented and that decision to not implement it, for whatever reason, is just wrong. Start by implementing everything RAMPS does, and does well, then go from there showing the world how uber this new piece of hardware is. If it hasn't been implemented by now in the years it has been out it never will be unless someone like you does it since you already have one.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
Dark Alchemist
I call them as I see them and if this feature isn't implemented then that was a retarded decision.

I really don't think you know anywhere enough about Smoothie's codebase, project goals, or development guidelines, to say anything like this ...

" I want this, it's not there, therefore it's retarded " ... unfortunately I hear that way too often.

Quote
Dark Alchemist
It might be a seldom asked for addition but it should have been part of the basic set.

WHY ?

How do *you* define what should be part of the basic set ?

Smoothie is Open-Source coded by volunteers on their free time. The basic set is "whatever they feel like adding".

You seem to like soft endstops. Some people do want to see them implemented, but that's definitely not a popular wish.

Nor is it critical for safety, or day-to-day operation. Most people simply do not ask their machine to go outside the boundaries ...

Some people want it, and so, maybe somebody will come in and add it, or some of the current devs will do it ... in the meantime, it's not there.

Quote
Dark Alchemist
Mimic that which works FIRST from Marlin

You know when I say "we want to make sure we know how to do things right before doing them, so the code doesn't become an awful unmaintainable mess " ?

Yeah ... Marlin ...

I don't mean to criticize, it's just : different project goals/priorities.

It'd rather have clean code and not have soft endstops for a while. We used to not have half the things Marlin had. Now it's only soft endstops that are missing. And soon even that we'll have. But we take the time to do it RIGHT.

Quote
Dark Alchemist
I know your eventually Arthur and eventually is a Euphemism from you meaning never

It's definitely not : the simple fact that a lot of work has already gone into *preparing* the code to make adding soft endstops easier ( it's much more staightforward now than it used to be, just not enough ), essentially shows that you are talking out of your ass here.

We've said "eventually" to litterally hundreds of things that are now in the firmware.

If it's not in soon enough for you, don't be a dick about it, just wait or help. Or give encouragments. Or give us money, why not. But calling the decision to wait rather than doing it wrong "retarded", is just mean and not that smart.

Quote
Dark Alchemist
I didn't say you were retarded just the decision that if it wasn't in it THAT was [a] retarded [decision]/

Yeah, you are not an asshole, you are just somebody that says the sorts of things an asshole would say So no hard feelings.

Edited 2 time(s). Last edit at 12/28/2015 01:00PM by arthurwolf.

Wow this topic went down hill in a hurry.

While I agree the software limits are important I also understand that Smoothie is used for many things besides 3D printing, so in a way it needs to be more flexible than Marlin. If the majority of their users haven't requested software limits, then I guess it's just not a priority for the developers and who am I to argue? I can certainly appreciate their wish to maintain a cleanly developed firmware...and if you dig into the code, Marlin really is a mess. Besides, since I didn't buy an official Smoothieboard I don't feel it's right for me to complain so for the time being I'm enjoying Smoothieware as-is. When I find the time I'll look into getting software limits working myself.

For what it's worth I actually REALLY like Smoothieware and plan to make a donation soon.

Dieter

Quote
w3drk
Wow this topic went down hill in a hurry.

Yeah sorry about that, didn't like the "retarded" thing, will try to refrain from commenting further.

Quote
w3drk
For what it's worth I actually REALLY like Smoothieware and plan to make a donation soon.

\o/ Thanks, those go to Jim, the main firmware dev, he uses the money to buy hardware to help with the dev, mostly, so it's a lot of help.

Quote
arthurwolf

Quote
Dark Alchemist
I call them as I see them and if this feature isn't implemented then that was a retarded decision.

I really don't think you know anywhere enough about Smoothie's codebase, project goals, or development guidelines, to say anything like this ...

" I want this, it's not there, therefore it's retarded " ... unfortunately I hear that way too often.

Quote
Dark Alchemist
It might be a seldom asked for addition but it should have been part of the basic set.

WHY ?

How do *you* define what should be part of the basic set ?

Smoothie is Open-Source coded by volunteers on their free time. The basic set is "whatever they feel like adding".

You seem to like soft endstops. Some people do want to see them implemented, but that's definitely not a popular wish.

Nor is it critical for safety, or day-to-day operation. Most people simply do not ask their machine to go outside the boundaries ...

Some people want it, and so, maybe somebody will come in and add it, or some of the current devs will do it ... in the meantime, it's not there.

Quote
Dark Alchemist
Mimic that which works FIRST from Marlin

You know when I say "we want to make sure we know how to do things right before doing them, so the code doesn't become an awful unmaintainable mess " ?

Yeah ... Marlin ...

I don't mean to criticize, it's just : different project goals/priorities.

It'd rather have clean code and not have soft endstops for a while. We used to not have half the things Marlin had. Now it's only soft endstops that are missing. And soon even that we'll have. But we take the time to do it RIGHT.

Quote
Dark Alchemist
I know your eventually Arthur and eventually is a Euphemism from you meaning never

It's definitely not : the simple fact that a lot of work has already gone into *preparing* the code to make adding soft endstops easier ( it's much more staightforward now than it used to be, just not enough ), essentially shows that you are talking out of your ass here.

We've said "eventually" to litterally hundreds of things that are now in the firmware.

If it's not in soon enough for you, don't be a dick about it, just wait or help. Or give encouragments. Or give us money, why not. But calling the decision to wait rather than doing it wrong "retarded", is just mean and not that smart.

Quote
Dark Alchemist
I didn't say you were retarded just the decision that if it wasn't in it THAT was [a] retarded [decision]/

Yeah, you are not an asshole, you are just somebody that says the sorts of things an asshole would say So no hard feelings.

I just know that for me and how I do things that is a deal breaker and I hope it makes it in sometime but I suspect, even by your own words, that if it does make it in will probably be only because someone remembers to do it while getting ready for the 300 dollar V2/Pro beast (the new microcontroller I presume that you mentioned).

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
w3drk
Wow this topic went down hill in a hurry.

While I agree the software limits are important I also understand that Smoothie is used for many things besides 3D printing, so in a way it needs to be more flexible than Marlin. If the majority of their users haven't requested software limits, then I guess it's just not a priority for the developers and who am I to argue? I can certainly appreciate their wish to maintain a cleanly developed firmware...and if you dig into the code, Marlin really is a mess. Besides, since I didn't buy an official Smoothieboard I don't feel it's right for me to complain so for the time being I'm enjoying Smoothieware as-is. When I find the time I'll look into getting software limits working myself.

For what it's worth I actually REALLY like Smoothieware and plan to make a donation soon.

Dieter

Never had one and doubt that I will at the price it demands but that lack of software endstops is a deal breaker for me personally but for others I can't speak for them.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
Dark Alchemist
I just know that for me and how I do things that is a deal breaker and I hope it makes it in sometime but I suspect, even by your own words, that if it does make it in will probably be only because someone remembers to do it while getting ready for the 300 dollar V2/Pro beast (the new microcontroller I presume that you mentioned).

Actually, part of the work on the v2 firmware does include some major refactors that are easier to do on a port than on a currently used version, and some of those refactors would indeed make adding soft endstops easier. We didn't wait until you pointed this out to start working on it.

About v2, we have no idea how much the v2-pro will cost at this point, it'll depend on loadout anyway. For the "normal" v2 however ( which is similar to the v1 board, but with the more powerful chip, more gpio, and some other improvements ), we are aiming at a price similar, or even lower, to the current v1 board.

Quote
arthurwolf

Quote
Dark Alchemist
I just know that for me and how I do things that is a deal breaker and I hope it makes it in sometime but I suspect, even by your own words, that if it does make it in will probably be only because someone remembers to do it while getting ready for the 300 dollar V2/Pro beast (the new microcontroller I presume that you mentioned).

Actually, part of the work on the v2 firmware does include some major refactors that are easier to do on a port than on a currently used version, and some of those refactors would indeed make adding soft endstops easier. We didn't wait until you pointed this out to start working on it.

About v2, we have no idea how much the v2-pro will cost at this point, it'll depend on loadout anyway. For the "normal" v2 however ( which is similar to the v1 board, but with the more powerful chip, more gpio, and some other improvements ), we are aiming at a price similar, or even lower, to the current v1 board.

Too bad you couldn't do something like e3d with a lesser cost version. Something JUST for 3d printers to bring the cost factor down.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

No worries, I've done PCB and software development myself so I completely understand getting a little peeved when someone criticizes a project without taking the time to understand all the facts. Like Tolkien wrote, "The Wise only speak of what they know..."

And please don't refrain from comment, I feel putting it all out on the table can have a lot of value.

Since it's fresh on my mind I just made a $20 donation. It's not much, but I hope it's useful.

Edited 1 time(s). Last edit at 12/28/2015 01:22PM by w3drk.

Quote
w3drk
No worries, I've done PCB and software development myself so I completely understand getting a little peeved when someone criticizes a project without taking the time to understand all the facts. Like Tolkien wrote, "The Wise only speak of what they know..."

And please don't refrain from comment, I feel putting it all out on the table can have a lot of value.

Since it's fresh on my mind I just made a $20 donation. It's not much, but I hope it's useful.

A little thing about Arthur is when I went looking for information on the SBASE I found him, and this other guy, constantly arguing all over the Google groups to the point it turned my stomach so I see now why I saw all of those posts.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
w3drk
Since it's fresh on my mind I just made a $20 donation. It's not much, but I hope it's useful.

Thanks a lot ! We'll be setting up a "donations" page like Slic3r's to thank everybody at some point, working on it.

Quote
Dark Alchemist
A little thing about Arthur is when I went looking for information on the SBASE I found him, and this other guy, constantly arguing all over the Google groups to the point it turned my stomach so I see now why I saw all of those posts.

Why did you see all those posts ? I'm not sure I follow ( I'm probably not alone in that ).

What first brought the SBASE stuff to my attention was there was strong suspicions somebody was pretending to be a random happy SBASE customer, when he was actually somebody that knew the sellers ( also known as false advertising, pretty much ).
The guy later admitted this was true, so ...
If it wasn't for that specific problem, I probably would have refrained from commenting.

Appart from that, I think I've mostly only been involved in technical discutions on the subject. I actually think the SBASE board is pretty good, and debating technical stuff is part of the reprap culture, I don't see a problem with that.
On the other hand, I have a problem with non-open-hardware designs profiting from hardware designs/projects, but that's not specific to SBASE and Smoothie in any way.

I'd also like to point out that for every single thing I pointed out about SBASE, I wasn't alone in having those opinions, nor was I the first one to advance them ...

If you are referring to something else, I'd be interrested to know what.

Quote
arthurwolf

Quote
Dark Alchemist
A little thing about Arthur is when I went looking for information on the SBASE I found him, and this other guy, constantly arguing all over the Google groups to the point it turned my stomach so I see now why I saw all of those posts.

Why did you see all those posts ? I'm not sure I follow ( I'm probably not alone in that ).

When I was using Google to search out more information/new information on Smoothie and SBASE you pop up a ton fighting with this one guy and it seems to be this one guy who has it out for Smoothie. Being able to find those messages didn't help and made me leery of both products.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
Dark Alchemist
]When I was using Google to search out more information/new information on Smoothie and SBASE you pop up a ton fighting with this one guy and it seems to be this one guy who has it out for Smoothie. Being able to find those messages didn't help and made me leery of both products.

So like, if somebody wanted people to stop buying the Duet, all they'd have to do would be to start stupid arguments about it everywhere all the time, and then people would thing "urk ! I'm not going anywhere near a board that has this many arguments around it !" ?

That's not a good way to know if a board is good or not Look at the features, at feedback from people that used it, at the code, at the source files, at how responsive are the support forums/mailing lists/irc, the price, how good is the documentation things like that.

Any idiot can start a sterile argument, and then the board's fans have the choice between letting that idiot say bad things ( in which case people will think those bad things are true ), or try to explain why those bad things are not true ( in which case there will be arguing ).
So as long as somebody starts an argument, either you think bad things of the board, or you think you don't want the board because there is too much arguing.

There's no winning here, either way you don't want the board. That's giving a lot of power to an argument-starting idiot ...

Essentially you don't want any board that idiots start arguments about If you find the miracle board that had no argument started about it, tell me, and I'll try to balance things out ^^

Edited 2 time(s). Last edit at 12/28/2015 01:50PM by arthurwolf.

You should look over the Linux kernel development boards and read some of the things Linus Torvalds has said in the past. But there is no arguing the positive impact Linux (and Linus) has made on the computing world.

Quote
Dark Alchemist
]When I was using Google to search out more information/new information on Smoothie and SBASE you pop up a ton fighting with this one guy and it seems to be this one guy who has it out for Smoothie. Being able to find those messages didn't help and made me leery of both products.

Edited 1 time(s). Last edit at 12/28/2015 01:52PM by w3drk.

Quote
w3drk
You should look over the Linux kernel development boards and read some of the things Linus Torvalds has said in the past. But there is no arguing the positive impact Linux (and Linus) has made on the computing world.

« I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones. »
-- Torvalds, Linus (1991-08-25). Post. news:comp.os.minix. Google Groups. Archived from the original on unknown. Retrieved on 2006-08-28. This was the launch of Linux.

What an idiot ...

One of my favorite Linus moments...(keep volume down if at work)

[www.youtube.com]

Quote
arthurwolf

« I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones. »
-- Torvalds, Linus (1991-08-25). Post. news:comp.os.minix. Google Groups. Archived from the original on unknown. Retrieved on 2006-08-28. This was the launch of Linux.

What an idiot ...

Edited 1 time(s). Last edit at 12/28/2015 02:00PM by w3drk.

Hi, I have quastion more on smoothie firmware not SBASE board probably, I have problem with ZProbe, (Proximity Sensor LJ8A3-2-Z/BX) which is wired (S - Blue, G - Black, V - Brown - directly to board z_min, 5V), its sometimes show (repetier host, with M119) that the Z axis is triggered or sometimes zprobe is triggered but sensor ir far away from bed, but sometimes all is ok, maybe some one see in code what I didnt see? If in code all ok, than I will double check wires.

Also if I have used the G32 command it only seems to probe the first point (z motion), the second and third points it just travels to but doesnt stop to probe, but in good day its works just fine.

This all happens after some time when printer is on, when I just turn on, its works just fine usually.

(printer is from stainless steel, un and bed also)

I was searching for some info in smoothie forum, and find out that I am not alone with that issues:
[smoothieware.org]

# Endstops
endstops_enable                              true             # the endstop module is enabled by default and can be disabled here
corexy_homing                                true             # set to true if homing on a hbit or corexy
alpha_min_endstop                            1.24^!           # add a ! to invert if endstop is NO connected to ground
alpha_max_endstop                            NC               # NOTE set to nc if this is not installed
alpha_homing_direction                       home_to_min      # or set to home_to_max and set alpha_max
alpha_min                                    0                # this gets loaded after homing when home_to_min is set
alpha_max                                    150              # this gets loaded after homing when home_to_max is set
beta_min_endstop                             1.26^!           #
beta_max_endstop                             NC               #
beta_homing_direction                        home_to_min      #
beta_min                                     0                #
beta_max                                     150	      #
gamma_min_endstop                            1.28^!           #
gamma_max_endstop                            NC               #
gamma_homing_direction                       home_to_min      #
gamma_min                                    0                #
gamma_max                                    150              #

# optional order in which axis will home, default is they all home at the same time,
# if this is set it will force each axis to home one at a time in the specified order
homing_order                                 XYZ              # x axis followed by y then z last

# optional enable limit switches, actions will stop if any enabled limit switch is triggered
#alpha_limit_enable                          false            # set to true to enable X min and max limit switches
#beta_limit_enable                           false            # set to true to enable Y min and max limit switches
#gamma_limit_enable                          false            # set to true to enable Z min and max limit switches

alpha_fast_homing_rate_mm_s                  50               # feedrates in mm/second
beta_fast_homing_rate_mm_s                   50               # "
gamma_fast_homing_rate_mm_s                  4                # "
alpha_slow_homing_rate_mm_s                  25               # "
beta_slow_homing_rate_mm_s                   25               # "
gamma_slow_homing_rate_mm_s                  2                # "

alpha_homing_retract_mm                      5                # distance in mm
beta_homing_retract_mm                       5                # "
gamma_homing_retract_mm                      1                # "

#endstop_debounce_count                       100             # uncomment if you get noise on your endstops, default is 100

# optional Z probe
zprobe.enable                                true            # set to true to enable a zprobe
zprobe.probe_pin                             1.28^!          # pin probe is attached to if NC remove the !
zprobe.slow_feedrate                         5               # mm/sec probe feed rate
#zprobe.debounce_count                       100             # set if noisy
zprobe.fast_feedrate                         100             # move feedrate mm/sec
zprobe.probe_height                          5               # how much above bed to start probe
#gamma_min_endstop                           nc              # normally 1.28. Change to nc to prevent conflict,

# associated with zprobe the leveling strategy to use
leveling-strategy.three-point-leveling.enable         true        # a leveling strategy that probes three points to define a plane and keeps the Z parallel to that plane
leveling-strategy.three-point-leveling.point1         0.0,0.0     # the first probe point (x,y) optional may be defined with M557
leveling-strategy.three-point-leveling.point2         125.0,0.0   # the second probe point (x,y)
leveling-strategy.three-point-leveling.point3         75.0,120.0  # the third probe point (x,y)
leveling-strategy.three-point-leveling.home_first     true        # home the XY axis before probing
leveling-strategy.three-point-leveling.tolerance      0.03        # the probe tolerance in mm, anything less that this will be ignored, default is 0.03mm
leveling-strategy.three-point-leveling.probe_offsets  0,0,0       # the probe offsets from nozzle, must be x,y,z, default is no offset
leveling-strategy.three-point-leveling.save_plane     false       # set to true to allow the bed plane to be saved with M500 default is false

Edited 2 time(s). Last edit at 01/04/2016 03:59AM by shteecers.

So I think I've figured out why I constantly have to re-seat the Ethernet cable on my SBASE to keep it connected to my LAN. The clock source for the Ethernet PHY (SBASE component designation Y2) appears to be ceramic resonator. I just studied the PHY datasheet and according to Microchip's specification, the 25MHz clock source needs to be at least as stable as 50PPM (parts per million). From what I can tell the stability of a ceramic resonator is measured in percent (parts per hundred) and therefore don't even come close to that figure. Typical figures I see for 25MHz resonators is .3 to .5%....WAY out of the range of what's required.

I've contacted the seller a 2nd time to see if they are willing to swap it out for me, but I'm considering robbing the 25MHz quarts oscillator and matching load capacitors off another Ethernet device and retrofitting it in place of the ceramic resonator to see if it makes a difference. It must be right on the edge because the problem manifests itself less frequently when the room is warmer. Even warming the ceramic resonator with my fingertip is usually enough to get the Ethernet connection to sync up for a while.

Quote
w3drk
So I think I've figured out why I constantly have to re-seat the Ethernet cable on my SBASE to keep it connected to my LAN. The clock source for the Ethernet PHY (SBASE component designation Y2) appears to be ceramic resonator. I just studied the PHY datasheet and according to Microchip's specification, the 25MHz clock source needs to be at least as stable as 50PPM (parts per million). From what I can tell the stability of a ceramic resonator is measured in percent (parts per hundred) and therefore don't even come close to that figure. Typical figures I see for 25MHz resonators is .3 to .5%....WAY out of the range of what's required.

I've contacted the seller a 2nd time to see if they are willing to swap it out for me, but I'm considering robbing the 25MHz quarts oscillator and matching load capacitors off another Ethernet device and retrofitting it in place of the ceramic resonator to see if it makes a difference. It must be right on the edge because the problem manifests itself less frequently when the room is warmer. Even warming the ceramic resonator with my fingertip is usually enough to get the Ethernet connection to sync up for a while.

Now that is the type of reporting I like to see. If this is correct, and I don't doubt it, that is pretty bad and typical coming out of China who does whatever it wants and the devil to specs (if you buy anything electrical from China off of eBay or Banggood, etc... immediately test if the ground wire is hooked up because most items have the wire but they never bother to attach it for some nasty reason).

I hope someone else will chimes in about this as this is bad enough, at least for me, to not even consider it because if they did this here then what else did they do to cut corners and cause grief with?

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Yeah it certainly makes me less happy about my purchase, that's for sure. My primary reason for going with the SBASE over a Smoothieboard is the 1/32 micro-stepping, but ending up with flaky Ethernet isn't a worthy trade-off.

At first glance the quality of the PCB looks REALLY good for a clone...name brand components, high quality FR4 PCB, very nice layout, easy to read silkscreen, etc. I've done 10-15 hour prints back to back and it's been rock solid too. But choosing a loose-tolerance clock source for a device who's datasheet (and IEEE Ethernet requirements) clearly says at LEAST 50PPM stability or better just to save a few pennies per board is really cutting corners in my opinion, and a real shame how nice of a board it is otherwise. Yeah cheaping out on the clock source may turn out OK most of the time, but that doesn't make it right. And if it turns out I'm correct this is exactly the kind of issue that can arise when cutting corners like this because of something called tolerance stack.

Now I may be completely off-base on all of this, but as far I can tell nobody makes a quartz oscillator in the same footprint is these tiny resonators so I really don't see how it can be anything besides one. And based on my (admittedly small amount of) knowledge on the topic--ceramic resistors ARE NOT suitable for an Ethernet PHY reference clock source.

Edited 4 time(s). Last edit at 01/04/2016 03:39PM by w3drk.

Quote
w3drk
Yeah it certainly makes me less happy about my purchase, that's for sure. My primary reason for going with the SBASE over a Smoothieboard is the 1/32 micro-stepping, but ending up with flaky Ethernet isn't a worthy trade-off.

At first glance the quality of the PCB looks REALLY good for a clone...name brand components, high quality FR4 PCB, very nice layout, easy to read silkscreen, etc. I've done 10-15 hour prints back to back and it's been rock solid too. But choosing a loose-tolerance clock source for a device who's datasheet (and IEEE Ethernet requirements) clearly says at LEAST 50PPM stability or better just to save a few pennies per board is really cutting corners in my opinion, and a real shame how nice of a board it is otherwise. Yeah cheaping out on the clock source may turn out OK most of the time, but that doesn't make it right. And if it turns out I'm correct this is exactly the kind of issue that can arise when cutting corners like this because of something called tolerance stack.

Now I may be completely off-base on all of this, but as far I can tell nobody makes a quartz oscillator in the same footprint is these tiny resonators so I really don't see how it can be anything besides one. And based on my (admittedly small amount of) knowledge on the topic--ceramic resistors ARE NOT suitable a an Ethernet PHY clock source.

From what I remember from years ago (so it maybe off) ceramic resonators are not stable and are easily swayed with temperature.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Well, I was right. It's fixed now. It's an ugly hack--but I swapped the resonator out for a 25MHz quartz oscillator that I robbed from an old HP JetDirect 10/100 Ethernet interface. I also added the matching load capacitors (C52 and C53) where they were left out by Makerbase since the resonator evidently had them built in.

Network connectivity is rock solid every time on all my Ethernet switches I test it on. Before it only worked about 20% on the time, and only on an old 10/100 Switch.

I'm actually a little embarrassed to share these pics...but it gets the job done. I want to keep the hot-air use to a minimum to avoid damaging the old re-purposed components with unnecessary heat, so I'm leaving it exactly as it is now that it's fixed...ugly reflows and all.

Whatever it takes and as long as it works who cares what it looks like? I don't and I never got into reflow work just not my thing.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

I will say this, the eBay seller has been good at communicating with me and relayed my concerns back to Makerbase. They are evaluating my findings and apparently may revise the SBASE to correct this issue.

In the meantime, if anyone else runs into a flaky Ethernet port on their SBASE, they may want to consider swapping the resonator out for a 25MHz quartz crystal. Any old 10/100 NIC should have one.

Another option may be to use a Murata 25MHz "Hybrid Crystal Resonator" part # XRCGB25M000FAN00R0, available for about $0.55 from Digikey and likely any other electronics supplier. The footprint is completely different but I think mounting it diagonally might fit OK.

Quote
w3drk
I will say this, the eBay seller has been good at communicating with me and relayed my concerns back to Makerbase. They are evaluating my findings and apparently may revise the SBASE to correct this issue.

In the meantime, if anyone else runs into a flaky Ethernet port on their SBASE, they may want to consider swapping the resonator out for a 25MHz quartz crystal. Any old 10/100 NIC should have one.

Another option may be to use a Murata 25MHz "Hybrid Crystal Resonator" part # XRCGB25M000FAN00R0, available for about $0.55 from Digikey and likely any other electronics supplier. The footprint is completely different but I think mounting it diagonally might fit OK.

I looked all over and basically the fix is under a dollar for a one off so they should step up and change over as their cost would be magnitudes less than our one off fix (which we shouldn't have to do in the first place).

Edited 1 time(s). Last edit at 01/05/2016 12:12PM by Dark Alchemist.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

It's quite somehting you discovered this in the 1st place! What capacitors should be added to c52 and c53? I'm still waiting for my board to arrive, and i'm also an intense user of the ethernet port.. Did you give octofab a try on this board yet?

I transplanted the load capacitors directly from the 10/100 board I robbed for parts right along with the quartz crystal. I didn't even bother to get the LCR meter out to measure them, but I imagine they're between 12-30pF, which seems to be about the norm for this purpose.

If you end up needing to do this mod, I'd suggest getting NP0 capacitors for their high thermal stability. Regular bypass caps can drift and are probably unsuitable for this use.

Hopefully yours works fine as is! I may have just been unlucky enough to get one with a ceramic resonator too-far out of the required specification to work correctly.

Hah I've been too distracted with this issue to try Ocotofab yet. I plan to do that this weekend!

Board has arrived, Looks very decent. Trying to get the LAN to work, doesn't seem to be as friendly as planned.

The board hangs as soon as I enable the LAN port. Somethings wrong with it. Too bad. Lets hope the seller is willing to help me out. Another 3 weeks of waiting down the drain due to lack of QC in china.

Upon closer inspection, i'm pretty confident that the components currently used for the crystal/capacitors are a handmade/half assed mod; you can clearly see the flux.. I'll see if i can get a close-up on that.

Edited 4 time(s). Last edit at 01/06/2016 09:56AM by to3dornottobe.

It's frustrating because using Ethernet with Smoothieware rocks. I bet the issue is exactly the same as mine was. Replacing the resonator with a quartz crystal would probably fix it, but without the proper tools it's not exactly trivial without risk of damaging the board.

Hopefully they make things right for you...did you get it on EBay? The seller has actually been very understanding and is sending me a 2nd board to experiment with, but at the moment it seems sticking with an official Smoothieboard is the only safe option unless you don't mine tinkering like me.

To support Arthur's earlier points, this is one reason why these closed source copies are bad. Cost savings backfired, QC missed it, and customers who think they are getting a great deal end up stuck with a flawed board. If Makerbase properly licensed the board as Open Hardware and provided all of the supporting documents (not just schematics...KiCad/Eagle/whatever files, bill of materials, etc. too) issues like this could be avoided because the community would hold the fundamental design to a higher standard and not only focus on profit.

Unfortunately I bought mine before I fully understood how the SBASE came to be. I wanted 1/32 microstepping and liked how it would be a drop in replacement for my MKS-GEN since the sockets are identical, so I chose it over the Smoothieboard. Fortunately I was able to correct the Ethernet flaw and the board works fine now, but next time I'm sticking with an official board. Needless to say I'm looking forward to Smoothieboard v2.

Edited 3 time(s). Last edit at 01/06/2016 10:09AM by w3drk.