Smoothieboards unreliable - what new controller to support?

General hangout discussion area for other non-printing stuff
User avatar
626Pilot
ULTIMATE 3D JEDI
Posts: 1720
Joined: Tue May 14, 2013 12:52 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by 626Pilot »

I have been in contact with Arthur Wolf.

All is forgiven.

I am still going to move to another platform (probably BBB) and then take a look at returning to Smoothie next year, when they have their version 2 board out.
User avatar
teoman
ULTIMATE 3D JEDI
Posts: 1783
Joined: Sat May 24, 2014 5:43 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by teoman »

626Pilot wrote:I have been in contact with Arthur Wolf.

All is forgiven.

I am still going to move to another platform (probably BBB) and then take a look at returning to Smoothie next year, when they have their version 2 board out.

It should be an easy switch between the two if the code is well written. They both run Linux. (except if code is very heavily ported on to the aux processors of the BBB).
When on mobile I am brief and may be perceived as an arsl.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by Generic Default »

I'm still glad I bought the smoothies, there have a been a few problems with them over the year and a half I've had them but they work great most of the time. I think I'm the only person driving a lathe with smoothie, and I haven't seen any 4 axis TAIG mills with smoothies either. The stepper current of 2 amps is nice and it allows for bigger steppers with faster travel speeds.

However, with some of the alternatives coming on the market, I think smoothie will have to adapt to keep up. The duet boards linked on the previous pages can support up to 9 stepper motors, which is enough to operate a 5 axis CNC mill turn center with an automatic tool changer and a bar feeder/puller. Or anything else that the firmware can be adapted to operate. The newer one that came from the kickstarter looks good too.

I hope smoothie will eventually support full closed loop servos, as that would open it up to a whole new customer base.
Check out the Tri hotend!
arthurwolf
Prints-a-lot
Posts: 21
Joined: Sat May 24, 2014 3:36 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by arthurwolf »

Generic Default wrote:I'm still glad I bought the smoothies, there have a been a few problems with them over the year and a half I've had them but they work great most of the time. I think I'm the only person driving a lathe with smoothie, and I haven't seen any 4 axis TAIG mills with smoothies either. The stepper current of 2 amps is nice and it allows for bigger steppers with faster travel speeds.
Smoothie driving a lathe ???
Ok so now I'm very upset.
This is an official message from the Smoothie police. You are in violation of Smoothie law #1, which states that if you do something awesome with a Smoothieboard you must send me a video of it.
Please resolve this situation as soon as possible !
My email is [email protected], pictures will give you some more time to comply, but videos are mandatory ultimately.
Generic Default wrote:However, with some of the alternatives coming on the market, I think smoothie will have to adapt to keep up.
Working on Smoothie v2 ! https://plus.google.com/u/0/+ArthurWolf ... ByuzX2wmkK" onclick="window.open(this.href);return false;
Generic Default wrote:The duet boards linked on the previous pages can support up to 9 stepper motors, which is enough to operate a 5 axis CNC mill turn center with an automatic tool changer and a bar feeder/puller.
We have a re-factoring plan written up to add this to the new version of the firmware.
Generic Default wrote: Or anything else that the firmware can be adapted to operate. The newer one that came from the kickstarter looks good too.
I hope smoothie will eventually support full closed loop servos, as that would open it up to a whole new customer base.
You can use closed-loop servos if you have external drivers for them, right now. "Native" closed-loop support is something that the addition of the FPGA on the v2 board should make much easier.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by Generic Default »

Damn Arthur you have forum ninja response time syndrome. Smoothie Five-O be profilin' me.

I have videos of the CNC stuff, I might have the original files on a camera SD but I transferred all my stuff to a laptop which has been unusable for a couple of months. When I was prototyping my Tri Hotend last year, I used the CNC mill and lathe to make the parts. The lathe CNC stuff works fine but the lathe itself is not up to my standards. I'm making a cheap DIY one with live tooling, 5 axis, tool changer, bar puller ect. and I need a board that allows a ton of steppers and stuff. Trying to compete with the $500,000 machines. ;)

A lot of the CNC DIY'ers stick with parallel port mach 3 and gecko drives because it allows faster operation. I decided to go with the smoothie because it is all-in-one and allows the same capabilities (plus about 500 dollars cheaper), except for servos and closed loop stuff.
Check out the Tri hotend!
arthurwolf
Prints-a-lot
Posts: 21
Joined: Sat May 24, 2014 3:36 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by arthurwolf »

Generic Default wrote:Damn Arthur you have forum ninja response time syndrome. Smoothie Five-O be profilin' me.
There's this button that when I check it, the forum sends me an email when somebody answers. So, blame the forum :)
Generic Default wrote:I have videos of the CNC stuff, I might have the original files on a camera SD but I transferred all my stuff to a laptop which has been unusable for a couple of months. When I was prototyping my Tri Hotend last year, I used the CNC mill and lathe to make the parts. The lathe CNC stuff works fine but the lathe itself is not up to my standards. I'm making a cheap DIY one with live tooling, 5 axis, tool changer, bar puller ect. and I need a board that allows a ton of steppers and stuff. Trying to compete with the $500,000 machines. ;)
Yay for the video !

For multiple steppers, we have the obvious XYZ stuff, for something simple like a tool changer you can use the Extruder module really ( it's a hack but it'll work ). But for true 6-axis support ( ABC axes ), that's probably only coming with Smoothie2.
User avatar
626Pilot
ULTIMATE 3D JEDI
Posts: 1720
Joined: Tue May 14, 2013 12:52 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by 626Pilot »

arthurwolf wrote:Working on Smoothie v2 ! https://plus.google.com/u/0/+ArthurWolf ... ByuzX2wmkK" onclick="window.open(this.href);return false;
I have a question about this. In another thread, someone very clever pointed out that you might implement a "Z probe" equivalent by measuring the current to the motors - if they stall, it goes up. I thought I could remember something about stepper drivers having missed step detection. Do any of your drivers have this, and if so, are the requisite pins connected to anything that can read them?

If so, your controller would have built-in Z-probe capability "for free."
arthurwolf
Prints-a-lot
Posts: 21
Joined: Sat May 24, 2014 3:36 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by arthurwolf »

626Pilot wrote:
arthurwolf wrote:Working on Smoothie v2 ! https://plus.google.com/u/0/+ArthurWolf ... ByuzX2wmkK" onclick="window.open(this.href);return false;
I have a question about this. In another thread, someone very clever pointed out that you might implement a "Z probe" equivalent by measuring the current to the motors - if they stall, it goes up. I thought I could remember something about stepper drivers having missed step detection. Do any of your drivers have this, and if so, are the requisite pins connected to anything that can read them?

If so, your controller would have built-in Z-probe capability "for free."
We'll be using those drivers : http://www.trinamic.com/products/integr ... ver/tmc262 for XYZ on the "pro" board ( normal allegro drivers on the non-pro board for cost reasons ).

It has plenty of fancy features, including what you describe, and everything will be wired to the microcontroller, so this is totally something that can be implemented.

Though you have to make sure it's sensible enough and the machine is built in such a way, that it won't break something in the printer before triggering. But good idea, several people have already spontaneously suggested it, so there is a lot of interest in that :)
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by Generic Default »

That would be a nice feature to have. Not sure if it would work for my big delta because of the lead screws, but belt/cord driven printers would benefit from a stall-zeroing system.

If the detector is reliable enough you could effectively use it as a closed loop system. If it can only detect that "steps were missed" and NOT "how many steps were missed", it could still home all axes and resume printing.

Tricky stuff to program but it would be worth it for overall system reliability!
Check out the Tri hotend!
User avatar
626Pilot
ULTIMATE 3D JEDI
Posts: 1720
Joined: Tue May 14, 2013 12:52 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by 626Pilot »

Generic Default wrote:If the detector is reliable enough you could effectively use it as a closed loop system. If it can only detect that "steps were missed" and NOT "how many steps were missed", it could still home all axes and resume printing.
I guess you could see if the pins are high or low in whatever code (probably a timer interrupt) generates the steps. Other possibilities would be to have the FPGA do it (assuming the lines run there), or wire the pins into hardware interrupt lines, if enough are free. That would close the loop. Your steppers would essentially have the same guaranteed reliability as a servo, if not the resolution.
User avatar
Eaglezsoar
ULTIMATE 3D JEDI
Posts: 7185
Joined: Sun Apr 01, 2012 5:26 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by Eaglezsoar »

mhackney wrote:Brian and all, I also run a small business as you know - http://www.eclecticangler.com" onclick="window.open(this.href);return false; - so I know it can be a challenge. But not once in 6 years of operation have I ever not responded to a customer in a timely manner. In the 2 instances where I was actually out of the country for 3 days, I immediately sent a response and explained that I was out of the country. I've NEVER had a customer send me 2 or more request emails for help. My philosophy is, if you don't have the time, energy or patience to put the customer first, you probably should not be in business. I've had my spikes in sales too and it is difficult to navigate through. But, communication with customers is #1 priority. People will understand if you are up front and honest "we had a big spike in sales and it is going to take an extra X days to get your order out. As a thank you for your patience, I'm including an ABC." That's all it takes.

I think the issue here is more one of lack of communication, although 2 boards with problem capacitors is indicative of a quality control or design issue. Come clean, fix it, respond honestly and punctually and you will build a business and customer base that will support and promote you. It works.

It is good that Arthur is willing to send replacements from Europe. But that is just broken in my opinion. If he can't communicate with Uberclock and get them to send replacements, find a new supplier in the US. Or, simply buck up and communicate with customers. If getting dug out of a backlog issue is the reason, customers will understand if they are told.

That said, I have 5 smoothie boards. My first one from early 2014 had gremlins. The community and Umber Clock did support me and communicated at an appropriate level. I'm running 3 now with no issues for any of them. The other 2 will go into future printers.

Now I am really confused. A post a couple of months past this one ( this one is in august, the other post is in October)
anyway, the post in October stated that you have decided to setup your printers to all run on the Duet. What happened in two months that made you decide to get off the Smoothie and put everything on the duet.
The reason I ask is that I am going to have to purchase 4 controller boards and I want what in your opinion works the best and is the best for the money. Thanks Michael :)
User avatar
mhackney
ULTIMATE 3D JEDI
Posts: 5412
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by mhackney »

I like my smoothies and still have 2 printers setup with them. I started with Duet simply to give it a try. Now that I have it running, I love it. The built-in calibration is fast and works great. I also like that it calculates every point on the move path, not calculating short line segments like ALL the other delta firmware does. You can hear the difference in the movement but it is very difficult to observe (NOTE: STLs converted to gcode end up with line segments to create curves. Even in firmware that supports the arc command, the slicer can not use that if the source is an STL. RepRapFirmware does not have native arc support so to do comparisons you have to use very high triangle counts when creating the STL from the CAD source to get very short line segments.)

I also really like the PanelDue touch screen and the built in web browser along with the EASE of configuration and tweaking configuration. That, to me, is painful on smoothie. I'm on a mac and there are issues with mounted SD cards via usb to transfer files, I don't know where the IP based simple FTP file transfer work ended up but at the time this was not implemented. There was just a little too much friction. To be fair, flashing the firmware was dead simple - put it on the SD card and reset. On Duet, you need to run a utility (came with the Arduino SDK) that connects to Duet via USB. You click a button on the Duet to erase it and then run the utility to upload the new firmware. This is a little more challenging than smoothie's method. But, I spend way more time tweaking configuration than uploading new firmware. Once you are setup to do it, its easy the from there but the initial "flash" has a lot of friction to set up. Also, gcode uploads via the webserver are simple and fast.

On the flip side, the smoothie wiki and documentation are very well done. I've talked to the Duet developers and suggested they do something similar. They could be even more focused since they don't have to deal with other applications like laser cutters, milling machines, routers, etc like the smoothie community does.

I have a little (2x2") wifi adapter connected to the ethernet port on my Duets so I can access it from ANYWHERE. I move my printers around as I work on one or another and I don't have enough host computers to access them all. Having them available via wifi gives me more flexibility in relocating them and ability to use a single computer to manage them (when I'm not using their panels).

Its a bit expensive to move everything over to Duet and there is the old saying "if it ain't broke, don't fix it" so I will leave my smoothie systems alone for now. There is no clear winner in the delta controller world so it makes sense to keep current with several of the contenders.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler
bot
Printmaster!
Posts: 993
Joined: Thu Sep 25, 2014 12:18 am
Location: Vancouver
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by bot »

Agreed on all points. I was originally going to do a smoothie vs duet print quality test, on identical machines, but now that I have duet set up, I'm in love and have no inclination do to such. You seem to be in more of a position to do a comparison, would you be able to? I wanted to mostly compare the segmentation moire of smoothieboard prints to the results from duet. Duet still does have SOME moire, but it's not the same pattern as the segemntation moire. I think the much smaller moire on the duet is from the extruder pulses. Either way, I'd love to see some comparisons of identical g-code on identical hardware setups. :mrgreen:
*not actually a robot
User avatar
mhackney
ULTIMATE 3D JEDI
Posts: 5412
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by mhackney »

Yes, I'll do some side by sides. I developed a test object I published in the moire pattern thread earlier this summer that works well. The Duet printed version is much much cleaner.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler
User avatar
Eaglezsoar
ULTIMATE 3D JEDI
Posts: 7185
Joined: Sun Apr 01, 2012 5:26 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by Eaglezsoar »

mhackney wrote:Yes, I'll do some side by sides. I developed a test object I published in the moire pattern thread earlier this summer that works well. The Duet printed version is much much cleaner.
Michael, I have decided to purchase the duet for my next controller and may I ask you a couple of questions?
Is the color touch screen display worth it or do you feel the web server built it is what I would use the most.
Hopefully your write up will include the firmware so that us neophytes won't feel so overwhelmed.
I will let you go to prepare for those Boston Winters! :P
bot
Printmaster!
Posts: 993
Joined: Thu Sep 25, 2014 12:18 am
Location: Vancouver
Contact:

Re: Smoothieboards unreliable - what new controller to suppo

Post by bot »

I feel the touch screen is really nice, and the potential to make it even greater is there, because as Michael said, its a separate unit altogether. Huge potential there. At the same time, the web server provides all the functionality of the touch screen, and it might be easier to use because the touch screen is currently set up with tiny buttons and no room between them, can be easy to hit the wrong thing.

If you're wondering if it's worth it -- set up the web server and use that, and you can always very easily plug the touch screen in at a later date if you feel you need it.
*not actually a robot
User avatar
626Pilot
ULTIMATE 3D JEDI
Posts: 1720
Joined: Tue May 14, 2013 12:52 pm

Re: Smoothieboards unreliable - what new controller to suppo

Post by 626Pilot »

I don't get why the Duet and BBP teams both decided to write their own firmware - both based on Marlin, of all things. It seems like a pointless duplication of effort. They both could've ported Smoothie to their hardware, added whatever modifications they needed (segmentless moves, nice web interface, etc.) and been done with it in a fraction of the time.
bubbasnow
ULTIMATE 3D JEDI
Posts: 1064
Joined: Fri Aug 02, 2013 4:24 pm
Location: Dayton, WA

Re: Smoothieboards unreliable - what new controller to suppo

Post by bubbasnow »

626Pilot wrote:I don't get why the Duet and BBP teams both decided to write their own firmware - both based on Marlin, of all things. It seems like a pointless duplication of effort. They both could've ported Smoothie to their hardware, added whatever modifications they needed (segmentless moves, nice web interface, etc.) and been done with it in a fraction of the time.
"my firmware is better" said every 3d printer board manufacturer
Post Reply

Return to “The Lounge”