Welcome! Log In Create A New Profile

Advanced

High-end control boards: worth it?

Posted by anvoice 
High-end control boards: worth it?
April 27, 2015 06:43AM
Noticing the large quantity of the new and expensive 3d printer boards on the market today. Someone even told me that the RAMPS is "outdated", so I began wondering what exactly that means here. I'm sure they have a bunch of bells and whistles to justify the price, possibly some extra protection and sometimes direct 24V support with no further tinkering, digital current setting for drivers but the basic functionality seems to be the same as, say, the very basic RAMPS. Considering that I can get a RAMPS 1.4 with drivers, along with an arduino for it for something like $25 total, is it worth paying the extra $160 or so for a RAMBo 1.3 for example?

With a bit of soldering and a cheap 12V power source, almost any RAMPS/arduino combo can take 24V, dual extrusion is also relatively painless. It's true that the Polulu pots can be a pain, but once you've got the current set once you can forget about it forever unless you mod the printer. Basically with some rather easy tinkering, it seems like the RAMPS can do about everything its more advanced counterparts can. So is it worth the extra price to upgrade, or is it just the cool factor (granted the standalone mobos look pretty good)?

P.S. One thing I definitely agree is wrong with the RAMPS is the heater bed input screw terminal: that thing is not rated for the 11A it's supposed to carry, and it's probably safer to solder on the power wires from the start to avoid a possible fire hazard. Unpleasant, but fixable if known ahead of time.
Re: High-end control boards: worth it?
April 27, 2015 07:08AM
Certain printer types (Delta and CoreXY) push 8-bit processors (RAMPS, RAMBO, Megatronics, etc) to their redline and beyond doing complex math. Add a graphical LCD, and they can't keep up. In those cases, it makes sense to upgrade to a 32-bit offering, such as Smoothieboard and compatibles, Duet, or (gasp) even one of the DUE-compatible shields like the RADDS (but stay away from RAMPS-FD as it is defective/dangerous out of the box).....
Re: High-end control boards: worth it?
April 27, 2015 07:25AM
The smoothieboard is worth it for the ease of configuring the printer alone. ALL variables are in a single config.txt file on the on-board uSD card. The config.txt file is read by the board each time it boots up, so you make your changes and reboot the board and you're done. There's no hunting through multiple files to find the variables, no compiling the firmware, no having to deal with Arduino compatibility issues. Updating the firmware is also done by simply putting the new firmware file on the uSD card. When the board boots it will see the new firmware and flash it.

A lot of people seem to like the idea of remotely starting prints- smoothie has built in networking (not wireless) and a web server.
Re: High-end control boards: worth it?
April 27, 2015 07:26AM
There are a number of disadvantages to RAMPS besides the inadequate connectors and heated bed mosfets on some boards. The performance is poor compared to modern 32-bit boards, and while it is adequate for a Cartesian printer, driving a delta from Arduino/RAMPS is pushing the processor beyond the limits of what it can do well. The USB-over-serial connection is slow and doesn't haver any flow control. It doesn't have an SD card socket, but printing from SD card is more or less essential because of the slow USB connection. The 5V regulator on the Arduino is fragile, runs hot if you draw even a little current from the 5V rail (e.g. to power an LCD) or if you turn the supply voltage up to 14V (too increase heated bed power).

I regard Arduino/RAMPS as outdated, because it is technically deficient in these respects, and there are many technically better solutions available. Against that, you can get Arduino Mega/RAMPS clones extremely cheaply. The plug-in drivers offer some advantages over the integrated boards: if you blow one you can replace just that driver, and you can experiment with different drivers.#

The other big difference is software. Marlin is packed with features, but showing its age. The modern 32-bit boards come with firmware that is easier to configure. Instead of changing the firmware source code, rebuilding and re-uploading, you just edit a configuration file on the SD card. If you want to support a different type of thermistor, you don't need to construct a table for it, you just tell the firmware the R25 and B values.

There is no way I would spend $160 on a RAMBo because it shares many of the same issues as the RAMPS, and for the same or loess you can get a 32-bit board. These are the boards I suggest:

1. Duet: GBP65 + tax from Think3DPrint3D, or $80 from Replikeo + shipping + import tax if applicable. 32 bit processor, 96K RAM, 4 built-in drivers, digital stepper current control, native USB port, SDHC card socket, and Ethernet. The firmware for it supports an excellent web interface, and after using it you will never want to use a USB host control program again. Also supports the PanelDue colour touch screen control panel. Does not currently support the low-cost LCD control panels that most other boards do. You can get expansion boards to drive additional extruders, but they are not cheap.

2. Smoothieboard: better known that the Duet, but more expensive. Similar spec but less well developed web interface. Available in a 5-driver version, so cost-effective for dual-extrusion machines.

3. AZSMZ Mini. Based on the Smoothieboard design, but no Ethernet port, and plug-in drivers (so no digital control of stepper motor current). Costs less than the Duet, but you need to add the drivers.

It's also possible to get RAMPS-like boards for Arduino Due, but then you don't get Ethernet or digital stepper current control, and you don't get to break free from Marlin.



Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: High-end control boards: worth it?
April 27, 2015 07:07PM
So basically convenience: easily add LCD, SD-printing, wifi, multiple extruders, no firmware reloading, etc. Or if you have a non-Cartesian printer. Which is nice, but if you have a bit of engineering skill, some patience, and not an extra ~$50-100 for those boards it's not critical. I'm thinking that for the moment, the high cost of these boards is making them prohibitive for use with some simple and affordable Reprap machines.

There are also some drawbacks that I see to the smoothieboard for example: the integrated drivers are rated for something like 2A, which is enough for 3d printers, but is underpowered for something like a DIY CNC router/mill of any reasonable size. So it's feature-packed but expensive for a Reprap, and underpowered for some other CNC applications.
Re: High-end control boards: worth it?
April 27, 2015 07:30PM
Most CNC machines are going to need more powerful motors than the smoothie or any other integrated board can handle. You typically use higher torque (higher current) motors, DSP based stepper drivers, and higher voltage supplies for CNC work, and it's easy to connect to smoothie because they bring the step/direction/enable pins out to 4 pin headers. You can connect most stepper drivers without any additional circuits, just a 4 wire cable. The 2A drivers on the smoothieboard can drive a lot of NEMA-23 motors directly, and they are well heatsinked and current is set by an option in the config.txt file, so there are no tiny pots to break.
Re: High-end control boards: worth it?
April 27, 2015 10:14PM
The stronger Nema 23's often go above 2A (3A common for 425 oz-in), and something like a geckodrive g540 (almost a gold-standard for CNC mills) provides 3.5A without an additional breakout board, making the Smoothieboard unnecessary. It also can't function as a variable frequency driver, so it's basically an unnecessary link. Plus some possible safety concerns (e.g. [smoothieware.org]) when used with heavier CNC machines

All in all, seems like a turtle/rabbit hybrid that's not quite as fast as a rabbit and not as sturdy as a turtle. Probably good on a high-end 3d printer or a generously built custom printer, but I can't think of another place I'd put it.

Edited 1 time(s). Last edit at 04/27/2015 10:14PM by anvoice.
Re: High-end control boards: worth it?
April 30, 2015 09:06AM
I don't know about the other boards, but the ability to print over bluetooth is a great advantage for me on RAMPS.
Re: High-end control boards: worth it?
May 08, 2015 03:48PM
Quote
eci22
I don't know about the other boards, but the ability to print over bluetooth is a great advantage for me on RAMPS.

That's not specific to the RAMPS, you can essentially do that with any board.

Somebody documented one of the adapters here : [smoothieware.org] for use with Smoothie.

But there are also adapters that will just be plug and play.

All you need is a board with the serial port broken out, and most boards have that.
Re: High-end control boards: worth it?
May 08, 2015 03:53PM
Quote
anvoice
The stronger Nema 23's often go above 2A (3A common for 425 oz-in), and something like a geckodrive g540 (almost a gold-standard for CNC mills) provides 3.5A without an additional breakout board, making the Smoothieboard unnecessary. It also can't function as a variable frequency driver, so it's basically an unnecessary link. Plus some possible safety concerns (e.g. [smoothieware.org]) when used with heavier CNC machines

All in all, seems like a turtle/rabbit hybrid that's not quite as fast as a rabbit and not as sturdy as a turtle. Probably good on a high-end 3d printer or a generously built custom printer, but I can't think of another place I'd put it.

The AZSMZ Mini, and the budget 32-bit board I have designed, support plug-in drivers just like the RAMPS. So with these boards, you can choose to whatever drivers you want.



Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: High-end control boards: worth it?
May 12, 2015 08:42AM
Quote
arthurwolf
Quote
eci22
I don't know about the other boards, but the ability to print over bluetooth is a great advantage for me on RAMPS.

That's not specific to the RAMPS, you can essentially do that with any board.

Somebody documented one of the adapters here : [smoothieware.org] for use with Smoothie.

But there are also adapters that will just be plug and play.

All you need is a board with the serial port broken out, and most boards have that.

It's not the serial port that's broken out, they are using the soft serial library to make generic pins take on the role of TX/RX. If you connect directly to the TX/RX (pins 0 and 1 on Arduino) you wont be able to send over USB.

I'm sure you are right and its possible on alternative boards, like I said I haven't used other boards- but you do have to enable BT in firmware- I haven't studied the marlin code for what its doing but It doesn't seem as simple as adding a BT module to the serial pins
Re: High-end control boards: worth it?
May 14, 2015 01:10AM
I have to say that reading these types of discussions is an amazing learning experience. Some of this went right over my head but slowly starting to get it. Upgrading components is both terrifying and exciting. I can't wait. Thanks for your knowledge.
Re: High-end control boards: worth it?
May 14, 2015 11:25AM
If your on a tight budget, then No. Ramps/Mega will do what you need it to do, and the savings used for better quality parts elsewhere. If you can afford it, then you gain easier setup, faster, extra abilities, more room for growth...

I'm a pretty hard Ramps/Mega fan, but I'm willing to admit that the 32bit stuff or better is our future. Hopefully dc42 makes some progress on his cheap 32bit board.

Edited 1 time(s). Last edit at 05/14/2015 11:25AM by madmike8.
Sorry, only registered users may post in this forum.

Click here to login