It's time for RAMPS-FD v2...

Hello everyone

I'm gearing up for another round of development. It's time to bury RAMPS-FD v1, and find a way to get FD v2 out there.

The No 1 complaint about v1 is the safety aspects, and I believe v2 fixes those problems. v2 introduces a few other things like onboard EEPROM, but is otherwise much the same.

There was a plan to implement thermistor protection, but it needs a more sophisticated circuit than I can design, and the circuit that was implemented messed up thermistor readings, so I have removed that part.

I'll be finalising a prototype design soon, and will get some PCBs made. I will also look at getting a small batch assembled.

I have been out of the loop for quite a while, so if there are any other features/bugs you want fixed, let me know.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Can we perhaps consider renaming it to something else? I think Geeetech has really burned people on the RAMPS-FD name.

i have always been interested in the RAMPS-FD but all this with Geeetech and the mistakes has gotten me away from it.
And i agree with Cefiar becuase of Geeetech. But i will look forward to se it be revived

I think a fresh name would be a good idea, since a new revision number is easy to overlook. The only idea I had so far is "RAMPS 2016". It's probably not widely known, but RAMPS-FD should also work with a Mega2560.

I'm not sure if the situation with Geeetech could be avoided again - if it is OSH then we can't dictate who can make it. I think nearly all the actual problems I have seen reported with FDv1 are due to bad manufacturing quality. I purchases one to examine, and it is shockingly bad. One of the tracks on the PCB was etched through, inferior components substituted, terrible soldering quality.

Perhaps an effort to persuade Geeetech to sell a new version could be made. I imagine their setup cost is fairly small, once they have exhausted stock of v1, I can't see why they wouldn't sell it.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

If you can make sure it works on a mega2560, why not just call it ramps2.0, its long over due

Edited 1 time(s). Last edit at 04/20/2016 04:30AM by Dust.

Quote
Dust
If you can make sure it works on a mega2560, why not just call it ramps2.0, its long over due

That's a good idea! The latest RAMPS version on the wiki is numbered 1.4.2, it has a few improvements over 1.4. It might be hard to displace 1.4 - just looking at the prices on ebay, lowest price is about $5 including shipping. But it's worth trying!

The main problem I found with RAMPS-FD and Mega2560 is mechanical - the USB connector shorts the MOSFET pins. On RAMPS 1.4 the PCB is left clear in that area. The clearance is also not great on Due, the power connector causes one corner of the FD to not sit properly.

I've put a revised schematic for FD v2.B on github [github.com]

A list of changes over v1:
- improve PCB layout around endstop connector
- Add eeprom
- All MOSFET outputs are non-inverting
- fix problems with outputs inadvertently going high
- add a second connector for Z axis
- improve thermistor readings with better analog ground
- add header for 5V
- up rate the crowbar diode to 10A so that it has a chance to blow the fuse

and possibly a few I forgot. I think this covers all the problems people have reported, except for manufacturing deficiencies outside my control.

As ever, a list of potential changes is quite long, and board space is hard to find. Although, controller boards of 150x100mm are now common, so perhaps the size could be extended, which would help the layout a lot.

Some things to consider:
- 12/24V switching regulator
- onboard 5V regulator
- RC filter for endstop switches
- use 0603 or even 0402 components
- current control using Digipots? Needs "enhanced" stepper drivers
- microstep selection by software?

Despite being "unsafe", the RAMPS-FD beta seems to have been surprisingly popular FWIW, I have been using a self-built FDv1 for over a year without problems, but as a policy I never leave printers unattended.

Now that there are several ARM based controller boards, including cheap Chinese clones, the market for 32 bit controllers is quite well served, but I think that there is still a niche for a RAMPS type board, with familiarity of RAMPS plus separate stepper drivers, which some people prefer.

While I'm about it, I'll also update the project to KiCad v4 and give it a new repository.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Some thoughts:

1. Maybe we should review the thing from scratch and re-layout everything? There were issues with current loops in the ground traces on V1 that caused problems, and while vastly improved I'm not 100% sure they're gone in the existing V2. The thermistor and end-stops were real bug-bears, and the traces from the Arduino to the related inputs in many cases were quite long (eg: thermistors, since they have to go to the Analog inputs). Will require a new pins.h but that's manageable.

2. Maybe keep initial board layout dev closed (or not publicly announced) till you've actually constructed and tested the thing, then open it up publicly? ie: Run a private (or not actively public) repo for the board layout (schematic could stay public) till protos are tested and you're happy with it. The big issue seems to be due to companies jumping early on free designs before they were ready (and while they had various bugs, like the inverted heater logic on V1 and thermistor ground loops) and using sub-par or wrong components. Delaying the openness for board layout avoids at least the first part of this issue. I know this is a pain, but till it's well down the debug track, it might be the only way to avoid this sort of thing happening again. Note that I still FULLY support OSH, and the only reason I suggest this path is due to the issues with companies jumping the gun. Once they get stock of something, they'll probably keep hammering it out till they run out of boards. That number could be in the hundreds for all we know.

3. For input protection, it might be simpler to use MOV's. [www.digikey.com] seems suitable for 3.3V inputs, while [www.digikey.com] seems suitable for 5V inputs. Only requires a simple 0603 pad across whatever input you want to protect, and should not interfere with the input signal (and can be left unpopulated if it's an issue). Datasheet: [www.littelfuse.com]

FWIW: Given their size and cost, very tempted to buy a few. Might hack/solder them across various inputs on all my existing controller boards just for the fun of it.

Only issue with calling it RAMPS2 is that RAMPS was developed by Johnny Russell. IMO the name sort of belongs to him.

That said, we don't need to keep the M in there (since we're not tied to the Mega anymore) so why not some sort of acronym that drops the M? ARPS?

Edited 1 time(s). Last edit at 04/20/2016 03:42PM by Cefiar.

1) That is tempting.I did some work on Decapede, having a larger board certainly helped to get a tidy layout for 10 steppers, 10 mosfets, a 60A fuse, and also segregating logic and ADC inputs. [github.com] It is a big board though, and the Due is a bit lost underneath. I had some boards made in 2oz copper, but never got round to populate one. I would like to revisit that project sometime. Anyway, the layout of the Decapede worked really well, except that the stepper connectors sit in the middle of the board.

2) I think I am with you on that. My theory was that to encourage interest and collaboration, to be as open as possible. Obviously that backfired! Funny thing was I went from thinking "how the heck am I going to get this board made?" to "How do I stop this board being made?". I think the best idea is to keep a low profile in development and make it very clear when the board is considered released.

I am not sure how many of the problems were down to very poor manufacturing, but there is a limit to how much it can be improved, given that the Due does not have a separate analog ground, but if the analog and logic could be moved to a separated zone that would help a lot.

3) I looked at our circuit for the motor drives we make at work, they are protected not only against ESD up to 8KV but also misapplying 24V... The analog inputs are protected by an extensive scheme including MOVs and op-amp buffers, I can manage MOVs but analog stuff like opamps is beyond my skills. I note that Duet has a simple thermistor input circuit, so I wonder how necessary it is. I am sure that on RAMPS because nearly everything is a 0.1 header it is very easy to misplug connectors. At work we follow "poka-yoke", which basically means having different connectors so you can never plug in the wrong one.

On the name, RAMPS is almost like a generic name now and has instant recognition, but I think we need a suffix to distinguish it. I was toying with something like RAMPS-NG.

I've been looking at 32 bit Marlin again, some of the devs there are keen to get a 32 bit version fully supported. The impression I get is that people generally like the idea and low price of RAMPS-FD (at $30, who wouldn't), and the fact it uses a stock Due (they like the Arduino IDE) but don't like the bugs.

So my main goal really is to get a bug fixed version of FD, and leave the serious board development to others - there is Duet-NG and a new ARM board from Ultimachine in the works, I hear.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

The issue with the thermistor inputs comes from the fact that the hot end has a voltage greater than the thermistor can get. It's easy to get a loose thermistor wire shorted to a hot end wire, blowing your input. I've seen it countless times and even in a recent 3D Printing Today podcast (ep #133) one of the guys mentioned it happening to him.

While it can happen with end stops, it's less likely. You're more likely to short the end stop +ve to ground, taking out your regulator.

I had a look at a Smart Diode Controller chip (to drive an N channel FET instead of the crowbar diode) but they're like $3-3.50 for the chip, let alone the FET and the charge cap (which keeps the FET switched on when the supply is reversed).

As for name, RAMPS-NG does make it sufficiently different IMO.

I found some variants of FDv2 being sold on OSHPark. This is a surprise for a couple of reasons, 1) v2 is a beta and was not released either, 2) they appear to have violated the Open Source license, as they haven't published their modified source. I also recently found some of my source code has been literally stolen - they removed my copyright notice and put their name on. And these people are developers on Github, who really should know better. I expected this sort of thing from Chinese cloners, but not fellow developers.

There are also people influential in the 3d printing community spreading FUD about RAMPS-FD, based on some problems that "might have been" to do with RAMPS-FD, but they weren't sure.

It seems at no time did people think to consult the designer about suspected problems, help to fix the problem by raising an issue in github, or at least give the designer a chance to respond. It's this sort of bullshit that makes me seriously wonder if I should continue with OS or just do what others do - take whatever I can get for free and give nothing back.

But anyway, I didn't really come here to rant, I meant to talk about v2

It looks like RAMPS-NG is trending so that name might stick. I also thought to add a micro-SD socket if there is space.

I've been playing with a new larger layout, and it looks promising. I can fill the keepout zone over the USB connector with surface mount fuses, and also use surface mount MOSFETs. I might have something to show over the weekend.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Here is a render of "RAMPS-FD3" from OSHPark which I tracked down on github. It seems that people copy projects on github instead of forking, which makes it harder to follow the evolution. I'm sure people aren't being malicious, just a bit careless and maybe ignorant of the GPL, but if we don't follow our own rules, it's a bit hypocritical to point a finger Eastwards and say "you guys aren't following the rules!".



It looks like a good job, he's got onboard Trinamic drivers, more SMT components, etc. Board size is 160x130 mm.

I'd like to introduce stepper current control. I have seen a couple of "stepstick" drivers that have an extra VREF pin, unfortunately in different places. One is SureStepr by Panucatt [www.panucatt.com], the other is SilentStepStick by Waterott [github.com]

I think it would be possible to accommodate both types, but I was wondering if there are any other stepstick design with an external VREF?

Edited 1 time(s). Last edit at 04/21/2016 04:55PM by bobc.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

New directions...

After a review of the "state of the art" and a rethink, I've decided that it is not worthwhile to reinvigorate RAMPS-FD. There are many other 32 bit options now, and RADDS is a very good alternative to RAMPS-FD. The Smoothie ecosystem now seems to be a well accepted alternative to AVR based solutions as well.

The BeagleBoneBlack plus printer cape is a very capable system, and not much more expensive than Due, especially if you include the cost of a Pi.

I will probably do a little rework on RAMPS-FD v2 to fix some bugs - I have the feeling of "unfinished business", and maybe flesh out a concept for a "high-end" RAMPS, but to do it as a background activity. I have an idea for a more general purpose board which can be adapted to different processor base boards.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

I have made a few more improvements, which I'm calling v2.2:



I've added a 5V regulator, reduced some component sizes. Size expanded to 100x100mm, and layout improved to provide better access and labelling around jumpers and connectors. Connector and jumper pin functions should be same as before.
The board is slightly more tricky to assemble by hand, 0603 and TSSOP require a little more care, but are still quite doable with or without reflow.

Next time I order some boards I will get some made, but I've no immediate plan to progress it beyond prototype stage.

In conjunction, I've been looking at a "no-compromise" concept design I will call RAMPS-EV - EV for evolution. It was suggested that RAMPS-NG may be confused with DuetNG, and I've no wish to tread on other people's toes.

Board size is extended to 150x100mm and all surface mount except connectors.
Software microstep selection, digipot VREF current control, FAULT monitoring (designed for Panucatt drivers)
Power stuff is at top and right, ADC and digital IO is at center and bottom.
There is a little spare space for extra stuff.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Oh hey, that's my variant up there!

Sorry about that - you called it though; mainly just ignorance on my part. I've at least added the GPL text to my repo and I just now deleted my repo and resubmitted as a fork. Tbh I never intended mine to be anything more than a variant implementing the features I wanted that may or may not be desired by others (like the kind of costly Tmc drivers and lcd board integration) but after I got really into it I figured I'd share it in case someone else liked the design and overlooked a few things.

Anyway I'm really digging your "EV" design. The layout is far more graceful than mine (which is fairly chaotic tbh). If there's anything I can do to help feel free to let me know here or via github.

Edit: forgot to mention; I recently got my protoboards and actually populated them the other day. I'm going to test them soon; just have to find the time (super busy for the next few weeks!)

Edited 1 time(s). Last edit at 05/14/2016 09:49AM by quixotic120.

Would like to assemble and try RAMPS-FD V2.
Where is the best place to get a pcb board ?

Thank You

i just finished putting in a quote to etch some Ramps-FD V2.1 boards (i have a 1.0A i want to replace to keep using my 12864 screen) then i saw this thread... bobc did i miss this v2.2 on your github?
i like the look of the the ev model you posted, but i don't see a way to hookup my 12864 display.

I am still doing some finishing on v2.2, I redid the routing completely, I might be able to publish something soon.

For the EV board, I hadn't quite round to the display interface, I was planning to implement something with a single connector dedicated to a display, possibly similar to RADDS.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Hey,

Hope you are making progress on this last fix-up. I have been wanting to make this for some time.

I had the problem that the github files can't be read in eagle, do you have any suggestions on how to get about that?

Yes, I was just doing some work recently, I think I am nearly ready to get some boards made.

Regarding github, all my projects are in KiCad.

Arduino fans may have noticed the Star Otto, recently announced by the "other" Arduino. There are some question marks about it, and price unknown, but it has a nippy Cortex M4 processor and wifi using ESP8266, which has piqued my interest. [www.arduino.org]

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

i hope that is as pin compatible with the due as it looks as i ripped the programming port off of my due, and don't want to buy a new one.

Porting a reprap firmware to a new cpu is more than just a matching pinout. Timers and interrupts may be different to handle.

Yes I understand that but this thread is talking about the hardware being designed by bobc, if it's pin compatible with the new arduino then it's only a matter of software as the hardware will already be there.
Edit: I should clarify the due has a native port that I can still program with I'm in no rush to replace so if it takes a while to get a firmware port I have a working board still

Edited 1 time(s). Last edit at 07/08/2016 02:22PM by timmatthews1.

I haven't seen a schematic for the Otto, but as far as I can tell it follows the Due pin functions quite well. The same can not be said for another STM32 board [www.seeedstudio.com], they have screwed up the pinout by replacing signal pins with power pins, really dumb. Otherwise the board is quite cheap, and I think the 3 pins affected aren't too important for RAMPS-FD.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

For the Arduino STAR - OTTO I have phone with Mr. "repetier".

We wil try to adapt the Repetier-Firmware (Hall + Pins maybe GUI for the Star LCD) for RADDS / RAMPS-FD ;-).

[www.arduino.org]
[www.arduino.org]

I give you Information for this.

@Bobc.
Can I send you some new Stepperdrivers for testing it with RAMPS-FD ?
(we have the Raps128 (THB6128) and now we -> (willi) has developed new one with new Stepperdriver IC,
who maybe is finished in the middle of August 2016).
If Yes, please PM me.
It would make me very proud, if you like it.

Thank you
Angelo

---

Mein Club: [hackerspace-ffm.de]
RADDS-Shield -> Commercial [max3dshop.org]

For anyone interested I have put a preliminary v2.2 on github : [github.com]

Please note : this is not a production release!



Edited 2 time(s). Last edit at 07/19/2016 02:27PM by bobc.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Nice work!

Is it possible to extend the MS1-3 with SPI? There are new driver in the market called TMC2130 silent stepstick. You can find the schematics etc. on GitHub. Search there "Watterott" a German open source hardware shop. The driver are similar to TMC2100 but has some great features like coolstep, interpolation to 1/256 u-steps, and up to 1/256 native u-steps.

---

Triffid Hunter's Calibration Guide

--> X <-- Drill for new Monitor

Most important Gcode.

Quote
bobc
For anyone interested I have put a preliminary v2.2 on github : [github.com]

Is this for the existing Arduino Due or were you able to confirm this would work with the new Arduino STAR - OTTO?

Quote
PDBeal

Quote
bobc
For anyone interested I have put a preliminary v2.2 on github : [github.com]

Is this for the existing Arduino Due or were you able to confirm this would work with the new Arduino STAR - OTTO?

I can't tell yet, Arduino.org haven't published any schematics or other details, I've only got the web pictures to go on.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Quote
Wurstnase
Nice work!

Is it possible to extend the MS1-3 with SPI? There are new driver in the market called TMC2130 silent stepstick. You can find the schematics etc. on GitHub. Search there "Watterott" a German open source hardware shop. The driver are similar to TMC2100 but has some great features like coolstep, interpolation to 1/256 u-steps, and up to 1/256 native u-steps.

It's an interesting idea, but would take too many layout changes at this stage. Might be something for v2.3.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

@bobc Great to see this under development again, been holding off buying a RAADS board which seem difficult and pricey to get hold off anyway.

---

[www.coder-tronics.com]
[www.thingiverse.com]