Welcome! Log In Create A New Profile


RAMPS 1.5??

Posted by aka47 
RAMPS 1.5??
October 05, 2012 11:37AM

I might be in the wrong place with this post as it seems to be mostly support. If this is the case I apologise in advance.

Secondly sorry for the post tittle I don't know if there is a 1.5 in planning or not. It was just using the next number onwards.


As there is an abundance of unused pins on the Arduino Mega/RAMPS combo why are we still using jumpers to select the microstepping sizes on the stepper modules ??

Is it worth considering doing away with the jumpers, desigining the board with pull up/down resistors and driving them all from spare digital io pins on the arduino. This makes microstepping settings a fimrware option and eliminates disasembly to change them.

BTW this is certainly not meant as criticism as I like the RAMPS boards being a user of one myself.

Are there any other things that would be worth seeing added into a future RAMPS incarnation ??


Edited 2 time(s). Last edit at 10/06/2012 06:41AM by aka47.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 05, 2012 07:43PM
I have never once needed to change microstep settings on any of my machines. It isn't something you should need to change in the course of normal usage, or even when hacking at the machine.

Help improve the RepRap wiki!
Just click "Edit" in the top-right corner of the page and start typing.
Anyone can edit the wiki!
Re: RAMPS 1.5??
October 06, 2012 04:05AM
I believe the same counter arguments were used during the discusion as to whether it would be beneficial to store PID values in eeprom and be able to change them via M codes. But here we are today with just this.

Ultimately it realy depends on what sort of user you are.

If you are using a specific design and or kit with all the variables tweaked to perfection for you, because you are focused solely on printing other things then I believe your argument is good.

If on the other hand you are an experimenter building your own machines which will be at best mongrels with everything to adjust your argument will apear to be flawed. As you will often need to adjust things to get the best mix. Given that many of them are a touch interdependant.

My own work is very much the second, being able to change parameters on the fly is very much desirable.

There is also an unobvious aspect.....

The arduino based equipment has a natural upper limit to the rate it can issue step pulses. Ergo a natural upper limit to the speed a machine can run at given a certain level of microstepping.

Stepper motors have time constants asociated with their physical construction and windings.

If you look at stepping waveforms and do some rough calculations re current flow and time constants for the motor windings etc. You will find that once you get above a certain speed, microstepping is pointless and a waste of processing cycles until you decellerate to a certain point.

Here then (With stepper drivers that will tolerate it) is an oportunity to claw back a bunch of procesing cycles at higher velocities.

Maybe even an oportunity to wring a bit more speed out of the machine and shorten the print time.

This is done by changing switching modes on the fly from micro-stepping to full stepping on a transferable boundary in the switching cycle. This is realy not a new technique. I am given to understand that certain commercial stepper drivers do exactly this to get higher speeds whilst retaining higher resolution at lower speeds.

Microstepping like most engineering solutions is a compromise.

Having said all the above,certain stepper driver chips initialise their modes at power on and need a reset to change mode so may not be able to do this on the fly without some clever reseting etc. (Again maybe not be doable for certain drivers due to time taken to reset being longer than the current stepping period)

However for others it is doable, but the firmware and hardware does'nt currently support it.

No one would realy bother to write firmware for hardware that does'nt exist.

So the argument is that by enabling the hardware a further stage in evolution becomes possible. That at the moment is precluded.

Edited 4 time(s). Last edit at 10/06/2012 06:45AM by aka47.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 06, 2012 05:56AM
Pololu stepper drivers allow to change microstepping on the fly. Just make sure you switch that in a motor position common to both resolutions.

As you apparently have big plans: go for it, it's open source!

Generation 7 Electronics Teacup Firmware RepRap DIY
Re: RAMPS 1.5??
October 06, 2012 06:33AM
LOL, my plans have always been big, that's why I am still actively contributing after all these years.

It is a project worth contributing to.

There are though more contributors and view points to consider than you and I.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 13, 2012 09:58PM
You guys seen the RAMBo board, from UltiMachine, creator of ramps? It has digipots for microstepping and current settings to all 5 on board drivers. It soumds like exactly what youre looking for, and firmware has the M codes to adjust it, on the fly, if wanted!
Re: RAMPS 1.5??
October 14, 2012 01:53AM
i got lost when it says motor current controlled by digipot, what does that mean? so if all things are busy working, what is a reliable woking output for the stepper driver? 1A 20v ?

johnoly99 Wrote:
> You guys seen the RAMBo board, from UltiMachine,
> creator of ramps? It has digipots for
> microstepping and current settings to all 5 on
> board drivers. It soumds like exactly what youre
> looking for, and firmware has the M codes to
> adjust it, on the fly, if wanted!
Re: RAMPS 1.5??
October 14, 2012 03:04PM
Yes you might think the RAMBO had it all, for everyone, from a quick look around.

Again though as an experimenter I want a setup that is modular so that when I trash bits of it (Which I do, and looking round the forums other experimenters do too) I dont have to replace the whole thing. The current RAMPS setup will allow me to swap any of the drivers and or the Arduino with off the shelf items. Without having to pay out for the whole thing again.

Havign said this though for folk buying a Kit machine that just want it to go and for them not to have to worry about the technicalites I think the RAMBO has it all.

The RAMPS board as it stands even the surface mount one is repairable. I guess as the ramps board itself is available relatively inexpensively I could also replace just that bit and re-use my Arduino and drivers. If I was oo lazy to fix it, or it had done to much magic smoking.

BTW a digipot is an IC that will accept digital instructions and emulates an analogue potentiometer (Variable resistor) hence the name. I guess the RAMBO board is use ing these instead of the variable resistors to set the current limit.

My steppers work fine with a 12 Volt supply and the current limiting set between half and three quarters. (1 to 1.5 amps ish)

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 21, 2012 02:30AM
I too would like an update to the RAMPS board, but I'm not at all interested in programmable microstepping.

I'm interested in rearranging the board so that it's a lot easier to connect to hardware (makes the board larger though), and also getting rid of those horrible polyfuses and replacing them with automotive blade fuses.

FWIW: I run my RAMPS at 24V with external fuses. I replaced the polyfuses with wire links because they get too hot, and I lose too much voltage across them. Add to this that the 11A polyfuse isn't even rated for 24V, and IMO you're just looking for trouble. Another downside - if you draw well in excess of the current for a polyfuse (eg: +20A thru an 11A polyfuse @ 12V), they can actually catch fire!
Re: RAMPS 1.5??
October 21, 2012 05:21AM
Yup I pretty much removed mine and put shorting links in as well.

Auto motive blade fuses is one I did'nt think of. I am not entirely sure why I would want the fusing on the board anyway. a pair of inline fuse holders with cartdridge fuses would be doable too. Where the fuseholder is, and it's type, I guess is down to personal preference.

Talking to folk, screw terminals would seem to be a popular would'nt it be nice upgrade.

The little crimps are OK for pre done cables provided by kit suppliers with the clever crimping tool but are a PITA for anyone else using thin nosed pliers.

Personaly I don't think those little terminals are realy up to the sustained current draw for heaters. Even though they are used on the stepper modules. Can't remember what the manufacturers rating is, I seem to think it was around 1 Amp. Happy to be wrong about this.

24V, intreresting. The motors coils are usualy rated at 3V ish. Did you use different steppers (6V coils) ??, How does it compare for performance with 12V ??

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 21, 2012 05:23AM
hmmm i agree blade fuses do have advantages.

ok got the part abt digipot. my feet looks dry after a few years of being out of touch with electronics lol

i think if there is a ver 1.5 ... i would suggest better stepper drivers, that handles and sinks more power. i read somewhere re 1 that does 3A? already i am clocking abt 70degC on the tiny heatsinks so .... i be trying larger sinks next week :p
Re: RAMPS 1.5??
October 21, 2012 06:13AM
I agree too about the stepper drivers and have been thinking about this one for a while.

An interesting bunch of observations about them:-

I don't know of many folk who actualy run them at full current, mine are set about midpoint so should be about 1A ish. But they still get hot enough not to touch them. Which I don't like. There may be PCB design techniques that could/should be used to help with this.

Adequate cooling is an issue and is made worse by the pot being next to the chip and taller than the stepper chip. I put cut down RAM heatsinks on mine but have no confidence in the thermal transfer capabilities of the double sided sticky pads provided for the job. Close positive pressure contact and good thermal compound should always be better. I guess lifting the heatsink by a small aluminium square or spacer might allow larger heatsinks, but makes it less likely to make good thermal contact with those useless double sided sticky pads. It also covers the pot and prevents adjusting the current limit after assembly.

If the pot was not there, there are still no mounting holes to fasten a heatsink to. Idealy a heatsink should be mechanicaly foxed and sprin loaded in some way so as to make a flat close contact with the chip to be cooled. I wonder if a useful upgrade to the ramps design would be to space them out a little and put heatsink clamping holes in the ramps board at either end of the Pollolu modules. I guess this would also mechanicaly fix them in place too.

Given that they run hot I am curious to know if they are limiting in some way at peak points in their operation. In reality there is no way we could tell. The junction temperatures are always much hotter than the case.

Having said all of the above (they are things I have never liked about them from the start) The modules are convinient, readily available and easily swapped when they go belly up. This I think explains their popularity despite their technical short comings.

Toshiba do stepper drivers that could be considered electicaly equivalent and have a higher current rating. They appear to be packaged for maximum inconvinience though. The package with best heatsinking capabilities is way too big to shoe-horn into a RAMPS configuration. Toshiba's drivers are also more expensive.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 21, 2012 06:21AM
Just thought I had winged on without realy talking about solutions.

An ideal stepper driver would do all that the Pollolu does, in the same plugin format.

But have a greater curernt limit (Can't see the need for more than 3 Amp), have serialy programable current and microstep, a larger chip surface for heatsinking and a couple of holes for mounting the heatsink on. If matching holes were placed on the RAMPS PCB these could also double up as mechanical fixing for the module.

The pins that currently select microstepping could be given over to the serial comms.

Step, enable and dir etc would remain the same.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 21, 2012 07:16PM
Standard motors. They only run slightly warm (about 40-50ish Deg C), which is not an issue in my opinion.

The extruder is the ones that annoys me the most, because my crappy old wades is PLA. By itself the PLA around the screws deforms enough that the motor will move. I mounted a fan on the back of the motor and I've had no problems with it.

Fuses on the board are good for end users who don't realise they need fuses or any sort of protection, then do something silly like short out a heater and fry some part of the board. If you provide your own fuses outside the setup, or use a decent PSU with a reasonable current limit and short circuit protection, then it's not an issue. Knowing those things requires some knowledge, which many people who have bought a RAMPS board simply do not have.

I personally like the 0.1" pitch locking molex connectors, but then again I have a crimp tool for the pins. That said, there are similar connectors with larger pitches, and you can also get screw terminals in those pitches (so you choose the connector you want).

I think the most annoying thing about the v1.4 RAMPS board is that a lot of the motor cables come out from the middle of the board, which apart from obstructing side airflow (especially in v1.4, where the pololu's are all in line and the cables actually block side airflow to the heatsink of the next pololu) make putting a fan ABOVE the board a pain. Better IMO for these cables to come out the sides, possibly with the option of using right angle connectors or screw terminals.

FWIW: You should be able to make something workable by mounting the Pololu's OVER the pins to the Arduino Mega, motor side pointing out. It's not like you need access to those pins (except for debugging perhaps), as there's no way you could actually mount another shield on top of the thing.

As for the heater/fan/heatbed outputs, on mine I simply put in the same connectors as used on the power input connectors (with the removable screw-terminal part). Had to file a bit of the connector shell away, but they work reasonably well, and dead easy to disconnect while being reliable. If at all possible, changing the position of those holes so that you don't have to file those connectors to fit them (less than 2mm) would be very useful.

Will probably have a stab at mucking around with the design at some point after this week is over.
Re: RAMPS 1.5??
October 22, 2012 07:04AM
Hmm now there's a useful idea....

--- quote ---
FWIW: You should be able to make something workable by mounting the Pololu's OVER the pins to the Arduino Mega, motor side pointing out. It's not like you need access to those pins (except for debugging perhaps), as there's no way you could actually mount another shield on top of the thing.
--- unquote ---

It is something that I had paused to think about was how to get the stepper outputs on the outside so as to use larger pitch screw terminals. o.1" pitch terminals are available but tend to be expensive for what they are. But that would certainly be a way of doing it.

Connectors for SD and LCD control panel could reay do to be reworked into dual row 0.1" pin headers. Simple because Ribbon cable and IDC ribon connectors are about the worlds easiest way of doing this. I guess it is doable to recycle PC ribbin cable and connectors for this as well. Lowering the bar to entry.

On Molex crimps, I too have a crimper and it makes life bearable. But there again I do a lot more than Reprap so it is worth me having one. Facilitating a local Reprap user group I get to see first hand the results of the inexperienced trying to make the crimp method work without the correct tooling, and it is'nt pretty. Screw terminals again lower the bar to entry and improve reliability.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 22, 2012 05:36PM
I too like the idea of IDC connectors, especially if they're edge mounted or you leave room for right-angle connectors, which gives you plenty of options for connector choices. No reason that things like the thermocouples and the min/max inputs can't be IDC connectors either, especially since you can use standard pin headers on them if you want.

The guys at our local hackerspace made an an all-in-one board with an LCD, rotrary/push encoder knob, SD, and 3 thermocouples on it, which connects to RAMPS using ribbon and an IDC on the board. I'll see if I can dig up the IDC side pinout.
Re: RAMPS 1.5??
October 24, 2012 07:21AM
Just thought of something, prompted by the latest Arduino news about the Due.

The Due board is 3.3v io not 5v. So a useful mod would be for the RAMPS board to read ioref from the Arduino and set the logic voltage/level translation to suit. (If it is neecesary).

It would be cool to have the option to plug a Ramps sheild onto a DUE at some point int he near future.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
October 29, 2012 11:05PM
aka47 Wrote:
> Just thought I had winged on without realy talking
> about solutions.
> An ideal stepper driver would do all that the
> Pollolu does, in the same plugin format.
> Step, enable and dir etc would remain the same.

I've been reading this post and others for a few days and I am thinking it is possible to design and manufacture a pololu/stepstick-compatible stepper driver.
So far, I got some chips that can handle the logic for microstepping, but they do not contain the MOSFET inside. These chips can also work with logic voltages between 3V and 5V, so that would be compatible with both Arduino MEGA old and new.
I'm still researching which MOSFET are most appropriate, but I want something that can handle at least 5A.
Although I am enthusiastic about this idea, I'm no expert in electronics, so any advice with this would be helpful.
Re: RAMPS 1.5??
October 30, 2012 06:34AM
Yes I think there would be some mileage in that approach.

I guess when you look at the pricing of say microcontrolers like avr's 168 there is an argument for rolling your own code for a microcontroler to be a stepper driver and add the power stages. There are also the PIC range. The new Xmegas have event circuitry in built that makes them much faster if used cleverly, than their native processign power.

Single sourced dedicated silicone is fine and has it's advantages, cost is not usualy one of them. Most vendors like to try and get from you a percantage of what you want to use it for, in advance of you using it. I think the arduino folk found this out with FTDI, and ended up making their own USB interface device using another microcontroler.

Other than for smoothing/noise-reduction their is not realy much need for microstepping given the resolution most machines work at. In most cases the 1/8th step is as small as you need to go. As perhaps the GenX controler boards have demsonstrated. Happy to be wrong abou this one.

There was an offshoot of the OpenServo project that went the direction of making a Stepper board. It could be worth taking a look at that. Think it was called phenom step or some such. It has been a long time since I last looked.


Other than that it is a case of experimenting using your own driver stages and an arduino.

I guess another option is using CPLD's (Complex Programmable Logic Devices) or some such (this is not an easy route as their is a big initial learning curve). They are very fast though.

On the power driver stages, you can get up to maybe 1 or 1.5 amp peak drive by repurposing powerfet driver half bridge chips.

Take for example the mic4426bn :-


Above that you are looking at dedicated silicone again and it gets expensive.

Rolling your own is always a lot cheaper but is more labour/math intensive. Have a google around the robotics sites etc for Full and Half bridge driver circuits. Particularly BEAM robotics. There are also some very good tutorials on using and understanding fet specs, which look a little wild at first.

If the full bridge circuits (You will need two for a bipoloar stepper) look a bit complicated start with Half bridge circuits then times by 4.

International Rectifier do some useful transistors and full/half bridge IC's. They are inclined to claim you need to buy a whole bunch of their stuff to go aorund anythign of theirs you are interested in though.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
November 05, 2012 07:22PM
Of note regarding going to 3.3v (in our case, for use with a Due), the Gen7 guys are looking at transition to an ARM CPU as well, which means changing to 3.3v.

In regards to the thermistor, this is interesting and worth noting:


Basically using something like a 470 or 560 560 or 820 ohm resistor in the thermistor voltage divider would definitely improve readings (even with 5v setups), though the thermistor tables would need to be regenerated (not really that hard).

Edit: 470 would push the readings too much to the high side, but dropping it lower could be useful, especially if you use the same resistor divider values for the heated bed (otherwise you sacrifice one input for the other).

Edited 1 time(s). Last edit at 11/05/2012 07:39PM by Cefiar.
Re: RAMPS 1.5??
November 06, 2012 05:17AM
Hmm interesting.

Apparently there are moves towards using the Texas Instruments DRV8825 as a replacment for the Pololu usual.

I wonder if that would make a board using it into a RAMTS board instead of RAMPS........ LOL

Or more corectly RADTS if used with a due.

I wonder if that one can change modes on the fly. Having had a look at the datasheet it actualy has an output that would generate a signal when the stepping cycle was at it's Home point.

ie the place where you can cleanly change modes without getting part cycles that cause skip in the stepper spindle.

My understanding fo the datasheets is that this is basicaly the full step boundary that all the microstepping modes share.

That is if it can change modes quick enough to be useful.

Some devices have a settling time on reset or mode change that gets in the way.

The DRV8825 also looks to be useable at 3.3 logic and logic supply as well. To be sure we would have to have a mess with one to be sure though.

It is almost tempting to discard backwards compatability at the steper module and arduino levels and do something quite novel under the new name RADTS...

Programable stepper modules (Current, microstep modes, decay etc...) now that's an idea.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
November 06, 2012 06:52AM
You can change microstepping mode at any step angle common to both modes. It's also pretty much trivial to do this in software, just connect MS1, MS2 and MS3 to microcontroller pins (you'd need 12 of them).

IMHO, the challenge is to do the calculation on _when_ to switch modes. This would make acceleration calculations quite a bit more complex and this calculation is the speed limiting bottleneck already.

I'm pretty convinced acceleration calculations can be reduced to a 16 bit calculation instead of the current 32 bit one. This could speed up things a lot. The task would be to split up each movement into an acceleration, constant speed and deceleration part. Acceleration and deceleration have never more than 65535 steps, so they can be done in 16 bits. Constant speed movements need no integer divide, so dealing with 32 bits doesn't hurt.

I've attached a number of files. They apply to Teacup's current Gen7 branch and do this split. Put a #define ACCELERATION_SPLIT into your config.h and enjoy!

Generation 7 Electronics Teacup Firmware RepRap DIY
open | download - 0001-ACCELERATION_SPLIT-get-started.patch (2.4 KB)
open | download - 0002-ACCELERATION_SPLIT-make-sure-the-queue-has-always-en.patch (1.1 KB)
open | download - 0003-ACCELERATION_SPLIT-for-a-start-add-the-general-case.patch (4.4 KB)
Re: RAMPS 1.5??
November 06, 2012 07:38AM
On switching stepping modes, my understanding is as follows:-

The decision on the need to change mode is governed solely by the frequency of steps being issued.

ie if above frequency X (dependant on motor, so will be a setable parameter) we change down a mode, bit like changing gears in a car. if below frequency X we change up a mode.

The decision as as to when to action the change is on a common step mode boundary (ie the step cycle home).

When droping the resolution, we are fortunate in that i think they divisors are nice binary units ie devide is an easy and quick shift.

So the above implies that a quick and dirty method is for the motion planner to assume the high resolution has'nt changed (ie is completely unaware of the mode change) but the bit of code that wiggles the step pin does the mode change, and divides the incoming requests to pin wiggle. Keeping the motion planing as is, but adding smarts to the pin wiggling.

Stepper chips which can tell you exactly when the oportunity for mode change is (ie stepping cycle home), can make this a bunch easier and remove the possibility of everythign gettign out of step.

Granted, the above does'nt save as many processing cycles as hacking the motion planner would, but does have the advantage of driving the motor harder at higher speeds making it less likely to miss steps and able to atain higher speeds. I think the stepper chip should disipate less as well.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
November 06, 2012 08:03AM

Thanks for the pointer to the Gen7 research traumflug did. I pretty much concur with all his findings my understanding has been mostly similar. the biggest surprise was the thermistor work. Well done though to traumflug.

Adding that robust research back into this discusion then a couple of points stand out.

1. Yes a rethink on the thermistor circuit is a good idea.

2. The situation when driving mosfets etc at 3.3v is perhaps going to be made worse at 3.3v than it is for 5v. The solution for this is perhaps to use more apropriately sized fet switches together with a fet driver chip (half bridge). Where the drive voltage is the motor voltage. A fet driver could of course be assembled out of discreets and may be cheaper. Dual driver IC's though are reasonably common. They certainly work better witht eh gate capacitances and protect the microcontroler. Gate resistors as observed are very much a compromise that sort of gives you the worst of both.

Necessity hopefully becomes the absentee parent of successfully invented children.
Re: RAMPS 1.5??
February 07, 2018 06:58PM
Anyone have a good schematic of Ramps 1.5 wondering what some of the pins are, J5? maybe a few others.
Re: RAMPS 1.5??
February 07, 2018 10:09PM
"schematic" ?

Re: RAMPS 1.5??
February 09, 2018 11:57PM
smiling smiley very few pics or info... so far I have everything working, except E0 which is very near those J5's but it cant be a problem as i couldnt connect motor wrong to them, so thinking of trying to use E1 instead, but it was other pins like the 12v missing so i wondered what those 3 pins j5 where for...I cant read the board dont have a good light magnifier.

just bought a few £3 ramps....(perhaps asking for trouble) one of these days i'll go 32bit, almost got a duet to try the polar printer or mill but decided to buy a ton of other stuff instead...but one day.

Edited 1 time(s). Last edit at 02/09/2018 11:58PM by MechaBits.
Re: RAMPS 1.5??
February 10, 2018 02:12AM
ramps 1.5 is not open source or even official, its just some Chinese supplier trying to one up on the ramps 1.4

No documentation... typical lack of support and documentation.
Re: RAMPS 1.5??
February 10, 2018 01:00PM
I was seduced by the flat mosfets tidier look, didnt know it had bigtreetech on the back which looks a bit shoddy, probably would have tried the BIQU one, but thats probably not official either?
Would like the one with the blade fuses or a pro one but as pay much more seems pointless if you get rearm or go mks, also got a few cnc shields as they where buttons...not sure i'll get around to using them.

If seller cant explain why extruder doesnt work, or provide a wire diagram to help....paypal will sort it smiling smiley

Just discovered I bought 2...So have another to test.

Edited 1 time(s). Last edit at 02/12/2018 02:08AM by MechaBits.
Sorry, only registered users may post in this forum.

Click here to login