How to: Re-ARM and RAMPS 1.6+ plus with TMC2130 drivers with full graphics LCD
ฝัง
- เผยแพร่เมื่อ 6 ม.ค. 2019
- How to (kinda) run a Re-Arm board with RRD graphical LCD panel with Bigtreetech tmc2130 drivers in smart control mode via SW SPI on a RAMPS 1.6+. "Seems to be working" (tm), but with a few graphical glitches. If you run into any issues, let me know. Hopefully I will come up with better solutions as I delve more into the software.
TIME INDEX:
1:50 background info
6:15 details of problem
7:38 How-to step by step
My contribution links:
MONTHLY SUBSCRIPTION:
/ alexkenis
ONE-TIME CONTRIBUTION:
PayPal.me/AlexKenis
AFFILIATE LINK:
www.amazon.com/?tag=alexkenis...
Background music by my band Solamors, you can contribute by buying a copy on bandcamp.com, or iTunes, Google Play, etc etc:
solamors.bandcamp.com
aletheian.bandcamp.com
PRODUCTS and blog:
AlexKenis.com (redirects to my blog at the moment) - วิทยาศาสตร์และเทคโนโลยี
Thanks for the overview. Look forward to the next videos.
Again an impressive video tut, fast‘n‘nerdy...thanks for that
Impressive totorial 👍😄
Thanks for sharing 👍😀
nice tutorial man :)
Hi Alex, thanks for this video - was waiting for it since re-arm should arrive next week, and 1.6+ is on its way to me too. Two questions though:
1. Since they're making special stepstick version for 1.6+, does that mean that others are not compatible? (like, are the MISO/MOSI/SCK pins in different places?)
2. If we forgo display, can we just run all of that stuff in hardware SPI? (if yes, can't we just use softSPI for display?) I'm asking this cause i got a ton of issues with TMC2130 in softSPI mode (shifted layers and stuff), and display is not that critical for me... (running it with octoprint)
Thanks for great material (nerdy is good).
The Marlin 2 sanity check only prevents hardware SPI to be used on Trinamic drivers and the full graphic LCD so you are definitely free to remove the LCD and go with hardware SPI. I've not tested this yet as I'm just building the printer it's going in and want the LCD for convenience.
Yes, the check is just to make that particular LCD play nice with those drivers. The issue is the LCD driver chip, so if you either don't use that LCD or run the 2130s separately, you are free to use whatever mode works best for you.
Nice!
Heyo thank you very much for this! now I'm having a hard time deciding on this combo or the recently released 32 bit SKR 1.1 (just a Gen L but with 32 bit CPU) or Sgen (Sbase really but with interchangeable drivers). Would you be possibly buying this and trying them out and comparing with this?
They way I do things here (uber-nerdy) I can't really get in too much depth with closed source motherboards. There is just too much guesswork involved. Even the reArm is tough and i have all the source files. I have seen a few threads about the MKS stuff, and they are pretty much trial-and-error until they figure it out, but you could check on Marlin's github to see if there is support there yet. If MKS release the source (pinouts from the mcu, pcb file so I could follow the connections, etc) I would be willing to put in the work on an in-depth teardown. My guess is that since those boards (as well as the rearm) are descendents from Smoothieboard, they will have similar strengths and weaknesses because they are all using the same processor family and pin footprint, and similar pinouts to the Smoothie (as well as the same software HAL)
Dear Alex, I enjoy your videos. Thank you. May I ask for a guess? My present project is to klipper my xcore tronxy. So I started with my ramps 1.6.plus and 3 tmc2130 in spi. Fine.
You think I can run them along with two 8825 for the extruders. I do not find any information around about a mixed setup of spi and dir/step ... thank for a brief yex/no/maybe
Alex,
Do you really need to remove the CS jumpers? Since the missing pins are N/C I imagined plugging the leads into AUX2 on the RAMPS board rather than the drivers directly. I haven't tested to see if those lines are truly N/C or if they are connected to ground or something else inconvenient. Mind you, when I thought that I hadn't thought forward enough to realize there will also be "flying leads" for the endstop lines.
I've been assuming that hardware SPI takes less resources and/or would be quicker to respond. Does hardware SPI have any benefit over software SPI? Are any of the 3 SPI clients fast enough to matter? If there is benefit to hardware SPI another option is to remove the 3 pins from AUX3 of the ramps board (we all know how easy pins come out of header plastic) and solder 3 pins to the top of the LCD adaptor board and run software SPI just for the display and SD card. Still a hack but much cleaner I think.
Sorry that I keep lagging behind you but the printer this is going into is taking me much longer to finish. Once I get the CoreXY movement dialed in I can actually try stuff before talking to you. Soon I hope.
The jumpers can stay on for the pins that are N/C (like the X axis: my first test I jsut reassigned the X axis CS and left the other jumpers on). They don't SEEM to be hooked up to anything else. SPI speed (and i2c speed for that matter) is a tough question because of the peripherals and protocol version/standard. The problem is not in running data fast, but rather that the ST7920 has to be fed in chunks, with specific delays... so essentially 'slower', and running it in a slower SPI mode helps, but that may be due to timing functions on the specific hardware that fix 'worked on'. 32-bit is the wild west still because of all the different variations, and X nanoseconds delay may not be X nanoseconds from one to another, especially if we are talking about blocking commands and thread conflicts (or just overhead from a call... crude example, but you could put in a NOP to delay a few NS, but if you put in a few NOP in a row it may fail to inline a function call and present a delay overhead that will then put the signal out of spec for the device). The st7920 is also goofy about init order on a shared bus, doesn't respect the CS low signal properly, and for some reason it can stay in a strange state between reboots, making troubleshooting even harder. I may do a separate video just on that panel since it is so quirky and so popular, but I don't even know all the details, but the 'last resort' for me has been using a physical gate to handle the CS pin strangeness (had to do that on 2 different projects) which 'works' but I would consider it a hack.
What is your opinion about the BIQU TMC clone drivers? I have heard some mixed things about cloned drivers so I'm generally debating if i should go for genuine ones or Biqu drivers
I only have the inverted ones, and they run hot, not sure about the others. I have a stepper comparison video coming out in the first part of the year, so I'll make sure to include them in the comparison.
Very nice video, as usual !
I have pretty much more glitches on my RRD graphical LCD panel.
How do you fix it ?
Did you even bother to watch the full video?
Hi! Nice video , did you know how wire the bltouch?
Good tahankyou
Got a question if you use a 2004 reprap discount and that adapter chip (328p based arduino) have the same software spi problem with the 2130's and what about the 2208's seems like software uart seems like the right uart pins are not broken out on the re-arm board, hate when designers cripple chips grrrrr.
terramir
Great video, how did u you know my Ramps 1.6 plus arrived yesterday? Would you consider to make an index for all your videos on one page? You got pretty unique content
I have plans to make a website with more complete video notes and an easier to follow video index. I already grabbed the domain, I just have to find the time to build it.
Hi, Alex! I am currently on my way of setting up pretty much the same thing, but without the plus on the ramps, so a ramps 1.6 with re-arm and the lcd and the TMCs. First, I got the LCD running which was quite a no-brainer. Now I am struggling with the TMCs, I tried to run them in software SPI with the pins on P1_00,P1_08,P1_09 but it is not working, I get a tmc comms error on the lcd. When I put the scope on the pins, I do not even find a clock. OK, so now you came along with your awesome video which presents an alternative solution. If I get you right, you put the software SPI on the hardware SPI pins and run everything on software? Would you recommend to further look into your plan and just use pure soft SPI or should I keep trying and keep the LCD on hardware and try to get the software SPI working with the Trinamics?
Not sure why your original SW SPI assignments didn't work, but I would uncomment PINS_DEBUGGING in configuration_adv.h and then send a M43 in the terminal to see if there are assignment conflicts (for instance, if they LCD you used trigged a macro to assign that header to ethernet serial, etc). And yes, the workaround for the 1.6+ and the MKS drivers I presented is to just run everything off the common HW spi pins, but in SW mode, but it shows a little screen corruption, so it may be better to do it your way if you can sort the pin issue out. Also, there are some conditional macros that toggle in the background based on the panel (not sure which you are using), so they MAY be blocking HW SPI from working properly, depending on the panel. I have a knee-jerk reaction against bit-banged busses, BUT tht may be unwarranted these days, and I'll have to do a speed check, because I don't know off the top of my head, what the speed difference is between HW and SW SPI in it's current form with regard to the SD card read times, or the mcu cycle overhead cost. If they are negligible, then soft SPI is just easier, but it is not something I have looked at in a while. Marlin is also using home-brew profiles for their graphical libraries, and I heard the SW is quite fast compared to the stock u8g, but again, I haven't tested, so I can't say for sure (I know from tests that on a raw arduino, HW was 4-5x faster for GLCD panels using u8g lib, but that was a few years back with the old library).
@@AlexKenis Hi Alex! Thank you for your answer! I eventually got it sorted and it off course was a case of severe stupidity. I am not used to the 2130´s and thought they wouldn´t mind and allow me testing at usb power. Off course i did not connect motors since I did not want to drive them at that point, but I still got that stupid error that would not go away...
Looking forward to the stress test results. I got the exact same drivers and the little testing I have done so far did seem to indicate overheating issues - the Y axis seems to stop and completely disable its stepper motor for couple seconds every now and then. I am in the middle of a upgrade/rebuild and switching from Marlin to Klipper at the same time, so I have a lot going on that could contribute to the wackiness.
I suspected as much. Trinamic chips are great, but they do tend to get hot, and they have some other issues that can put them in a temporary shutdown state as well. Klipper is really picking up steam in the last few months. I plan to include a Klipper video later on in the 32-bit series since they have added a lot of the features lately that people have been wanting, and it is getting closer to being a viable complete alternative.
@@AlexKenis A quick update. After upgrading active cooling on the drivers, my issues have completely gone away. Flawless prints every time. Very happy with these cheap clones.
Hi Alex, Does this RAMPS 1.6+ supports 24v input power?
hi Alex i have some problem with my control with ramps 1.6 and arduino mega 2560 the power supply 12v can't run them ,do you have solve with this problem , and if no what is type of control you advise me to use ?
what is the Pc-on 5V vcc jumper on the board for?
Hi Alex! Great vids man. I have a Ramps 1.6 (not +) with DRV 8825 drivers. I have smoked two Tango v1.0 boards already so I am very cautious with the Ramps. Do you think you could make a short tutorial on setting it up for Anet A8 with Bltouch sensor? I would really appreciate the help. Thanks for all the valuable info so far anyhow!
That sucks. How did you blow it out? Was it the BL Touch or something else? You're right though, I should do a sensorless homing video sometime
Alex, is it possible to run TMC2208 UART with this Re-Arm + RAMPS hardware combo?
They will work in standalone mode, but as far as UART control... Still not positive. Technically it should work, but in my one test I did, it was detected and I could write to it via console, BUT I had communication errors and could not get it to move, so it's possible there is a software issue? I'll have to look into it further when I get that far. Another issue is that you are a bit pin-constrained, and there may not be enough pins with the proper features available to do a full 2208 setup with UART since you need 2 extra pins per driver.
@@AlexKenis I was told that the re-arm did not have enough interupt able pins to drive tmc2208, but then again i also wanted to drive 5 drivers
hijack Q: is it only with the 1.6+ that its not possible to drive both lcd+sd+tmc2130 in hardare spi mode?
@teikjoon do you know of any proven 32-bit board (Due?), Ramps and TMC2208 UART combination ?
@@boherrmannsen8219 RAMPS 1.6 plus has some convenience lines wired from the hardware SPI pins to the correct pins on the tmc2130 drivers. It is no different from the wire spaghetti normally used.
Hey Bo... long answer: Like David said, The 1.6+ is different because it hardwires the SPI MOSI/MISO/SCK and CS pins, so the 'hack' was to re-route the CS pins to inputs that exist on the LPS1768, and use shared softSPI on the HW SPI pins that are shared by the LCD/SD/Drivers. You are free to use flying leads to wire it up any other way if you don't want to use the on-board convenience routing (you just save some pins and a nest of wires this way).
As far as the 2208's, I have gotten them to behave 100% properly YET on any LPC-based platform, and since they were my 'go-to' chip for a long while, I am very familiar with their inner-workings, so I don't think it is user error. I know that the SW UART object is created by the TMC driver library, so it may be either that the tmc lib needs a small tweak, or that there is a peculiarity about LPC that is causing intermittant communication.
I suspect that it will be fixed eventually, but I have not gotten to looking at the serial stuff yet. And yes, as far as I know(??) the serial needs an interrupt, and from what I can see in the docs, 0 and 2 are interrupt-capable (at last pinchange capable) so you will need to leave at least 5 of those open for the RX pins if you want to run all 5 2208s (and I imagine they have to be 5v compatible since serial is a 5v protocol? Or is it running with 3.3v... I never thought to check until now). The TX pins can be whatever (if I understand it properly). I really wish more of these projects used something like Doxygen and detailed annotation, because it is very difficult even to jump in and contribute to small issues if the inner-workings are obscured.
There are other options to solve these issues, but they would require a driver library re-write (which may be prudent seeing as so many people are running into pin count issues, especially for those printing with 4-color heads or pin-starved boards like the teensy/esp32, etc). If we could daisy chain trinamics (SPI or UART models), or parallel feed and then add an identifier bit, or run them with an addressing logic, etc... we could reduce the pin count greatly. (But I am not going to bother teematlut about it though because I think I annoyed him a while back asking for an additional error flag to be implemented). Same goes for end stops: a diode/resistor ladder fed to an analog input with a separate unified output to trigger an interrupt on state change would save potentially 5 interruptible pins if you are running all 6 stops (even though we have enough horsepower these days to PROBABLY not have to worry about the cycles used up by non-interrupt endstop state polling). Etc etc etc. There is much that can be done, which is why I get frustrated when people say 'we have taken this about as far as it will go'. It reminds me of the "640K of RAM is Enough For Anyone" statement
But it is also possible to run them in standalone mode until the details are worked out, which is not a perfect solution, but at least you can work with the drivers while the community works out the details. Another option is to temporarily use something like TMC2208Pilot and a separate Arduino to control them outside Marlin
Nice video! Thanks ！We will adapt the SPI mode of the Ramps1.6 plus based on the RE-ARM .And then we have the new version of TMC2130: it's going to be chip-down, Non-soldering the Pins , and then cheaper, everybody like that?
If possible please don't make a RE-ARM only version of the Ramps 1.6 plus. Change it so it will work with either the Arduino or the RE-ARM. Perhaps a set of jumpers which can switch between the two, if not how about solder pads to do the same. My first idea is to have 3 jumpers between the hardware SPI lines on AUX3 and the SPI lines to the drivers. This way I could remove those 3 jumpers and use 3 wires from the RE-ARM board to the driver SPI lines.
@Alex Kenis dude you got noticed :-)
@BIQU 3D Printer
hmm interesting i will be watching
it would seem i will not put in the order i was about to and hold off for a bit to see what you do
When will be released?
Also if you make a new RAMPS board, please make sure it is 24v ready both for RE-ARM and Arduino. The latter would require a jumper to disconnect power from the RAMPS board to the Arduino board. A lot of users have been looking for a 24v ready RAMPS board which doesn't require modification.
I've read somewhere that the TFT displays have the ability to somehow store print progress to memory, so in the event of a brown out, you can easily reheat and pick up where the print was left off. I'm still trying to find the details, but the model was the MKS TFT. IDK if other TFTs can do it, but the MKS uses an STM32 (perhaps the linked one?) and has it's own firmware and itself is 32 bit. Have you worked with this chip, or displays of this nature?
www.st.com/en/microcontrollers/stm32f107vc.html
I don't do much with touch panels, but I know the STM32F chips pretty well. You don't need a fancy chip for that feature though. Touch panels like that basically have their own firmware, and they take care of parsing all the controls and then send out the end commands to the printer board. You can pretty much program them to do whatever, so it is very possible that they have a power-out-resume feature that does something like logging the last line of code parsed and then re-homes and picks up from there when it boots back up. Sometimes it works, sometimes it doesn't, but it has been getting better. Marlin has power loss recovery built that does that if you enable it.
@@AlexKenis I did not know Marlin had it too, I would prefer the Marlin solution. I was under the impression, that after the brown out, the machine wouldn't need to re-home due to knowing the current positions of the axes? I can't picture successfully homing Z with a half printed model on the bed.
Yeah, not sure how well it works. I have not messed with it really, and I know it has been re-done a few times because it is a tricky thing. Prusia has it in their Marlin fork with a capacitor that stores a small charge, and a sensor board. It goes through a little routine when it senses power out and then lifts the z axis, moves off to the side and re-homes.
Hello Alex, I applied all you said to Rep Rap Discount SMART Controller (with modded cable) and TMC2130 with SPI are working fine. Also no glitches on the LCD. The only problem that I have is that SD card is not working anymore, I mean I can see it but I can't browse it (LCD just shows empty). I had to re-route P1.10 and P1.14 because are used by RRD and used P1.00 and P1.08 instead. Any hint? Thanks a lot!
I didn't try the SD on the panel, just the one on the rearm, so I'm not sure if it works or not. I think I made a note of that in the how-to section, but I don't remember. I know that Marlin has the reArm SD socket enabled by default, so it may be as simple as enabling the SD on the panel, but again, I didn't try it. I THINK the option is called LPC_SD_LCD in the pins file.
@@AlexKenis thank for your replay. Yes, that's the option, LPC_SD_LCD but no joy. I'll give a try to the one on the rearm, do I need to change anything in pins_RAMPS_RE_ARM? I mean, after enabling LPC_SD_ONBOARD?
EDIT: ok, I just enabled LPC_SD_ONBOARD and internal SD is working fine. Too bad SD LCD is not working, it was easier to use for me
Glad you got it working. Sucks about the panel LCD, but this is a workaround anyway, so hopefully everything gets ironed out in the firmware soon.
@@AlexKenis Thanks for your videos, they are really helpful! Ciao from Italy!
Hi, can it be used with Arduino Due?
Hi, great video! could you tell me what are those extra thin blue(D78),orange(d76) and green(d74) cables behind of the cs pins (J12 ethernet / gpio)? thank you
Those are the default pins for software SPI in the rearm pins header in Marlin: TMC_SW_MOSI=P1_00 (D78), TMC_SW_MISO=P1_08 (D76), TMC_SW_SCK=P1_09 (D74)
@@AlexKenis thanks for the reply. Where are the other ends connected on the ramps 1.6+? 😁
HA, well I did a few versions in this video and I probably should have made it easier to follow, but if you are using the drivers that are meant for this board, you don't need flying leads, you just have to set the jumpers properly, but if you are using watterott or other tmc2130s with flying leads out the top, you can use those pins for the software SPI to the drivers themselves
did U think about making your own board based on ramps
I have built my own boards just for personal use. I may put out a public design in the future, but I would more likely do a video series showing the process from beginning to end. I'll have to think about it
Could you make video about rearm + ramps 1.6 + tmc2226?
Please revisit and complete this Alex. I'm waiting ...:)
HA yes sorry. I am still playing catch-up. When I have finished the stuff that is piled on top of that printer, I'll get back to it
thanks
now to watch it at lest 4 more times to let it all sink in then to re-touch my hardware :-)
I saw the monster core XY machine you are working on. Hope this helps. I'd recommend grabbing some of the diode protector boards because the output clamping diodes can help prevent the trinamics from missing steps if a big heavy gantry pulses back EMF on direction changes. Sometimes it helps (tough to tell when becasue the overcurrent shutdown warning flag is tied to the ground fault flag).
thanks i hadnt thought of getting any diode protector boards but i guess i will just to be on the safe side
i am building 2 of the core XY units 1 for me and 1 for a mate (all the parts are being printed on my TronXY X1)
what do you think of the Z homing? (100% my design )
th-cam.com/video/Ff13Il7rLs8/w-d-xo.html
that's actually a really nifty idea. as long as it doesn't rattle around, that looks like a good way to do it without messing with sensors or servos. Maybe it could even double as collision detection. i like it. keep me posted on how it works out
here is a bad pic of the CAD drawings so you can see a little bit more detail and there wont be any "play/slop" (well none as yet)
i.imgur.com/YzirDap.jpg
@Alex Kenis
"what if / can you" move the screen and SD reader to I2C so the TMC2130's can have exclusive use of hardware SPI ?
Hi Alex,
could you share your Marlin code for this setup on GitHub please?
Will it be fine if i connect rearm with octopi?
It shuld work, but I have not tried it yet. I'll try to remember to do that for the next part of the video. I think the SD card onboard uses a shared mounting point with the USB, so that may be an issue... or not, I'll have to test it.
Looking forward to your review. And thanks a lot for all your reviews, they are great and answers all the questions
thanks for your very informative video, I have RAMPS 1.4 on and a MEGA and got the Re-ARM and the 2130s, of course, will run them without SPI, but I want to connect the PT100 sensors with the relevant amplifiers, connected at A11 and A12, but because they are 5v, how do I connect them on the Re-ARM... thanks a million
you can use E3D PT100 Amplifier board with 3.3v supply.
The signal will be 03.3v and Re-Arm Aref is 3.3V so the thermistor table 20 in Marlin don't need to be modified.
Or if you use an external 5v supply, you will have to step down the PT100 amplifiers signals, so use a voltage divider with rail-to-rail OPamps.
Something like: PT100 Amp board signal 0-5v => voltage follower 0-5v => voltage divider 0-3.3v => voltage follower 0-3.3v => Re-Arm ?
But, A11 and A12 are not available on RE-ARM, you have to use anothers analog inputs.
@@ketchutrez3457 excuse my electronics ignorance, my amplifier board is 5v and it takes the supply form the ramps. Do you mean I can use 3.3v instead I use this oneon it? is voltage divider thing an external circuitry to change the voltage of the logic to 3.3v, do I have to make it or is it, ready-made, somewhere. I use this one
www.aliexpress.com/item/Free-shipping-3d-printer-part-Accessories-V6-PT100-thermocouple-Amplifier-Board-up-to-450-Degree-Hotend/32836133708.html?spm=a2g0s.9042311.0.0.27424c4dBRCBlv
@@khomouda you are welcome.
Yes, your PT100 Amplifier board "is" 5V and yes on a Ramps + Mega2560, it takes the supply from the Ramps which is at 5V.
So with the Re-Arm + Ramps, you know that you can supply only 3.3v, and also that the analog inputs support 3.3v MAX.
And so your ask yourself if you must supply 5v to the PT100 Amplifiers boards and if you do that, how to deal with the 0 to 5v signals coming from them.
So, I say you that you can supply 3.3v to the labeled "5V" pin on the PT100 Amplifiers Boards..
So like this, the signal coming from the amplifiers will be in safe range of 0 to 3.3V.
And like the Re-Arm Aref pin is wired to the 3.3V, you don't have to edit the thermistor table in Marlin.
So.. you simply have to plug your PT100 Amplifiers boards, like you use to do on the Ramps + Mega2560.
Praticly nothing change, but you will have to use anothers analog inputs since A11 and A12 are not available on the RE-Arm.
(Don't forget to edit your Marlin config files)
Voltage divider split your voltage but add some impedance.. so I prefer use before and after the voltage divider, voltage follower. Don't bother you with this.
@@ketchu13 thanks again. I tried the pt100 at 3.3v but it did not work
@@khomouda "did not work" ? can you use a multimeter to check the pt100 amplifier board signal voltage?
Hey Alex! Love the channel as well as the content you put out! I've been looking for some time to find a creator that really gets into the nitty gritty nerd stuff. My question.. what would prevent someone from using servos like DMM-Tech, ClearPath, Yaskawa, or even Allen Bradley servos On boards like the Smoothie or Deutwifi. Being a Machinist, and doing industrial mill retrofits on the side. All you see is servos. Correct me if I'm wrong, are servos not way more Precise and accurate compared to steppers. Combine that with Quality linear rails and actual ball screws. Maybe even encoder feedback. You'd probably get rid of most if not all of the inconsistency problems in the finished print. I have come to the opinion that 95% of the hobby 3d printer crowd are just fine with capable stepper machines. So these questions don't come up, and in part, to build a machine with industrial grade hardware is expensive. I'm looking to build a bigger machine (3' x 3' square) for printing V8 intake manifolds for high hp applications, or really dimensionally accurate +/- 0.0005 and repeatable parts from expencive materials like PEEK or Ultem. For vacuum forming ABS or Polycarbonate. Maybe fixturing parts to use on my CNC bed mill. I also want to print molds to make sand cast aluminum and steel parts. I have all the Nippon Bearing linear rails i need as well as the THK ballscrews. I have a few Allen Bradley servos. I'm just stuck on making them move. Anyway, thanks you again for the time you put into making all of your Subscribers better at what we love.
Those Nippon raile are no joke! shaped rail and recirculating roller bearings, whew. But, I also wish there were more interest in servos. It is starting to head that directions somewhat, and I think we will get more progress now that 32-bit is coming of age. They have a ton of advantages over open-loop stepper positioning, and really only the disadvantages of cost and complexty. Heck, brushless motors and optical encoder disks have been the norm in office printers for years. But of what you mentioned, I only have experience with the Allen Bradley stuff from years ago, and I'm fairly certain they used their own communication protocol, which would likely be the barrier since 3d printers use the step-direct method. I don't even think any printer firmware has CAN-bus support yet: Marlin has an experimental inter-device i2c bus that never got off the ground, and Klipper has their own thing, which is probably the most promising since it is built in from the ground up and would let you more easily string together axis controllers/counters/encoders from separate mcus. There have been a few projects moving toward closing the control loop like mechaduino and uStepper (hybrid-servo systems), but the cost-per-axis has kept them out of the mainstream since open-loop stepper is 'good enough' for most use. But the answer to your question is that we don't have it for the same reason that we still use STL files: there is not much pressure to change since it is 'good enough' for most general users right now, but hopefully there will be more progress in the near future
Thanks for the reply Alex, i hope they get to servos in the near future. Nema 17 DMM-Tech servos are awesome and cost only ~$70
Aside from servos, what about closed loop steppers? The "hybrid" steppers. Seems that they alone would get rid of the loosing steps and torque issues.
What is your opinion on running a smoothie board that does 32 micro stepping paired with a stepper that has 0.9° resolution. Would there be any cons? The maths in my head thinks doubleing the resolution would be the way to go.
That 4 letter word LCD responds to all transitions on EXP1 pins 3 & 5 (SCK and MOSI). My suggestion is to not have any shared SPI pins by 1) using the onboard SD card and 2) dedicating 2 of the ethernet connector pins to the TMC software SPI's SCK & MOSI.
Sorry - I forgot that early in the video it was stated that the LCD and the TMC (and the LCD's SD card) all share the same SCK & MOSI. The easiest way to break them apart is to 1) use the onboard SD card and 2) dedicating 2 of the ethernet connector pins to the LCD's LCD_PINS_ENABLE (EXP1 cable pin 3) and LCD_PINS_D4 (EXP1 cable pin 5). In cases like this I use a breakout cable like the Adafruit www.adafruit.com/product/1199?gclid=Cj0KCQjwzunmBRDsARIsAGrt4msvk1pD9MoQrzka-TVOurL-EHjTEAo00WESeSVXmiKf3jwQafwPdoUaAj4lEALw_wcB
THat's a handy pre-made cable Bob. thanks for the link! Yes, that LCD driver is definitely a four letter word. I had luck in the past on a raw panel using a diode gate on the CS... kindof a hack, but it made the CS pin behave better. It's on the list to try in the next re-arm vid (that has been delayed for months for various reasons)
The clone protectors are they v1 or v2
I have had those for a while, so they are V1, but there are v2 clones out now for like 90 cents a piece if you get them from overseas. Just count the number of diodes on the board to make sure (the V2 have 9 diodes, and v1 have 8)
@@AlexKenis well I didn't find any v2 except watteroth, so I got those to ensure power on sequence.. and a sticker...
keep up the good work..
You can't possibly be getting that LCD to work on a plug and play basis even with marlin configured for it it still has skewed graphics but nobody ever mentions it .
Man I've clocked like 300 hours watching this video and all I get is an All HIGH error on all my drivers. Doble and triple checked wiring and pins and still no luck. Any help?
UPDATE: Replaced the jumper caps, got ALL LOW. In either case, drivers all show as disabled. Pins exactly as instructed in video, thinking of giving up and getting an SKR instead.
Sorry to hear about the problems. Trinamics can be unpredictably goofy when it comes to cables, etc. Sometimes swapping around or temporarily removing protector boards can be the thing that does it, and sometimes it is just wire routing... once I had problems for days with a setup that SHOULD have been fine, and it finally worked after I bundled the excess wire. Not sure if any of that is helpful, but the SKR is on the list to check out when I get through these vids.
Alex i have Question
Hey, if you've recently bought that Bigtreetch kit, you might get the new TMC2130 V3.0, and those come in two different "flavors", in my case they sent the one not set for SPI, so I went mad trying several pin and FW configurations to find only on this link: github.com/bigtreetech/BIGTREETECH-TMC2130-V3.0/blob/master/TMC2130-V3.0RM.pdf to their github site that I needed to solder two pads as some junky jumpers and remove a resistor from the board, which with a standard pencil style iron was really hard but actually works! Anyway a comment just to save time and frustrations...
Also, one of the 2130 overheated every time I turned on power, so I figured there was some short to ground, it was the driver itself because it would do it in any axis at idle. I tried everything, luckily putting it in the oven mid power for 15' fixed it
YOU are always so worried that WE might be worried that YOUR audio might be WONKY.
NADA PROBLEM!
GREAT VIDEOS!
I have got some important questions and ideas and for
anybody interested I will deal with it in my FB Group OpenIdeas 3dp&cnc
1.) Trinamics and Re-Arm on mpcnc
I think elstcam can drive the mpcnc wheather or not the firmware is grbl or marlin.
the automatic zero finder and x-direction finder can be made with a metal probe that
touches the work piece and detects some low currents
IDEA:
why not use Trinamics load angle detection here - so no extra sensor will be necessary?
QUESTION:
if I used the RE-ARM and Trinamics then I coud boot any machine with either
Marlin or grbl firmware - so why not compile a firmware that is Marlin AND grbl
in one piece - has anybody tried this before?
2.) pressure sensing
in many cases it would be wise to know the current pressure vecor the tool
applies to the workpiece [prX, prY, prZ] - based on Trinamics load angles -
and it would be wise to define actions for specific pressure values
e.g:
if the Z-Pressure gets over a defined threshold then - why not retract the cnc
tool a little bit and automatically drive the current path in 2 loops
controlled by firmware - not controlled by gCode ---
This is especially important in case of low cost machines that do not provide
such high rigidity as industrial machines.
QUESTION:
Who of you is interested and capable to add such
"PressureSense" to Marlin and/or to grbl?
Would the same problems be present with this panel? www.aliexpress.com/item/3d-printer-Reprap-LCD-MKS-MINI12864LCD-mini-12864-smart-display-Reprapdiscount-controller-Full-Graphic-for-mother/32913529241.html
Also, what's the difference between TMC2130 and TMC2130 DIY?
The DIY version are for standalone mode. You can solder them up for SPI control, but it is a pain. I don't have one of those panels to test, but they use a different driver chip, so double check that it is supported (I don't know off the top of my head)
@@AlexKenis Thanks for your answer, love your channel! l I just got the "new and improved" TMC2130 SPI, let's see how they do.
Do you play ps1 ps2 ps3 ps4 or ps5
Keeps coming up with errors (30 errors in total) such as:
In file included from Marlin\src\HAL\HAL_LPC1768\../../inc/MarlinConfigPre.h:38:0,
from Marlin\src\HAL\HAL_LPC1768\persistent_store_sdcard.cpp:25:
Marlin\src\HAL\HAL_LPC1768\../../inc/../../Configuration_adv.h:1381:0: note: this is the location of the previous definition
#define TMC_SW_SCK 12
In file included from Marlin\src\HAL\HAL_LPC1768\../../inc/MarlinConfig.h:41:0,
from Marlin\src\HAL\HAL_LPC1768\persistent_store_sdcard.cpp:29:
Marlin\src\HAL\HAL_LPC1768\../../inc/SanityCheck.h:1870:4: error: #error "SENSORLESS_PROBING requires a TMC stepper driver with StallGuard on Z."
#error "SENSORLESS_PROBING requires a TMC stepper driver with StallGuard on Z."
Looks like you are compiling with sensorless homing. I haven't tried that yet (probe is on order, but hasnt arrived yet) try turning that off temporarily and see if it compiles without error. If it does, then its something to so with the sensor setup. if it doesn't, post the errors and I'll see if it is anything obvious when i get a chance
@@AlexKenis the Z stepper is the only one that comes up with the sensor less proving error :/
@@unspokendesign Is sensorless homing on the Z axis even a good idea? The nozzle will end up getting damaged.
@@DavidMulligan I haven't uncommented the option for Z homing. I have a BLTouch. Don't understand why it's coming up with the error :/
@@unspokendesign It says you have. Look at the if statement one line before the error in marlin/src/inc/SanityCheck.h. it says ENABLED(SENSORLESS_PROBING) && !HAS_STALLGUARD which means SENSORLESS_PROBING is enabled and HAS_STALLGUARD is not. HAS_STALLGUARD just means you have TMC2130 driver enabled.
You might have to play with timing for your screen :)
#define ST7920_DELAY_1 DELAY_0_NOP
#define ST7920_DELAY_2 DELAY_2_NOP
#define ST7920_DELAY_3 DELAY_0_NOP
Good call. I should have played with the speeds before posting the vid. I'll mess with it tomorrow when i get a minute
@@AlexKenis I had the same issue on my Einsy Rambo and for pretty much the same reasons I guess (TMC2130/SPI/...)
Tried cranking them up to 1000 with no effect, but I'll try some other things later.
@@AlexKenis I'll be following the videos on that subject, that was one of my objective to get that setup running (RAMPS/RE-ARM/TMC2130 with SPI & StallGuard), but alas, I lack the time to do the research, so I'll be following your videos with a great pleasure :)
next project, ramps on Adafruit Grand Central M4 Express (www.adafruit.com/product/4064 ) As if you needed another project :P
Man, I saw the announcement for that a while ago on the ADA livestream. I have been purposely trying not to think about it because I am so easily distracted by new projects, lol
Hi Alex,
what I am writing now is not meant ugly!
(not ugly to you or the readers)
But really,
do we have to do all this nonsense?
that is how everyone gets lost (I think!)
and why a new PCB board designed
that does not fit well on the other PCBs?
must rummage with dupond cables?
people will say yes I want to do that
okay for me, and go ahead.
but no this goes too far for me!
then I'll wait for a PCB that's more decent
and where everything stays clearer.
I think it is a waste of all the trouble
!
do you find me negative?...... maybe I am!?
I will then apologize but it is not ment to anyone personally.
Again friendly greetings from the Netherlands!
Rob.
Rob, at first there was no information exposing the incompatibility between Ramps 1.6 plus, re-arm and the RRD Full Graphics LCD. I can't speak for Alex but his first RE-ARM and Ramps 1.6 plus video came out just after mine were delivered so I think he had similar ideas. I wanted an affordable 32 bit option that allows me to use removable TMC2130 drivers in SPI mode with Marlin 2. The RAMPS 1.6 plus board appeared to let me do that without the wiring mess. The only reason I got the RRD Full Graphics LCD is it was easy. Alex previously stated it is the most common LCD out there which is why he wants it to work.
The 3 missing pins are a minor inconvenience compared to the standard wiring mess. The LCD display garbage might be able to be solved by software once the RE-ARM version of HAL_spi.spiBeginTransaction is implemented. This may well end up being the cleanest way to use a RE-ARM with TMC2130 drivers.
Doei!
I understand what you are saying HobbyRob. This video was just a workaround to help out people that have a specific set of hardware. 32-bit printing is still in it's starting phase, so I want to encourage more development, and finding workarounds for the temporary issues people run into will help keep development moving and eliminate frustration. Even though it seems silly to run external cables and put in software hacks, hopefully it will help move toward more development of complete solutions in the future.
@@AlexKenisam happy with you! Yes, okay, we understand each other!
and with the thought Rob will not hurt anyone
(no ugly thoughts etc etc ....)
BUT I thought for very briefly ........ HELP!?!?!?
and honest!
I hope that there will be a kind of univerzeel PCB board is coming,
maybe also with bells and whistles and .......
but then that you can ... go with jumpers in some more ways or something like that?
with the thought that you can switch LCD screen (with or without SD)
Or you can easily change stepsticks without fiddling......
kind regards from here again!
Rob.
I hope so too. Like I said, this is all in the initial stages, and manufacturers are still trying to figure out what platform to go with and what options are needed. the LPC based boards are in a strange spot, because all the ecosystem is based off the MBed platform, not the Arduino platform like AVR, STM32, and SAMD, so not all the libraries will just port over to 3d printer firmware as easily, and non-3d printing support in the community is not as deep as with the arduino-based systems. I think that is part of the reason that development tends to be a bit slower for LPC HAL
Do you play roblox
Clear as mud. Why use cheap Chinese knockoffs TMC2130? WHY ENCOURAGE THEFT?
Please change the title music ( if that's music ) on your videos. The content of your channel is being misrepresented by the annoying title
NO! Death metal needs to be spread.
@@ThePazuzu Unfortunately, I won't be enjoying your spread. Unsubed
First of all, it's not my channel. Feel free to create yours with whatever pop crap you listen to. And second, if you're so immature that you don't sub to a channel because you don't like the intro music, you have bigger problems than you think. No channel would miss you as a subscriber.