Z-axis inductive sensors for Cheap Chinese 3D Printers: 6-36v & 10-30v sensors work on 5v power
ฝัง
- เผยแพร่เมื่อ 4 มี.ค. 2017
- Tom's Video about auto-leveling sensors: • Autoleveling on 3D pri...
A Neodymium magnet falling through an aluminum tube: • Aluminum tube and magnet
STOP! Before you leave another comment about why I'm wrong in this video, go watch the video I made addressing reliability/repeatability of these sensors, and also why the foil tape backing is a mistake: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
Sorry to bother you I have the same printer and the same sensor. But I´m using Repetier Host and my Start Gcode is always incorrect. Can you guide me or tell me what should say with an offset of 0.6 on Z ? Thanks !
HEY MATE . CAN YOU TELL ME HOW WILL I LEVER THE SENSOR I MEAN HOW FAR IT MUST BE FROM THE BED ??? DOES IT HAVE TO BE AT THE SAME LEVER AS THE EXTRUDER ??? THANK YOU
MY FLSUN I3 WITH SIMILAR SENSOR DOES NOT AUTO LEVER AFTER I PRESS THE TOUCH SCREEN . IT JUST MOVES THE BED FEW TIME AND EXTRUDER MOVE UP A BIT AND TO THE CENTER OF THE BED . THE END . WHATS WRONG IF ANY THING WRONG HERE ?!
Just a hint, typing with capital laters (large) is the same as you are screaming. So please don't scream.
works flawlessly.
Seriously, thank you for this video. You have taken things very practically and yet in depth at the same time. You definitely talk my language at any rate, and I appreciate that.
You were practical at the beginning and efficient and that was lovely.
And the explanation afterwards was really good to have.
Thanks a lot, (that was amazing)
Basically Lenz law for any other engineers out there. The tube and neodymium experiment is one of the ways for demonstrating this. Kudos for solving the issue with Al tape. Now I can use my tronxy 6-36 V sensor and glass bed.
you've saved my day ! SN04-N
I've received mine today and i've tried to make it work... but it didnt work and sensor light was almost on!
Finally at 1 am, i've seen in your video that 2 cables(blue and brown) were inverted !
so i've remove those 2 cables and invert them back to their right place !
Thx for close view and full HD vid !
Thx you !
'Inductive sensor' pretty much sums up your explanation. Good video.
Confirmed. I have an A8, 10-30v sensor and I placed a dollar store glass sheet on the aluminum heated bed. It works great. Easy to calibrate and great print adhesion.
i have the MKII select plus. i'm about to do some testing with autoleveling. in my hobbie of robotics, i've picked up quite a few inductive prox sensors, and was going to try them out over the next coupla weeks. fortunately i won't have to wait long for these new sensors to come in. thanks a whole hell of a lot, both you & Tom!!!!! (but seriously, thanks).
Russ from Coral Springs, Fl. usa
I'd love to learn the outcome of your testing. I too have a Select Plus, and am looking to setup autoleveling.
Thank you for your video, always learning.
I find this very helpful, hope you don't mind but I am posting it as a reply in some fb group.
To level the bed, I believe Creality tells us to home the hot end, then disable the stepper motors to move from corner to corner and raise or lower these as needed. Problem is, when I follow this procedure I can never get the bed level the first time. It was infuriating. Then, I read an article someone wrote on the net. He said to stop using disable stepper motors. So, to move from corner to corner, I had to use the Move command. Doing it this way, I was able to level my bed the first time around! My print is perfect.
Very good comments. Thank you for the video
By god sir you have earned yourself a sub. Good stuff.
Brilliant
Thank you
Interesting video - let's see if we can make Tom test the 6-36V sensor as well. Seen from a technical point of view (i work with electronics all day, have made my own inductive probes etc.) i can assure you that the voltage rating on the sensor has nothing to do with it's sensitivity - it's something that's engineered into the sensor itself when being designed & manufactured. The problem can be running these sensors outside their intended area of operation (5V). After all, then you may be lucky that it still performs good and repeatable, or it may be all over the shop. I've found some 24V-only sensors to work fine on 5V but others starting to oscillate wildly when under 10V, so YMMV. Keep up the good work.
You can wire these sensors directly to the 12v output of the board as long as it is an NPN type. The NPN is open circuit when not active and pulls to ground when active. As a result, with pullups enabled the Arduino will see either 5v or ground depending on the state of the sensor. The PNP type sensors will apply 12v to the output when they are active and float when inactive, so not really suitable for 3d printers as they require some clumsy voltage dividing circuit to be installed.
wow bro. amazing!! thank You. need this info
and good way o explanation
was very informative. thanks!
My 10-36 volt sensor didn't work with the Anet A8 board, but it was an easy fix with an LM7805. Works likes a charm now! Sensor gets a full 12 volts for operating, but signal is dropped to 5 volts so the board is protected.
If your printer is anything like my Tronxy X3, the Melzi board on the X3 supplies 12v to the sensor. I haven't had much luck detecting through glass.
It will detect on 5v but the detect radius will reduce with 50% and detect object closer then 2mm while 4mm is recommend to detect
Very nice job on this video!
DIY3DTECH.com Thank you! I made another video on the subject yesterday.
Tom's testing was conducted with reliability and accuracy in mind.
Your videoonly shows that the plate can be recognised. There is no accuracy/reliability in your tests because you have no reference, nor repeated tests.
The next point is that not everyone used/has a heated bed. From what I have read, PLA does not require (or want?) a heated bed.
Next, when running current through the heated bed, the bed will not be consistent with temperature over the whole bed. There will be hot spots, and as Tom showed, heat does affect the sensor readings.
And, on a heated bed with current running, remember it is switched on and off by the driver to keep the temperature consistent. The reading (distance) will vary by whether the current is switching, on, or off, when the reading was sampled.
Just because a light comes on at some indertimate distance, this does not prove that the sensor will be reliable, nor accurate.
Sorry, but your video proves nothing.
I have no affiliation or know with Tom. I only found his videos a few days ago.
Tom says that they don't work. He does not say that they work, but are unreliable. I was merely proving that they do work. I might go through the effort to make a testing apparatus to measure the sensor's repeat-ability and reliability, but probably not. I have better things to make videos about. The sensors work. I have them on machines which I print with on a regular basis. Any fluctuations in readings are so small that they don't affect real world performance. The accuracy/reliability issue that some folks want to bring up is purely an academic argument. I put the sensors on my printers and they work. I get prints with good bed adhesion. What more do you want?
Tom tested inductive capacitive ir and mechanical switches, they all worked with varying levels of accuracy. Inductive sensors with the correct range for your specific application with or without a heated bed are extremely reliable and repeatable, capacitive do work but do have large variances, ir work well depending on surface type but can be affected by lighting, mechanical for cost to accuracy and repeatability are probably the best for most circumstances because they aren't affected by surface type. The problem with hall effect sensors is they can be extremely sensitive to magnetic fields not just in the proximity to them but also the earths own field strength and the location of your printer or orientation in your home or office.
Gareth Price correct ! anyway a well made machine doesn't need this and its associated problems !
I know that this is 2 years later, but there is another type of sensor that could be used, and it would be extremely accurate. We used them to detect the tension on tape storage systems ~40 years ago. It’s an iron plunger inside of a glass tube with a copper cup on the outside of one end of the tube. If you spring loaded the plunger, you could use it like current switch types, but it would be repeatable to at least 1 micron, of not smaller. It’s like a miniature metal detector. A tiny unit with circuit board could probably be mass produced for under $10.
You could also do something similar with the Hall Effect sensor and a spring loaded rod. Just detect an object on the other end of the rod, the end that is not touching the bed. There’s always at least 2 ways to skin a cat, and it doesn’t need to be direct. 😎
Thanks Kaelum. This is a good comment!
Thanks for sharing 👍😀
Awesome video man!
Loved it!
I have an ANET A8 & I just picked up a 213*200*3mm Borosilicate Glass Plate for the bed off amazon for under $20
So I am just looking up all the details before getting started with autoleveling...
I have the "ANET 3D" main board, So I guess I will be needing a new board to get this all working great.
What main board do you recommend?
Hi. Thanks for explaining the different readings that can occur between a cold bed and heated. I am using a 3mm aluminum bed and a 6-36 v nc inductive sensor. Virtually every time I sent a g29 on a cold bed, the hotend would hit the bed and want to keep going! With the bed at printing temp though, 70 - 90 C works everytime. I can live with it but it seems that the 10-36 would have been a better choice. I can sleep nights now...Thanks!
doug barnes Make sure you watch my other video on the subject. It gives new information from what I show in this one.
Nice video. Would you mind providing the model numbers for the sensors you used in the video ?
So Good Vid Brother But as maintenance in production environment . The slop the mess grassy fingers can change VERY fast
I've been using the proximity sensor, tube one, on my geeetech and it detects the PCB just fine I have stuck aluminium foil under the glass like you have but it's not made much of a difference I just wanted to get a more uniform result
I just found your channel and I like it! This is some great information and although I am not the expert on sensors I am willing to try your concepts. Although, it seems to be a controversial topic, I'm still willing to explore it. I have the i3 clone (geeetech) and although I like using just the glass, I recently received the Model D Maghold Magnetic Build Plate 8"x8.5" off Amazon and would like to use it. However, I wonder if the magnetic base plate for the build tak with interfere with the sensors??
That's a good question.
Design Prototype Test
Maybe you could put together a test on that?
Most metals will react to a magnet in someway even if the magnet does not pull them. Some metals don't seem to do much when a magnet is near. The only reason we use iron with magnets most of the time is because it is cheap and one of the best known materials for a magnet to stick to. Other metals will react with a magnet in a different way. Slide a silver coin down a rare earth magnet if you have one big enough and you will see the coin move more slowly than if there was no magnet. It is one of the tricks you can do to help test if a silver coin is real or not, just don't rely on that alone.
Thank you, amazing test, hmm I was thinking about this "electrical" bed problem, did you test it on bang-bang or PID?
Bang bang. The board I was using at the time was not able to handle PID, and beds are a large thermal mass, so PID really doesn't give all that much improvement.
Thanks for the great work. I order one of the 6-36v sensor from gearbest do you think it is good for the anet a8 aluminum bed?
seetcas Yes, it will work. I have one setup on my anet a6. I'm making a video on that. should show up tonight.
Totally understand.
I've bought the Anet A8 in Gearbest with the optional sensor (rated 6v to 36v) and bracket. It doesn't work at all at 5v. I tested it and it works at 12v but it's sitting idle because I don't want to release any magic smoke! Hehe.
Hmm. That is strange. I purchased the 6-36v sensor in the video from the same place as you did. It should be the exact same sensor as what you have. There are too few moving parts in the sensor for this to be a quality control issue. In other words, if you have the exact same sensor as I do, and if you know it works at 12v, I can't believe that it won't work the same as the one that I am showing in the video. It has to be some other issue. Are you sure that you plugged it into the correct spot on your controller board, and also that you did not plug it in 180 degrees backward?
It doesn't fit backwards, I tried, so I plugged it to a 12v power supply and it worked. This is the sensor (imgur.com/a/voLjj). Could it be that the cable are flipped around? I'm new to electronic so I didn't want to fry anything..
Yes, That is the exact same sensor as I show in the video except that it looks like you have scratched off the "Tronxy" logo and model number. What happens when you plug the sensor into the board, turn on the power, and then hold a metallic object against the sensor? Does the light come on the sensor?
Hey! I tried with a 5v power supply and it lights up but if I plug it to
the motherboard It doesn't do anything. Can I just change the cables in
the connector and try it without frying anything? Or, how do I know the correct order?
You can switch the cables around. It won't fry the board. That is the beauty of 5v. The order is correct on the connector. Black=Signal, Blue=(-), Brown=(+). You may need to rotate the connector 180 degrees so it works on your board. You should look up a "pinout" image of your board to figure out which pins correspond to signal, (-) and (+).
can you provide a link to this sensor´s specs? how is it with the thermal durability of this piece? it will operate few milimeters above a hot surface..
What about a piece of aluminium with a silicon heatpad taped to it, does it affect the probing (cold vs heated) ?
I have a 2-part question. I have a Hypercube with a 330mmx330mm 4mm thick aluminum plate and a Keenvo 24v silicon heated mat attached to the underside of the plate. I also have a 4mm thick Borosilicate glass plate. My question is, will the 6-36vdc sensor work for either the plate alone or with a aluminum taped glass plate if I add a step-up inline transformer to take 5v from control board and bump it higher for for metal plate only configuration (ie, no glass) at 10v verses adding the glass plate and dropping voltage back down to 5 volts? Looking at options to allow me to run with or without a glass plate. ~Thanks~ Doug W.
It sounds like you need a sensor that directly measures the bed. This way you will not be having to constantly adjusting the probe offset distance when changing the bed type. I recommend the BLTouch sensor. However, if you want an inexpensive DIY solution look into "allen key probe."
Do you have a recommendation of what sensor to I could use with the Tevo Little Monster? I'm using the black glass since my PEI sheet was rubbish.
About my machine and plans:
I'm going over to Duet2 WiFi (no smart effector) and redesigning the effector to be more compact and remove some unnecessary weight/peripherals from it.
I'd like to use a compact sensor like the ones you showcase but am concerned about repetition against a black glass bed. I'm not a fan of the BL but if it's the only choice I have, then so be it.
Thoughts?
I highly recommend that you do not use a capacitive sensor, or an inductive sensor. The capacitive sensor will give variable readings that fluctuate with the humidity in the air. So you set it up one day, then the humidity changes the next day and all the sudden it isn't reading the same distance. Or it could happen in the same day. You set it up with the bed cold, then heat up the bed. The heat radiating off the bed dries the air above the bed and you are now getting a different reading. The inductive sensor also gives inconsistent readings. It will vary it's readings based on how hot or cold the actual sensor is. Both are poor solutions. The only reason people use them is the price point. There are work-arounds but they are just patches being applied to a sub-optimal piece of equipment. For instance, watch my video about the Duet Maestro installed into an Ender 3 Pro and you will see how I made a chart to correct for temperature of the probe (th-cam.com/video/PpLEjwLZwBw/w-d-xo.html) Spoiler: It didn't work very well, and in the end I just switched to a BLTouch with is incredibly reliable, accurate, and measures the print surface itself by touching it. This means you can change out your bed from PEI to Glass and then Aluminum, and you will never need to adjust the "trigger distance." Because you don't want to use the DUET smart effector (Which is awesome) The BLTouch is the solution I recommend.
@@DesignPrototypeTest ohh gotcha. And it's not that I don't want to use the smart effector, I'm just in a budget crunch right now and with my printer being down, it's starting to hurt my production targets. So once o get back up and running, a smart effector will be on the To-Get list. For now, I managed to snag a used Duet2 for cheap.
I'll factor the BL touch into my effector redesign then. As for temperatures, being where it is on the OEM setup of the Little Monster, is it true the heat from the heatsink will distort it's performance or is that a wives tale?
IS it really that easy to install one of these sensors? I heard stories about it being awkward to set up the software to accept them and yet this and other videos show them being simply plugged in and used without any software modifications. Plus the microswitches used as standard with the Anet A8 give only a normally open connection until it's triggered and doesn't use any electrical power, so how does the board supply 5 volts to the sensor?
I was considering installing a similar sensor, they're all over Ebay so they must be popular. I intended to use a completely separate 13.8volt fully regulated power supply. I will be using a 3mm borosilicate glass on top of my heat bed. I figured the sensor with the highest measurement distance would be the best.
Is there a software modification I have to do and are there any problems I'd be likely to encounter please? Thank you.
Yes. You will need to modify the marlin installation file. If you already have that file to modify, it can be quite easy. If you have to write/calibrate the file for your printer it can be incredibly hard depending on your competency with programming in the Arduino environment. If you have an A8, you should be able to fairly easily find a copy of the firmware on the internet as that is a very common/popular printer.
that is because a 3-36v sensor has a detecting gap of 8mm on the other hand the 10-30v one has a 4mm detecting gap
I wish you would do a video showing how to set one up on a printer, calibrate and set upfirmware....
Ok. That will be the next video I make after this one which I'm currently working on (upgrades to the Delta Printer).
Its a bit silly to worry about not using 12v is you have 12v for the fans on the printhead anyway. Also if you want to use these with Glass use fairly thick *Copper* tape not thin Aluminum tape.
First: Do not use tape of any kind. It will throw off your calibration at the seams as seen on my other video here:: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
Second: If you feed 12v into the sensor it will feed 12v out of the signal line. This high voltage will fry the logic circuitry on your control board which is only designed to handle 5v. Meaning if you want to run these sensors on 12v you will have to do some extra work to bump that signal voltage down, and you now run the risk of burning up your board. It is easier and safer to run these sensors on 5v.
heh, well I have a voltage divider in there obiously. I don't use Tape anymore because of me getting an aliuminium heatbed - I just wanted to point out that copper seems to work several times better than aluminium. Also I don't want to imply that it works well ;)
blkhawk Surprisingly, copper has a very small advantage over aluminum in this use case. To increase sensing distance, the best substrate material would be ferrous (iron, or steel). Source: automation-insights.blog/2010/04/12/inductive-proximity-sensor-targets-material-does-matter/
ah - maybe my copper tape was very thick then - but again I do not use this anymore
SN04 does not work with magnetic removable bed, any work around?
The 10-30 is your best option with an aluminum heated bed, unless as you earlier proved, you have a piece of glass over it to smooth out your prints. Would not be surprised if the 10-30 can't read that any better than aluminum tape, which means back to the 6-36 NPN and that means zeroing it with a preheated bed. Or you're back to looking for something mechanical.
Great, so how do you WIRE one of them for 5v? There's three wires for
the sensor, but the mechanical endstop switch that cheap printers use
only has two wires. Plus they are always in different colors.
The sensor should have a pin-out diagram printed on it, or that came with it. The three wires are (+), (-), and signal (s). Find the pin-out diagram for your board (sometimes printed on the board itself) and simply plug the correct wires onto the correct pins.
So what's the verdict? Cost per performance what should I get to compliment a MP MS +?
There's a terribly large difference between "will work on 5v" and "will work on 5v with consistent, reliable measurements". These sensors function, but their drift is simply unacceptable for 3D printing on low voltage.
Agreed, they have specifications for a reason, going outside those will very likely result in inconsistent results, or possibly crashing into the bed if it failed to report altogether under varying conditions.
I don't have these exact sensors personally, but i have some experience with hall effect sensors as i've built them into cheap BLDC motors to use them for positional control operations. Basically get a motor with the benefits of a stepper high direct drive torque with continuous operation and no misstepping.
I tbh do not know why these sensors wouldn't work for you, besides the fact that they have a noticeable hysteresis (which can be worked around) but really is just something you need to deal with.
If you are having inconsistent behavior i'd recommend you try regulating your 5v more carefully, adding capacitors (100nF is enough) from power to ground in the line to the sensor should help. Depending on if they did so in the sensor module you might also have to add a pullup resistor to the output line to prevent a floating signal on your input.
If you are still having problems then you have an unreliable 5v source voltage and should add a more precise voltage regulator to be used for sensor reference.
If this seems like too much work and not worth the hassle, then dont buy chinese parts in general. I hope this helps.
Hi, and thanks for alot of good content :D Was wondering if you might have a diagram on how to wire it up? I currently plan to use the 6-36v sensor on my Tronxy X5s 400, but got kinda stuck on the wiring part :-(
Wow! I forgot to include the scene which I filmed explaining the Control board on the X5S 400. Bottom line is that there aren't any extra i/o pins on the control board of this printer. You will have to do some fancy soldering to traces on the board, and also build a new firmware from scratch as TronXY has not released the source code. Yes, they are in violation of the GPL.
@@DesignPrototypeTest Ok :-) I have tried to look for a signal around 5v on the board, and the one one I found was where you connect the Arduino on the 6 pins :-( I also doesnt quite understand the PNP NO (normally open) probes, as it seems like the Z limit switch goes to 0 when triggered :-(
@@DesignPrototypeTest Hi from Norway btw :-) Really love your content, and just ignore the haters :-) They are gone hate no mather what unfortunately :-(
Thanks for the info about the sensors.
I am going to try to install a inductive sensor in my Tronxy 802E. But I do not see any actual instructions for my software. I will be using RepetierHost_1_6_2. Do I have to reload the software with the new settings for the sensor to work?
All of the youtube videos talk about Marlin software not Repetier.
Where can I get instructions for fpr Repetier?
Thanks for any help.
I will cover how to do this in a future video. To get the firmware to work, you must modify the firmware for your board. This means making a custom build of Marlin. The slicing software (Repetier) has nothing to do with it.
I am a newbie as far as software. I am assuming you are talking about the software in the printer? When my printer boots, it says Repetier .91 not Marlin.
You are correct, the Repetier 1.6.2 is the software running with the computer to do slicing and control.
Is there Reptier software in the processor on the electronics board, I assume?
Ah! Yes, there is a firmware generation tool made by Repetier. I forgot about that. See if you can update your firmware through the tool located here: www.repetier.com/firmware/v091/
A quick note however. Once you flash the new firmware onto your board, you will not be able to reset it to the original firmware. It is difficult to backup your original firmware. Unless you can get a copy from the manufacturer, you might want to just purchase a new board and play with the firmware on the new board. This way if it doesn't work, you can always just reinstall the original board. For the lowest cost, and greatest compatibility, I recommend a Ramps 1.4 board kit such as this one for under $30:
www.ebay.com/itm/3D-printer-kit-RepRap-Prusa-i3-HSG-RAMPS-1-4-Mega-2560-A4988-LCD-2004-Display-/112205898484?var=&hash=item1a1ffe22f4:m:mWUUFpgvy8eaN5ZQ0q1aAFg
Many tutorials/forum posts exist explaining the process of getting a Ramps 1.4 board to work. Good Luck!
Hey I have the tronxy xy3 which sensor can I use?
cool stuff, but it's either inductive, reluctive, or hall effect. If it's hall effect, it's not inductive.
all you need is for the 6-36 volt 12 volts is best how you do this... wire it from your arduino .... so buy a step up converter for 2 bucks on ebay.. the negative from the arduino to the negative input on the converter.. then the positive from 5 volts on the servo lead to the converter on the other side of the converter you put positive and negative to there pads.. and use the signal just straigt to the arudino servo pin.. basically its in line.. then adjust the voltage.. well adjust the voltage outside first with sensor off then install it.. and voila 5 volts into 12 volts.. i hightly recomnend not doing any inductive use a bl touch.. marlin is set up for them now and you can swap out any plate annnnything and the bl touch will detect it....
He said you wouldnt get a reading off glass with aluminium tape added to the back. And your test proved him right, basically he was saying you cant get around the problem of the sensor not working by adding tape, aluminum tape will not detect-so no reading!! Also using that sensor in this arrangement is toward the edge of the sensors operating parameters, a different batch or some voltage droop from anywhere may affect some setups. I get that it works and in your case was handy as it was simple, but if all the processes and systems around us were built with a similar margin for error or deviance we would forever be troubleshooting or pulling our hair out. Imagine your pc hanging 1 out of every 1000 tasks but always for a different reason, because of the designers "good enough-it works" design mindset built in many less robust automated tasks.
Why go sketchy by design? We can easily engineer a solution which sits more comfortably within the components operating parameters and reduce possible issues some setups may encounter. We need to make 3d printing more robust in its processes not less, any fixes should be done good enough to move on from and avoid introducing another possible failure path in the already unreliable 3d printing workflow.
I have tested over 20 different induction proximity sensors. All of the ones with the square plastic shape work to detect foil backed glass. All the circular metal barrel types do not. Watch the follow up video I made to this one: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
Also, I am strongly of the opinion that induction proximity sensors are a bad solution. They don't measure the bed directly, so what happens if you have an air bubble in your PEI sheet? Your offset value would be incorrect in that spot on the bed. Also the readings they give are substantially influenced by the temperature of the probe itself. This is why Prusa had to add the thermistor to his sensor.,
As far as sensors go, the BLTouch is a much better option.
What if you have a tronxy with a glass bed over the aluminum heated bed? That's my setup.
buy a capacitance sensor.. same shitty thing.
I'd like to know, too.
won't work i just tried it since i got the sensor for $3 on ebay............ wont detect through glass but will with the heated bed if it has metal top. the heated bed doesn't cause it to change due to heat either. all in all its great for the stock bed but i like printing on glass its way better
once again.. use a capacitive sensor, not inducitve one.. they both have the same shitty accuracy. If you want accuracy on any surface, you need a touch probe like "bltouch".. www.ebay.com/itm/LJC18A3-B-Z-BY-PNP-NO-3-Wires-10mm-Capacitive-Proximity-Sensor-Switch-DC-6-36V-/161701766400?hash=item25a62d2500:g:QyoAAOSwWfFXjq~r or www.ebay.com/itm/BLTouch-Auto-Bed-Leveling-Touch-Sensor-w-Spare-parts-for-3D-Printer-Extruder-/262827728333?hash=item3d31c0ddcd:g:REoAAOSwo4pYZRjU
I use an inductive sensor with this setup and it works well (2.5mm gap).
Did you made any modification to the sensor connections to the ramp boards? Can you please tell me the order of the cables your sensor comes with?
Black=Ground(-) Brown=5v(+) Blue=Signal(S). Watch my latest video for a tip on how to make your own plugs to connect the sensors to your board: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
I have a Tevo Tornado, which has a 110 AC bed heater, would the collapsing mag field at 60 hz work better?
I made a second part to this video. You can see it here: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html In that video I did a little bit of testing to determine that the electricity running through the heated bed has no noticeable effect on proximity readings. At least not for 12v DC. power. It is a very distinct possibility that your 120v AC setup would effect the readings. I can't say for certain because I've never tested it. Just to be safe, I suggest you heat the bed up, turn the bed heater off, run your bed leveling algorithm and then turn your bed heater back on before printing.
What about Geeetech-3D-Touch-Auto-Bed-Leveling-Sensor?
My printer has a Aluminum bed, i put a mirror ontop of it, would any of this sensors work on that? maybe without the aluminum tape? Mirrors contain Silver Nitrate will the sensors sense that?
Like I showed in the video the aluminum foil is detected through the glass. Your Aluminum bed would similarly be detected. However I highly recommend against this type of sensor. While inexpensive the measurements they provide vary based on how hot or cold the sensor itself is. This means that if you set up your printer perfectly but the sensor is hotter the next time you try to print, your first layer won't adhere to the bed. A BLTouch is superior.
where did you get the diamond hotend printed extruder file ?
I drew it in Rhinoceros 3D. I will put the file on thingieverse once I'm sure that it works.
does it work on geeetch rostock deta 301 printer
is the brown the hot wire, black ground and blue signal for the 6-36v? i am trying to set it up oon an anet a6, right now there is a red far right hot wire, center empty and black far left ground.
Hi Brett, all sensor do not share common wire colors. I can't tell you which wire does what on your sensor. Most of my sensors have come with a little tag on them explaining the wiring. At any rate, if you are running 5v you can plug in the wires in different combinations until one works. Remember of the three wires you have power ground and signal. Black is almost always ground, and blue is almost always signal. The only wire left is positive power (AKA Vref). I made a follow up video to this one, it might help you if you watch it and pause at key moments: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
Unfortunately, these low cost prox switches aren't typically high quality, and are heat sensitive.
your voice sounds like Luke's from modern family
I had Tronxy XY-08N and the 2mm mirrored glass with aluminum foil under. It seemed to work fine til the heated bed rose to 50 C, the Tronxy XY-08N didn't detect the heated bed, though trying with/without the glass. Any ideas?
That is interesting that it worked until it got up to a certain temperature. I will test this in a future video. In the mean time, to solve your problem, I've read in forums in the past that some people have had better luck with copper foil. Copper is more conductive than aluminum, so it should provide a better signal for the sensor to pick up. Maybe try this: tinyurl.com/kkb7mla Then again, you can get an aluminum bed for under $20 on ebay.
Thanks for your reply. For the heated bed, I'm using an aluminum bed originally comes with an Anet A8. But having aluminum foil under the mirrored glass, the sensor works better. Anyway, I'll try copper sheet. Thanks again and waiting to watch a future test video.
Ah. Given that information, I don't think the copper will solve your problem. Watch the upcoming video. I will show a trick to make the sensor more sensitive. Hint: cut the plastic off the bottom, and expose the sensor itself.
Nice
It comes on Tronxy XY 2 pro printers and after few months starts failing and becomes a disaster. I had to remove it and replace it witha norma l switch and the printer is working fine, easier to level. At Tronxy they will never tell you is failing, they just say is miscalibrated, how can it be if you never moved it out of place?
It varies the sensing height based on the temperature of the probe itself. The varriance is huge: .5mm! I made a video on the subject: th-cam.com/video/_-fAMAnn3_s/w-d-xo.html
You connect direct the sensor to board? without Resistors?
Yes. As seen in the video at 0:36 the wires plug into the 3 pins on the board dedicated to the z-axis sensor/switch.
if you supply those sensors with 12v they will be more sensitive also , why is it such a big deal to give sensor a voltage within its rated range ?
The output from the sensor will burn up your control board. If you feed it 12V, it will output 12v. Thus you will need to do some complicated wiring to drop the signal down to 5v. It's easier to just run the sensor on 5v to begin with.
stick some of that aluminum tape to the bottom of the sensor and test again :)
The Tevo Tarantula comes with the SN04 sensor you showed. The detection distance is really small, mine won't detect at 0.93mm but will detect at 0.8mm against the stock Tevo bed. Bumping the sensor up to 12V never occurred to me, but that's probably a good idea to increase the sensor detection distance and reliability. Tech2C has also done tests of inductive sensors and found more or less the same results as Tom, I'm sure you've seen it.
How come I can't get the PNP to function on my hyper cube running Marlin 1.1.9... I have followed tom,s video and the fermware still will not work.
I don't know friend. I've stopped using Marlin in favor of the far superior RepRap Firmware and Duet WiFi control board. I suggest you do likewise. You will have far fewer frustrations.
I've tried sensors similar to the ones you show in this video and they did not work for me. However, they were cheap chinese sensors. Can you provide link to the specific sensors that you used so that i can try them? I'm interested in 6-36V type.
I purchased the 6-36v sensor from gearbest. It is labled "tronxy." I have ordered some other sensors from Ali-express for under $3 each, and I will test those more rigorously in a future video.
I went to gearbest website and used search engine to search for tronxy. All that came up were 3d printers, no parts. I then searched for "proximity" and "hall" and got all sorts of hits including cell phones and ceiling light fixtures, just nothing that resembles your sensors. You don't make it easy to reproduce your results by not providing links to specific product.
If you search for "Anet A8" and then at the bottom under "buy together & save" that is where I found it, but it looks like they are our of stock right now.
So basically this is a super compact "bltouch-without-touching" sensor? :D
i have a 10-30vcd sensor do i need to add resistors to it?
if you connect to 12 volt yes because you gonna resend 12 volt in the signal pin !
I tried to install 6-36 V Sensor on my Anet A8 and it didn't detect the Hotbed. When I tapped into 12V line the printer went crazy and some smoke came out of the LCD. I would NOT RECOMMEND this to anyone.
just i read this and but my c-lin pl-08N sensor to test bench, power to 12 V brown, blue to GND and black wire to signal and out come 12 V ok, i drop power 5volt and signal come out 4,4V yes this working. i no remeber were i orden this sensor, gearbest or banggood or were but not has connection blug to anet board, only 3 wires, must solden blugin to wires but i no know what connection blug and were can orden. i test whit aluminium+glass 3mm and working stay about 4-5mm air area. and metal work sure 8mm air area.i no sure what anet board have 5V "out" or under this, select 0 and 1 what volt need limit. but i remeber maybe limit has "0" 3,5 volt and "1" up this. must test then have connection blug future.
This sensor work 6-36V
Inductive sensors and Hall Effect sensors are two entirely different things ..... Inductive sensors sense changes in inductance of the exited coil inside. I have an 8mm range inductive sensor exactly like the one Thomas tested and I use it to sense nothing but a 250mm x 250mm PCB heater through a 2mm thick piece of glass with no pieces of tape to create an air gap and ruin the heat conductance from the heater to the bed Part of the reason this works better on a heated bed is a PCB heater is really nothing but a flat wound inductor, easily verifiable with an inductance meter and which is exited with a PWM signal and thus causes the inductance of the sensor coil to change easier and more reliably. It also acts as an antenna and throws out a fair amount of RFI
I've gone several months, almost 6 now, without a single failed first layer and as long as the hairspray remains uncontaminated the bottom is picture perfect every time
I don't recommend tape unless you tape the entire bottom because any air gap will kill you which is why you have to use heatsink grease in the first place, to remove any air gaps between the IC and the heatsink. If using just glass and a PCB heater I recommend locking it down on 10mm of cork with a gap cut for the thermistor wires to support the PCB and keep it tight against the glass plus it eliminates some of the bed wobble which means less artifacts and ghosting and being able to get clean prints at higher speeds
Sources for those sensors in the video?
I shy away from linking ebay sellers because the increased traffic causes them to increase their prices. You shouldn't pay more than $5.
10-30v sensor:
www.ebay.com/itm/ABS-5V-24V-DC-Hall-Effect-Sensor-Proximity-Switch-3-Wire-Normally-Open-W-LED-/131900296606?hash=item1eb5decd9e:g:oE0AAOSwImpXpbCf
6-36v Sensor on ali-express:
tinyurl.com/lt9w4jt
OK, I get your reasoning, thanks for sharing anyway.
I can now confirm that these are the same ones available from multiple sellers. That's what I wanted to know.
Thanks again. Interesting and helpful video.
Post links to the parts you talk about.
This one? www.ebay.com/itm/SN04-N-Approach-Detect-Inductive-Sensor-Proximity-Switch-DC-10-30V-/321697909691?hash=item4ae6b08bbb:g:pAcAAOSwWxNYuKwj
That one will work. The reason I don't post links is because the sellers get a lot of traffic, and then they raise their prices. This happened with my first 3D printing video where I linked to a complete printer kit. You are better off finding the parts yourself, and paying the lower price.
You will find a better deal by looking up "Inductive Proximity 10-30" on ebay, and then organize your search to be "Price + Shipping Lowest First."
Thanks.
What I had to flush my board in order to get it to work or can I just plug that bad boy up
uxcell SN04-N 3-Wire NPN NO 4mm Approach Detect Inductive Sensor Proximity Switch DC 10-30V 200 mA
it only works on a few sensor, I ordered one when I ordered my printer from gearbest. it Does NOT work with 5v.... my board put out 4.9v ... ( it is the 6-36v version)
Which exact sensor are you using? Is it Normally Open (NO) or Normally Closed (NC)? Does it work on 12v? What happens when you hold steel or iron up to the sensor, or a neodymium magnet? Does the light come? Are you sure you have it plugged in correctly? Sometimes, the plastic connectors between the sensor and the board will engage, but the wires will be switched. You must check the pinout diagram of your board, and make sure that your are correctly hooking up the sensor. Note: DO NOT plug it directly into 12v, and then into your board. You will fry your board. You must use resistors, or a voltage divider for 12v. As long as you are only using the 5v feed, you can switch the wires around without damaging your board. Best of luck, I think you can get it work. I have faith in you.
I have the Tronxy XY-08N NPN-ON I would get no light when using 5V, when using 12v the light comes on when close to something. I had to reduce the return voltage to below 5V. on first attempt I was getting 5.4v and it would not work... Everything was connected right I updated the firmware (to Skynet) 5V on mine does not work... I have seen other videos and they say it is a crap shoot if they work or no..
Thank you for your reply
And no it is not working yet.... dump me blow up my main board playing without unplugging. New motherboard came in today (was on back order for a month)
I bought the same model as yours and had to switch the black and the brown wire to make it work. It seemed to work fine til the heated bed rose to 50 C, the Tronxy XY-08N didn't detect the heated bed.
My only question is why is using 5V and avoiding 12V something people want to do? I mean maybe people with less DIY printers might be intimidated, but it seems to me that tapping into the 12V is super easy.
If you feed the sensor 12v it will then feed 12 volts into the sensor pin on your control board, which will fry the board. This can be corrected with resistors, but the wiring is easy to get wrong. In other words, it adds more steps and and you run the very real risk of ruining the printer. Why bother when 5v works fine?
Eddie currents
Why not just wait to turn on your heated bed until after you have set home
The bed heating up causes it to expand and bow slightly in the middle. This means that it will be at a higher place (in the middle) after heating up. If you measure before heating up, your measurement will no longer be correct once you begin printing on a heated bed.
Buy A Makergear M3 , no issues , no need for auto bed leveling due to design.
Ни хрена не понял, наверно по тому что не знаю английский.
Realistically this doesn’t work. It’s not consistent and repeatable. A ton of people have tried. Also you never calibrate your printer cold. It’s a waste of time.
Sorry, but the information here is not nearly enough to warrant any conclusions. At the very least use the firmware's capability to make repeated measurements and see if the std dev varies with applied voltage.
Just throwing out the idea that the electrical conductivity of Al varies with temperature without quantifying it is very misleading. In fact, it varies only a small amount with temperature in the ranges you'd expect to operate a printer. Conductivity is only one part of the equation, so to speak; magnetic permeability is probably more important and, again, over these temperatures it's not going to matter.
What is most important is how a proximity sensor actually works, which is nicely described here - automation-insights.blog/2014/03/05/basic-operating-principle-of-an-inductive-proximity-sensor/
It is a kind of electromagnet, which means the field strength varies inversely with distance cubed (r^3). The field strength is a function of the voltage and current. The stronger the field, the greater the distance before the sensor trips. So, if the voltage is less than the recommended, yes, it will work, but the effective distance will be much smaller than what the datasheet will indicate is nominal. If the distance is too small, and the std deviation too large, the accuracy and precision will likely be poor. It might be so poor that the hot end (which is lower than the sensor) will crash into the bed; can't say it won't, right?
I agree it would be nice if Tom tested one of these with his rig, but to take into account your points he would have to have temperature chamber and variable supply. That's a lot of degrees of freedom.
Please don't suggest that just because the sensor can work under-voltage that it's just fine for people to use.
5:16 -zzzzzzzzzzzzz .........good video otherwise very helpful...
They are both crap there are some test videos out there..
You mean like this one? th-cam.com/video/_-fAMAnn3_s/w-d-xo.html