I highly recommend you jump to voltlog #177 where I show you how to install a bltouch sensor which is the best option in my opinion. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-7Upc5HpE9gE.html
What qas the prooblems that you have with the sensor, in the video you have found a right sensor, why you dont have used the th3d firmware? by the way nice video and you have here a subscriber
The inductive version was a better choice, just reduce the glass thickness from 4mm to 2mm. The inductive versions are prone to error with minor changes in humidity and temperature, a second print of a hot bed will be way off...The bltouch is probably the best option but if you want accuracy, loose the glass and go with the inductive sensor ;)
Hey, while watching your video I grabbed my SN04-N switch (blue one), and filed sensor side so it can reach my aluminum plate through 3mm glass. After some careful finishing with exacta and putting protection layer of Kapton tape it senses >1mm over my glass. Just telling that they are not useless, just designed for different application.
For sure they have a use! Some people are printing directly on the aluminium bed with a thin insulation layer so then it works great. I just prefer the bltouch for printing on the glass.
Marcos Ramirez I have one on my cheap i3 and it works flawlessly since I've installed it... Had to make a decision weather I switch the buildplate from glass to metal and go with an inductive probe or keep the bed.and choose a capacitive sensor. I don't regret to have chosen the later...
Aaron Rogers To each their own friend. I personally like buildtak on bare metal with an inductive probe. 😄 Say any chance you might know how to connect TMC2130 drivers to an MKS GEN 1.4 board. I haven’t been able to find a proper walk through.
Marcos Ramirez nor really thinking about upgrading to the trinamic drivers myself. I think Tom sanladerer or maybe Marco reps had a general purpose tutorial to install this drivers on any board - bypassing spi setup from Marlin by using an additional Arduino nano...
Specifics matter. The 3D printers at one local maker space are all in insulated boxes with active temperature control. For those, the temperature will always be stable. If I were to set up a printer in my shed, it would see temperature swings of 20 degrees C. How much does the measurement change per degree C? How much across the range of likely humidity values? Changing humidity causes problems in the woodshop, so the humidity is also monitored.
A repeatability test of multiple approaches with the "winner" would have been nice (ideally using a dial indicator, but a thickness gauge or even a paper strip works too), especially if repeated on a few different days...
The practical approach, just using the sensor and printing stuff will give me a good indication of whether this is a good idea or not. I should have some more info on that in a few weeks. Right now I just printed the sensor mount and will be trying it soon.
The inductive sensor's rated distances use iron as the inductive sheet. Aluminum has ~40% the rated distance, so a 8mm sensor will only trigger with 3.5-4mm distance from an aluminum bed.
Prusa gets around this by using a spring steel sheet, which gives nearly 80-90% of the rated distance (due to the carbon and other added alloy metals, it isn't quite 100%).
Thanks for a video I have installed capacitance sensor (LJC18A3-H-Z/BX 1-10mm) but i found it not much accurate each time i need to adjust z prob offset. I read on internet this sensor change reading due to humitdy and temprator change so each time it come with diffrent number I am thinking to move back to manual meshbed leveling.
Capacitive sensors just create more problems than manual leveling because of their unreliability. Any industrial engineer would tell you they are the wrong type of sensor for this application, and could very well be the worst choice. You need to make sure there isn't too much of an insulating layer, or choose a more expensive sensor, it's up to you. Mine was cheap and works fine over a glass bed.
In my experience capacitive sensors are not consistent. I had much better luck with a inductive sensor that could sense through the glass sheet. It worked much better but wasn't a good option either. I ended up using a Triangle Labs BL Touch clone. It's by far the best option and is cheap.
thanks for the hint. I will have to recompile & upload the firmware to test that because it seems I had Z_MIN_PROBE_REPEATABILITY_TEST disabled in config.
How can I increase the distance and keep reading? With the small screw on top? I can barely rise it a couple mms. before stops reading. I can´t find anything related to this screw.
But the glass surface is very plane and it is very good property! I am thinking now could we cover the glass plate (very plane!) with a thin film like that one for printing with laser printer on it. If so, we could to glue some pieces of Al foil under it. And if they are glued accurately, we could use all the functionality of Prussia's autocalibration with such a combined plate. I think the best choice will be something like an adhesive sheet of transparent film for laser printing on it and very carefully cutted and glued pieces of Al scotch tape or even a simple foil. But it needs an experimental proving and very good accuracy. The second way is to cover a glass witn an aluminium paint via some stencil. The third way is to use a mirror with almostly etched metal layer which is leaved only in defined places. And the mask for the etching may be transferred from the paper with old good laser printed pattern by something like an iron. Or even with a photoresistive film. Or we can get a sheet of copper foil, transfer the mask with an iron on it. After that we can glue the foil on th glass and etch it. And finally cover it with a laser film. This difficulties can give us not just a bed auto-leveling, but full 3D geometry calibration and correction like Prussia developers did... Sweet DIY dreams :)
Hi! Cool video My 3D printer is old and uses an enclosed electronic board (Gen7Br) So, I would like to use the LJ12A3-4-Z / BX proximity sensor only with Arduino. Making an analog reading to measure distance through the Monitor Serial, can you help me with the code?
How about optical sensors, they at least are more repeatable on the same surface, and will work with most smooth surfaces. visible light ones are the easiest to set up, though the most common are IR ones simply because you do not see the spot, but a cheap cellphone camera will generally be able to see the IR beam spot well enough.
I'm thinking there must be good reasons why the capacitive/inductive proximity sensors are widely used in the industrial applications. For me it was cheap and easy to get one of these sensors, while trying to work with an optical sensor it feels like there is much more that could potentially not work as expected.
Industrially simply because they are not too much worried about oil, grease and dirt on the front of the sensor, but in clean areas the optical ones are the best because the repeatability of the sensing is almost as good as a mechanical switch for limits.
Proximity sensors reduce in range as they age, just keep that in mind to allow for adjustments. I do not know how repeatable the sensing distance is. Maybe a micro switch would be a better option, I don’t know if anyone has tried them.
These proximity sensors were the cheapest option and possibly simplest to use. If I have to buy some expensive switch like BL touch I think I'll just stick with manual adjustment.
That is a PCB ruler, I have a bunch of them from different places. Most of them are offered for free when you purchase something from a store like welectron.com for example.
could you show me how to conect it and get the distance. How do you do it electronicly. what do you need for 3d printer. Can i hook it up to my 1.4 ramps motherboard. LJC18A3-H-Z/BX 1-10mm Capaciteit Proximity Sensor Switch NPN GEEN DC Please show me more on how to do it would be verry help ful thx
Wouldn't it be much easier to use a steel sliding rod with several mm stroke mounted beneath the sensor so that it will lift towards the sensor as the z axis lowers?
Hy ! You forgot to comment on the ability of your sensor to repeat the same measurement. This will be the measurement error. And how much is this in mm? Is this enough to print with quality?
about the capacitive sensor, i've read some links about some problems it can trigger, about different voltages or overvoltages... sorry, link in italian, but look at images or at links inside it for english references
look at links, the italian guy uses a relay, in links others use optocoupler or some voltage divider to avoid some return feedback to the electronic board, or the fact that the LJC18A3-H-Z/BX is NPN and triggers if HIGH while the BY model being PNP and triggering when LOW, so for the BX model was needed something to protect the 5v board if sensor powered with 12v... sorry, is something seen 2 years ago and i can't remember the exact problem, and i'm not so fond in electronic to be more precise...
I have an optocoupler isolating the sensor from the motherboard, I wouldn't use a relay for this type of circuit don't know what that guy was thinking but my circuit is working just fine :-)
would the sn04 work with the wambam flexplate system, i know its springsteel but its the pei stuck down to it im curious about, looking into getting a sensor for bed leveling
You must be doing it wrong, your sensing distances are too short across all sensors. How are you powering the sensors? SN04-N is fine, and it will trigger at 5mm, but it really, really needs a good bit of supply voltage! Don't feed it off 5V , else it will trigger too narrowly and will be super unreliable! 12V are good enough for it. Use pullup on your controller board, and then connect the sensor through a diode, so that the sensor can only pull the board pin down, not up
I am powering from 12V, glass is 4mm thick, and build plate is aluminium so with a 4mm inductive sensor designed to work with ferrous materials it's only normal to get the results I am getting.
@@voltlog your inductive sensor should be 8mm. not 4mm you have a different part no in the overlay and description. LJ18A3-8-Z/BX 8mm NPN NO Inductive Sensor is supposed to be good for 4mm glass but your video worries me. unless it is not powered @ 12V as it should
Capacitive sensors are way over rated and a 3D printer fad. If you level the bed, the standard micro switch works very well. If your bed is not leveled, then it can be of some use. Then again, why not just level the bed?
It's an interesting idea that I'm sure many 3d printer owners are thinking about. I wanted to test and see for myself if it's usable or not. Because I had bad luck with the inductive sensor that I got first time I didn't want to abandon the project so I got the capacitive one instead and I'm gonna try to make it work.
The magnetic field from the heated bed, the heat itself, humidity and other variables will make these sensors unpredictable. The one that came with my Anet A6 printer, is still in its packaging. It is the blue useless one you dismissed at the very beginning.
I find that my endstop behaviour is not all that consistent. There is apparently some minute movement in the bed presumably with heating and cooling, and it's not large enough to cause a significant skew of the bed, but large enough that i can find the nozzle scraping the glass one time and the prints not sticking down another time. There is also some systematic movement of the second Z axis every time the printer is turned off, so i'll do separate Z drivers and dual Z endstops before i do a bed level sensor. Luckily i have plenty of endstop/sensor points on my board that i can use. And of course it will be inductive, not capacitive. Humidity doesn't matter. Heat is in fact completely predictable - first the head is parked, then bed is heated to target temperature, and finally the probing is performed at known temperature! Reputable printers like Prusa Original have no problems with an inductive sensor at all. Also there will be a steel sheet instead of glass.
