Design and build a spin coater
ฝัง
- เผยแพร่เมื่อ 8 ก.พ. 2025
- I describe how I designed and built a spincoater, and how you can use a similar motor control system in your own projects.
Firmware with I2C and Arduino library
www.blueroboti...
Compile instructions
lunarflow.com/i...
Flash tool to download firmware (and other helpful features)
lazyzero.de/en/...
Arduino code and assembly for motor ESC
drive.google.c...
en.wikipedia.o...
Some very cool tidbits on this. You aren't kidding about RC being a whole new world. That hobby has been around forever but as soon as LiPo batteries and cheap brushless motors became easily available, it seems like things exploded.
+Practical Engineering I have a 15 year old 8 channel Futaba 72 MHz radio that cost more than my first plane (with Saito 4 stroke engine!) and my FrSky Taranis was a whopping 275, and supports 32+ channels, depending on how you set up mixing. It's gone positively nuclear.
Depending on the quality and charging circuits for the LiPo, it may have literally exploded!
I hope the LiPos didn’t explode, though...
I guess im asking randomly but does any of you know of a method to log back into an instagram account..?
I somehow lost the password. I would love any tricks you can offer me.
@Hugh Zev Instablaster :)
Very nice build! As someone who really needs to work on improving my knowledge of electronics I appreciate your thoroughness in describing the how and why of various components.
+NightHawkInLight you read all the comments.. OK
do you watch and comment on all youtube videos too??
+NightHawkInLight Thanks! Let me know if you ever have a video topic suggestion.
@@AppliedScience Ben, maybe you still read comments here (I hope!). There is interesting idea called callutron for isotope separation, similar to mass spectrometry.
@@AppliedScience Also second ideas: electroactive polymers for robotic actuators
I really appreciate how you share not just your design, but also how you arrived at your design, including options you considered then rejected, and options you tried before rejecting. I don't expect to ever have use for a spin coater myself, but I learn enough from all of your videos that they are always worth watching. Thank you!
Now if you add a LED blinking always at the same time at each rotation for stroboscopic effect, you can even see how the setup looks like and the layer thins while in progress. Maybe not very useful while in high-speed mode, but while dispensing the liquid manually (you even can make it have a slightly lower frequency to have it look like rotating slowly - that way you have a natural indicator to avoid hitting the poking-out screw with your pipette).
+Henner Zeller Put the LED on the arduino and have it strobe in sync with whatever RPM you have the rotor spinning at so no matter what speed it's at it'll always appear stationary/slowly rotating.
+Henner Zeller
That definitely needs to be in version 2.
+Henner Zeller That's a cool idea.
+Henner Zeller Enable MOTOR_DEBUG to get such a strobe in real-time (from MOSI / PB3 by default), but it will have a 50% duty and you'll have to divide by the number of motor pole pairs to get a whole revolution.
DanieleGiorgino Perhaps you meant version 3
zener+mosfet ( or bipolar transistor) would be a more stable high-power clamp
Great use of RC gear!
Getting into the hobby has revealed a lot of methods of accomplishing tricky things cheaply/easily, of which I would otherwise would have been unaware.
Also, motor braking isn't just for cars. It's used on fixed wing aircraft to increase gliding performance, as an auto rotating propeller creates a lot more drag than a stopped prop (oddly enough).
Cheers!
I love how you are able to convey a lot of information per unit of time. There are youtubers that in 20 min basically say nothing and I hate it.
When you start talking about rc stuff I can really start feeling just how new you are to the hobby, great video nonetheless.
I'm guessing someone has already mentioned this but that weird little "adjustable zener" circuit reminds me of the dump resistors on superconductors. I think all you are missing is a load to keep the fet from popping if it ever turns all the way on. Like a piece of nicrome wire between the FET and Vcc as a current limiting resistor.
Seeing a new upload from this channel just get the inner happiness out of me
Nice work Ben! I always enjoy your video's, Big Thumbs up.
Great video as always, thanks Ben.
A good way to make an accurate shunt regulator is with a TL431 precision programmable reference and an PNP power transistor. They are cheap and very easy to use and temperature compensated. Often they can be recycled from an old ATX PSU.
It will have a much sharper turn on than the MOSFET or a zener circuit and will be repeatable to the accuracy of your resistors, so suitable for a production environment. Your MOSFET circuit will depend on the gate threshold which varies from device to device and with temperature.
RIGHT! I think this is the best solution.
And now 2 years later the RC world is insane around this stuff. Programming the controller and customising it is really easy. Once you understand all the weird lingo which is spoken.
Great project! I don't have a need for such a thing. But I like your tinkering and overall problem solving mindset.
I always love the ending "see you next time, baaaaaai" :') great video Ben.
Great explanation of all of the details, and neat result!
By the way, you can set MIN_DUTY lower (I usually set around 8) to get closer to the lower limits of what can be software-PWM'd. This can extend the range between minimum and maximum speed, or if you would like to just push out the low end even with a higher supply voltage (or higher KV motor).
Yup it's a lovely world for electric RC people. I've built many airplanes of my own design for well under $30 using hobby king parts and dollar tree foam board. Nice video.
You're back! I missed you. I mean your videos and science and stuff.
I just wanted to let you know that I really enjoy your videos :) I think every thing you do is awesome and even though I don't understand half of any of it, you do a very well job at explaining it. Keep up the good work!
First video found looking to make a coater for our lab. Very informative starting point. Very helpful. Thank you.
Great video , glad to see a new video from you😃
So true
Great tip with the circuit board. Wow your spin coater is much better than the lousy commercial one I was using when working in applied physics LOL!
18:10 That slow zoom on the spinning disc in combination with your voice was quite hypnotizing :-)
+Siegfried Müller Funny enough, the slow zoom was caused by the camera lens zooming due to gravity pulling on the movable part of the lens. I have two identical camera setups, and the lens on this one is a little looser, and it will not hold zoom setting when vertical.
This looks like a good idea for a diy centrifuge
Nice idea to add that mosfet for clamping the regenerated energy. It should have no problem except with temperature variations that will affect your threshold voltage.
is it dishwasher safe? also could you chuck up bread and use it to spin coat peanut butter and jelly because that could save on washing knives..
I have actually been wondering for months how I could control brushless motors with I2C :D Thank you!
Your "programmable zener" is amazing.
Just throw a big heatsink on the FET and use a big resistor to take up the current.
That was awesome, start to finish. First time I've used the replay button on TH-cam.
Multirotors actually do benefit from regenerative rotor braking. It helps the craft drop quickly and not feel so "floaty" during acrobatic maneuvers.
thank you for just another contribution to showing folks the beautiy of science
the new one looks like a manufacturer quality one, nice.
I took the same route of logic Ben. Hobby bldc controllers for sub 10$ w/ open src fw. For the motor, take a look at brushless outrunners, high torque, low top rpm. (5-6k rpm max). A prop motor, not a quadcopter motor. Great information in this vid btw, tnx!
+Hedley Rainnie Cool. What is your motor project?
+Applied Science Ben, a single motor quadcopter with variable pitch props. 4 geared driveshafts out to the props. One beefy motor in the centre.
Pleasant surprise to see a new video today, thanks!
I’ve seen commercial motor drivers that used transient voltage suppressors, which are basically two Zener diodes back to back, along with a series resistor for active shunts.
In a typical use case, they might have a 48VDC power supply with a suppressor that starts conducting at 52-56V to stay under the 63V rating on the electrolytic capacitors.
Great video again!
I understand like 50% but still amazing to watch!
Nice job on the video! Keep up the good work man!
That's absolutely astonishing! Amazing job!
The ABS cement is really strong stuff. I used it for years on drain lines.
ABS cement actually welds the abs together by dissolving them together.
Many of the motors for RC cars have hall effect sensors that lett the speed controller know the motor position and therefore can run at really slow speeds and give precise RPM readings
you just helped me solve so many 3d printer problems
i've been thinking of getting a small motor and making a portable universal friction-drive e-bike kit i could snap onto bicycles when i find them. never knew it was that easy to control them though.
Awesome build man and thanks for sharing! I will be looking to construct something similar!
Simply amazing! Eagerly waiting to see in what project you will use this. Better DIY LCD? Transistors?
One thing though, you say in the video you will show us the HDD motor connected to a debug port to demonstrate how fast it is able to change speeds? It didn't work anymore or did the footage just "fall off the editing table". :)
+Macovei Vlad I ended up not using the HDD motor. In the video's last couple minutes you can see the quadcopter motor changes speeds as quickly as it can.
Your n-FET plus potentiometer will act as a voltage clamp, but with a "very soft knee", without further calculation I'd say that its resistance will only decrease very gently. I doubt that one can dump much energy in that device. Also: No heatsink.
With a classical motor controller, you'd employ something called a "Breaking Resistor". Industrial motor controllers (commonly called VFD - variable frequency drive) will have separate clamps to which connect pretty large, mainly air-cooled, wire-wound resistors. But that's of course devices with many kW of power intake. www.google.com/search?q=vfd+braking+resistor&tbm=isch
For your device, probably a smaller power-zener on a heatsink would provide more efficient breaking.
braking*
Man I love your videos, you explore so many subjects! I'd love to be half as cool as you when I'm your age!
Stunning engineering as always, you rock sir!
Very cool. Now I want to come up with something to spin-coat!
Thanks.
Sweet, an applied science topic I can wrap my head around.
FUCK YES!!!! FUCK YESS!!!!! I've been telling you to do semi conductors for a while now!!!!! :D!!!! YAAA!
Thanks for the programmable zener trick. That is awesome.
I think that for the best coating, you should take into account the liquid's surface tension and viscosity because it might stick to the edges of the glass and be thinner in the middle and thicker in the edges
Anoter marvelous video fella, and your didactics is flawless as usuall!
the breaking mode on model e-motor controllers is important for electric gliders. they got folding props, which you want to stop spinning when you throttle to zero. because that is when the prop folds and causes less air resistance and better gliding.
I like planes...
I have never understood why everyone calls the quadcopter motors "DC Brushless" when really they are just 3-phase AC brushless motors. That's why they need a "speed controller" to run on DC power, it is really just a variable frequency 3 phase inverter.
With that said, the modern market for these devices certainly does mean that we now have access to an insane amount of well controlled rotational power output for very little money.
+Jensen567 It's true that the choice of labels is unfortunate, but there is an important distinction: "AC motors" are typically understood to be induction motors ie their rotor contains only conductors, no magnets, and the force is created by induced current flow. A permanent magnet DC motor has brushes to shift the current flow which produces torque on the permanent magnets. A brushless motor could be induction or permanent magnet, so we need another label. It might be best to call them brushless permanent magnet motors, but BLDC for brushless DC became popular when the controllers were integrated (eg 12V computer fans)
+Applied Science Is it true that a brushless motor could be purely induction based though? How would one power the coils on the rotor without brushes for more than a few revolutions? I did a quick search and did not turn up any designs that used an inductive rotor without brushes. So really, I think brushless would imply a permanent magnet rotor.
+Jensen567 Induction motors generally don't have brushes. They work by magnetically inducing a current flow in the rotor, so no need for brushes. However it's true that the "brushless" designation has come to mean "not induction" by convention.
+Jensen567 the AC asincronus induction motor , wich is brushless, is the most commun used in various aplications that need reliability but not lightweight, they have so called shortcircuited rotors or 'squirel cage' the stator incuces a current in a rotor winding that is in short circuit with all the other rotor windings there is an amplification of the current/voltage drop that generates faze shift and so on... , by the way try to short circuit the rotor of an universal motor to see what happens
+Jensen567
They are called BLDC, because, well, they are DC motors with external commutation. You can run them on 3-phase, but they will most likely slip as soon as you apply a varying load.
The speed of a BLDC motor is *not determined* by the phase switching speed! The speed is controlled by *changing the voltage* to the coil (almost exclusively done via PWM), like with a brushed DC motor. The time, when the next "phase" is activated, depends on the position of the rotor, as it would be in a brushed motor by the rotation of the commutator.
Not sure if this was your problem, but the pulse wire should work with your board, you just have to detach or otherwise not connect the 5v power. Connecting the signal and ground will work nearly every arduino.
8:17 The 3 wire motors can be wired for Delta or WYE. A Delta termination will have a kV 1.73x higher than the same motor & winding terminated WYE. For optimizing RPM per V (kv), the motor can be rewound increasing or decreasing the number of turns per pole and terminating in the configuration that gives the closest to the desired kv.
Is the platter balanced?
Thanks to you I delved into sputter coating and now need to do some spin coating. No idea of this Arduino thingy, one of the links isn't working but I'm sure I'll figure something out. Being able to run a program, slow speed followed by a timed faster spin would be great. :)
Yea I learned the PVC pipe cement trick when superglue wouldn't hold my headphones together, figured it was worth a shot as I had glued a plastic drain stop together that I thought would never hold.
They are still together, had them for years.
That is a very neat machine! Have you looked at the coating with a microscope to see, how even you can get it, as that might also be interesting.
i like your mosfet clamping method
Yea probably should put some kind of door or shield type thing over the opening.
+BioNerd Yeah, at 4k rpm that rotor looks like it could be pretty dangerous, let alone the glass slide
+BioNerd Useless worry. You should wear a helmet the second you leave your house.
dus toin It's all fun and games until you have shattered chemically coated glass rocketing towards your eyeballs at hundreds of miles per hour.
+dus “dustoin1386” toin I'd feel fine with Ben's design because I know he did it properly, but if you're recreating this and don't get the proper mass distribution with those screws (or the rotor itself), I wouldn't trust it.
+BioNerd A possibility of fast-flying pieces of glass don't seem all that fun to me either. A shield might be a good idea just in case.
Concerning welding the ABS: Simple Acetone works like a charm ;)
the Y config works fine with an ESC(common). Just feed it 12 volts or even 24V, and leave the COM. some ESC can be programed in depth, and you can make the accept that COM wire. I believe that you can just hook the COM wire onto a virtual ground made using two caps? yeah that's pretty complex since you can just simply feed it with an 3 wire ESC, and it works!
If you have some ABS or PS/HIPS/SB, you can simply dissolve it in acetone to make your own ABS cement. Doesn't need much acetone at all, the less you leave in there, the lower the chance of it warping the substrate.
Thanks for the video and info.
Any chance you have measured the run out of the shaft? These motors could be great as high speed spindles.
Also, does the temperature jump up? I wonder if the motor assumes it will get cooling from the propeller.
For ABS and PVC welding, I've had great success with MEK (methyl ethyl ketone).
Servo pulses range from 1 ms to 2 ms, with 1.5 ms being mid-range, or neutral for control surfaces on R/C aircraft.
I use acetone/abs mixture to glue ABS plastic sometimes. probably not as strong...but it works.
btw you can use these ESC controllers with some harddrive motors. works pretty well (if they have 3 pins)
I would like to see a infrared chemical analyzer, spectrometer I believe. Also a chemical analyzer using mass deflection of a ionized bead and Cl3D, as the solvent.
Another great video Ben!
Imperfect timing, I just recently finished my own spin coater. th-cam.com/video/hdgkTBljyfE/w-d-xo.html
I'm using a Maxon motor with sensors and their controller after playing with hard drive motors and RC controllers. I thought I might need slow speed control (nope) so that's why I went that way. The vacuum chuck is contains a porous aluminium (Metapor) insert and is itself held by vacuum with an o-ring seal for quick change. The motor has a rotating shaft, I drilled it but had to put the rotating body of the motor in a box for the vacuum connection, I also made a crummy shaft seal as the bearings were obviously leaky. The bowl is a water bowl for pets, I have a smaller plastic bowl with no bottom and a lip (formed with heat) to catch the resist and make cleaning easy. I don't bother ramping during but instead just drip on at full speed, it is the least wasteful method and seems to work.
+littlestworkshop Nice work! Thanks for the details.
+Applied Science save you self some money and use straight acetone to solvent weld ABS dont use nail polish remover though as it has other oils and aditives
+Applied Science Thanks! Seems the least I can do considering all I learn from your videos. I'm building a dip coater now for double sided coating, much easier to make but dust is harder to avoid.
+chaos I've had less than ideal results with straight acetone. It seems to act on the styrene far quicker than the acrylic or butadiene (not entirely sure it reacts w/butadiene at all) so you wind up with a crazed, weak joint. At least with my 3D prints. I've used the pipe cement before, and it's a pretty bulletproof bond.
Slurry is also pretty good - chopped ABS (failed prints/raft+skirt bits) dissolved in acetone - works pretty well as a gap filler and adhesive, but I find it's a bit of a pain to mix up vs. just buying something that works, for the amount of assembly I wind up doing anyhow.
Rion Motley it comes down to pre saturation man, your trying to bond different plastics that dissolve at different rates. try brushing the ABS surface with a few coats of acetone before bonding to give it more time. and when it comes to bonding ABS to ABS acetone works the best chemically again give the surface time to soften up before pressing it up to the other part. Ive been bonding all variety of plastics for a long time so ive got some industry experience with, fixing plastic car bumpers is more technical then you would think
Your braking circuit resembles a 'crowbar' circuit, I believe... used in big audio amplifiers, to prevent DC from appearing across the speaker terminals when a fault happens.
Great video as always! I'm curious how you would tackle measuring how thick and how even your coating is in your shop..
A system to control many high speed motors with precision? You're not planning on one-upping the nuclear boyscout are you?
Very cool project, I never would have thought to use RC parts for anything but RC.
a HDD is exactly what I was hoping to hear. I have been looking for something to do with my recently dead drive, this project is going to be fun //then again... an excuse to be a new motor is temping.
You can reflash the ESC with firmware to enable and adjust braking.
+Camaro Rick USBasp + Atmel chip flash tool is what I used with BLHeli software
How about trying to coat it with graphene layer!!!
A couple of years ago there were scientists that coated a spinning regular CD with graphene layer, potentially turning it to a super capacitor. That would be really cool if you could make this, and also show how it looks on your Electron Microscope and investigate its properties!
Very nice build. Thanks for sharing. :)
I always wonder why the coating doesn't turn out uneven, since the centrifugal forces must be much lower near the center?
I'd add a MOSFET and load dump resistor to dissipate the regen energy, if necessary PWM the gate drive to control the energy flow.
cooooooooooool !! Nice work Ben!
Looks like there's some vibration at the higher RPMs? Maybe if you move those two side brass screws so they are at the same radius as the opposing one that would balance the system?
Do you know Dr. Garret Moddel? Could you please build his cell with this device? He practically combined two known phenomena/principles together to get usable energy from the thermal background radiation. He combined a Quantum Tunneling layer with a Casimir cavitation layer in such a way that you get e net flow in one direction. Is there a way to get that cell done with household items and a DIY spin coater 🤔?
Great video, as always!
I know there is a way to balance that spinplate, but looking at the rpm readout, I think AvE did a video on that.
Great tip on how to mount the components! Where do you get the big chunks like that?
I wonder if a spindle for a hobby CNC would work, those can be had for $20-30 and likely have better torque (useful if your rotor or what you're spinning on it is heavier)
Do you think you could make large dichroic filters using a similar setup or do you need to use a sputterer for that?
Really nice video. Can you tell the difference in quality with a professional version?
neat build!
How would you use a small quadcopter controller to control larger motors to make an oversized quadcopter?
Very nice video! However, but I have a question regarding how you acquired the correct rpm speed from the ESC?
Hey really like your series of videos. I work as a mechanical engineer in the high tech industry and want to make DOE's and CGH's at home. Could a spin coater like yours control photoresist layer thicknesses within 30nm? There are all kinds of fancy measurement devices to measure the thickness of these coating like atomic force microscopes. But is there any cheap solution to measurethis? Chromatic convocal sensors can do this, but that is not a cheap solution.
you can get a dedicated program card for ESCs
Could you try and make a really thin diaphragm like what's used in microphones? And then maybe testing the quality of it used in a microphone?
Is there are reason to go trough all the effort of reprogramming just to spin the motor the other way?
You could use a relay to switch the leads around, reverse the motor or if that isn't fast enough some (high power) transistors.
Dont use a FET with a potentiometer. It is better to just use a Zener diode rated at that voltage +1-3V.
Power FETs have a much higher response time than Zener diodes. Usually the feedback from slowing a motor will produce short-time spikes in excess of 20-600V depending on the windings, but the current is very very small, unless you are driving massive AC motors. d(I)/dt = current/0 = infinity.
+Eternia Kerbal Capacitance on the bus can easily handle the spikes you are talking about, although a zener would be much more accurate and less temperature dependent
He's got that line going back to whatever the off the shelf supply is, it likely has its own zeners and protection from minor inductive spikes also.
You are a genius!
How uniform is the coating between the middle of the slide and the outside considering they are moving at different speeds?