Making PCB Motor v4
ฝัง
- เผยแพร่เมื่อ 27 เม.ย. 2024
- These are all the new improvements with PCB Motor version 4!
Get a free trial of Altium Designer - carlbugeja.com/altium
PCB Manufacturer - www.pcbway.com
Help me make more projects by supporting me on Patreon - / carlbugeja
My Electronic Kits - flexar.io
This project is Open-Source and licensed with Creative Commons Attribution Share Alike 4.0 International License
Open Source Files - github.com/CarlBugeja/PCB-Mot...
Time Makers:
00:00 - Intro
00:36 - PCB Design
01:39 - Unboxing
02:03 - Initial Testing
03:13 - Iron Rotor
05:24 - Shaft Mount
06:02 - Torque Testing
07:01 - Speed Testing
07:30 - Thermal Testing
08:24 - Onboard ESC
09:35 - Conclusion
Cameras/Tools I use (affiliate links):
Main Camera - amzn.to/30tRgY7
Second Camera - amzn.to/2ZLm2wy
Light Setup - amzn.to/3jtqckA
Heating Plate - amzn.to/36qYzD4
Ultrasonic Cleaner - amzn.to/3nfUGHY
Magnifying Lamp - amzn.to/3jrcaQv
Vernier Caliper - amzn.to/3eTju3I
Multi-meter - amzn.to/3hq6B2R
Digital Milligram Scale - amzn.to/30HIw1X
Digital Microscope - amzn.to/3hyzoFt
3D Printer - shop.prusa3d.com/#a_aid=0550494M
LCR Meter - cutt.ly/Wlmg0b0
Social Media:
TH-cam Channel - / carlbugeja
Instagram - / bugeja.carl
Twitter - / bugejacarl
Facebook - / bugeja.carl
Linkedin - / carl-bugeja-0b922a135
Website - carlbugeja.com
Music:
Sun Machine One - Loopop (TH-cam Audio library)
DC Love Go-Go - Silent Partner (TH-cam Audio library)
Brontosaurus - Topher Mohr and Alex Elena (TH-cam Audio library)
Here We Go Again - Silent Partner (TH-cam Audio library)
#electronics #motor - วิทยาศาสตร์และเทคโนโลยี
Your tenacity is commendable, the amount of iterations you've gone through on all these PCB coil projects is simply staggering to me, I usually give up after revision 3.
If theres a will, theres a way. And Carl has clearly shown that he has the will to make a PCB motor that works pretty well.
Make one mini fan cooler
@@timteecvhn it can give speed but no torque.. because of the coil limitations in that pcb..But it can be used in many places...but if you put a heavy proppler on it, the rpm will go down
@@b-beluga4510 maybe we can use some gears to increase torque
@@pratiklondhe5167 A harmonic drive, maybe?
Lots of good advice in the comments: hitting permanent magnets reduces their strength, use coolant when machining iron, a simple hand vice to put in the bearings would help prevent damaging the outer race,
True
Also the way he mounted the magnets in the steel part is shorting them, it would be better to have a magnetic plate as a flux carrier and sandwich the four holes on top made from pcb material so N-S of the magnets is not connected via magnetically conductive material.
@ Carl Bugeja & Robert F
You are conflating Iron and Steel, you should use coolant with STEEL, Iron is self lubricating because of its high carbon content which acts like graphite during the cutting process. In machine Iron, be prepared for a tremendous mess, the carbon dust will coat everything in a certain distance
One of the great things about online forums is that you can get a lot of good advice from people who have been there, done that. Recently, someone asked how to make a PCB motor and got a lot of great responses.
Some people suggested using permanent magnets, but others warned that hitting them with a machining bit reduces their strength. It's a good idea to use coolant when machining iron because it helps to keep the heat down and prevents the workpiece from warping.
also a good esc that keeps the motor pressing rather than jump around and hit the scale like that, not a diy arduino based one..
adding small air vanes to the side of your stater to encourage air movement though the PCB with holes in it might help control the thermals. A lot of high power motors use this design.
Good call!
To encourage air movement through the PCB with holes, small air vanes can be added to the sides of your stater. This will create a more efficient system by increasing the amount of air that is able to flow through the PCB. Additionally, this will also help cool the motor by drawing in cooler air from outside the enclosure.
Although I have absolutely no need for a PCB motor myself, it’s great to see your thinking process and seeing a solution evolve. Brilliant!
Thank you very much!
@@CarlBugeja ...Hammering bearings shouldn't be done. Try gently pressing them...Good job!!! 🇺🇸 😎👍☕
I like that you labelled the signal and ground and still got them mixed up. Seems like something I'd do tbh.
I have worked a little with a company that is makes PCB motors commercially. The primary engineering challenge is heat removal, the method you are using is pretty much the way to do it. The next step is to add an aluminum housing as a much larger heatsink. The rotor is typically on both sides on the PCB. Two bearings further apart assists in removing wobble of the rotor. The next engineering challenge is that you get a very low copper to rotor ratio. One way to get around this is making a motor with a very large outer diameter, the other is to sandwich multiple together. By having a stack, you can utilize the same rotor magnet on both sides on a sandwiched PCB. From a power/efficiency compared to weight, these motors are not competitive with traditional wound motors, however I see a future for low cost designs where size and weight is not as critical. Using an extruded heatsink ring that mounts to the end would allow the assembly to easily scale up or down my increasing length with common componentry sandwiched together.
For calculating torque, you should instead make an inertial dynamometer. Basically a fly wheel that you attach to and spin up with the motor and monitor RPM, based on it's angular acceleration and the fly wheel's known polar moment of inertia, you can build a torque vs. speed curve that's much more accurate. Ideally this should be done using 2 fly wheels of different mass so that you can calibrate out the rotor's inertia, but if the rotor's inertia is much less than the flywheel or the accuracy needed is low, then this can be done with only 1
Great option, I was thinking the same thing! Was kinda flabbergasted that he said "the second method was less accurate", since the first method also takes in the acceleration by the gravitational forces. Basically the same thing as hitting one of those carnival attractions, the one where you strike with a hammer trying to make a bell go "Dingggg" (don't know the name) ;p.
Cool, is this method described further somewhere? Is it called something?
this is good - can't change the physics behind this simple test - very good!
On your OtherMill -- you should use mist coolant and run at the right feed/speed with an HSS bit!
I think carbide will still do better with cast iron over HSS - HSS wants a lower SFM thus lower RPM during cutting or it tends to burn up. The Othermill doesn't spin slower than 10k- whereas I'd want to be cutting at less than half that RPM for a .125" endmill on a regular milling machine. Coolant could definitely help with the issue of burn up - spraying a bit of WD40 or a cutting lubricant can also work in a pinch.
You could make a little fan with that motor, they would be pretty useful in motherboards or small places that require coolig, great ideia m8
This is a really cool project! I was excited to see another video in this series. Keep making progress
This rocks. Thorough explanation of your steps, great troubleshooting, friendly vibes, very cool engineering!
I think a "shoulder bolt" might do what you need for holding the bearing better
+1 for shoulder bolts, saved me from a lot of headaches when making some of my RC projects.
yep. and you cant always get the right shoulder length, but you can get one too long and tap it further without much trouble.
Or just a part threaded bolt, it'll be too long but you can just cut the excess threaded portion off.
@@spankeyfish yeah that’s the go to when the shoulder length is right. And the proper solution being AN or MIS bolts
Great work! Good idea to use printed fixtures. About the bearing: When journalling a shaft using deep groove radial ball bearings, you need to use two of them, not just one (unless it is a cross roller bearing, which is not available in such small sizes). In your case you could get away with stacking one on top of the other. This would get rid of that wobble and should improve performance as fluctuations in the air gap distance would be reduced. Also, it is best to press the magnets in; a convenient way is to use a small pillar drill.
Hi Carl. If you extend the idea of decreasing magnetic reluctance from your iron rotor to the stator you could stick a tile of high frequency ferrite material to the opposite side of the stator. I reckon you should get significantly more torque that way. But since you then also increase PCB coil inductance you most likely will decrease the motor Kv value so you will the need a higher supply voltage to maintain max rpm.
Looking forward to your next improved motor!
Love your approach for testing. Excited for the next 2-wheel drive robot video!
I'm loving all these mechanical PCB designs. Keep up the good work.
Brilliant video. I love the determination. If you get the rotor recut, maybe look at making some curved channels on the back to act as a small low flow fan over the pcb.
a bit worried about whether the iron backplate has short-circuited some magnetic flux or not, because the magnets are sunk totally inside the backplate. maybe the holes should be shallower.
Right !
Great work! You can place thinner magnets (in hard plastic discs) on both sides of pcb for more efficiency.
I've heard that hitting magnets isn't the best idea; this may be for old-school magnets so possibly worthy of further investigation.
Maybe a vice or some other parallel compression method might improve things possibly?
No, still goes for exotic magnets. You reduce their strength. Those definitely need to be pressed in. The same with the bearing, but for different reasons. Ball bearings are designed to take a radial load, not an axial load. Hitting it in can deflect the cage and cause the rolling drag to increase.
Half tonne or tonne arbour presses aren't expensive, have plenty of feel and are designed for the job
KNIPEX PLIERS WRENCH makes a great hand held arbour press. especially for small parts like those.
True.
Magnets have been around for centuries and are still popular today. However, magnets may not be the best idea for making PCBs or motors. Magnets can be very strong and can easily damage or pull apart delicate parts. In addition, they can be difficult to remove if they are accidentally attached to a component. Magnets should only be used if there is no other way to hold a component in place.
Cool project.
Kicad is the software that you want.
For real. KiCAD is free, open source *and* cross platform.
Meanwhile Altium is none of those, and costs 3,500€ *per year*. That's 15 Adobe licenses. Totally absurd. They must be running a good racket on whatever businesses pay for this.
At this moment, KICAD lacks on some features comparing to Altium, but the kicad developers make great improvements on each version. I have already used it in some professional projects.
More than 10 years ago Eagle was a promising tool, you had a Free Software version (even available in Ubuntu 2008 packages). But since about 2017, they changed their strategy, they were bought by Autodesk, and from version 4.7 to 9, they have not improved anything at all. They have only made an ANNUAL PAYMENT LICENSE with very annoying authentication system by smarthphone and time limited code. And continues without proper hierarchical design.
KiCAD instead has improved remarkably from version 3 to 5, has hierarchical design, trace length matching (this works very well, so this brings you the possibility to digital high speed design), and python scripting.
If Kicad continues to improve, we will soon have a tool capable of competing directly with Altium.
@@ralvarezb78 eagle is now integrated into Autodesk Fusion360
@@maxhouseman3129 yes, I know
@@maxhouseman3129 I have one fusion360 license at work
Great work. For the shaft/bearing setup, it is recommended to use a pair of ball bearings for supporting the shaft and the shaft could be a shoulder screw instead of a normal fully threaded screw. These changes should minimize the wobble you are seeing.
Superb 👌.
Hats off for you patience while creating these projects
LOVE the way you design PCB's brilliant artist you are Carl!
very nice, can't wait for pcb robot v2 Thank you for sharing your nice work & ideas
I will order my next project definitely from pcbway. Looks like a real cool sponsor and they do way more interesting pcb.
I'm a very big fan of this project, you have my rapt attention, please continue
Your projects are amazing! I was curious about why it was so long since last video…. Now I see! Congrats!
Your effort is appreciable..
One of the most underrated youtubers. Love your content. Keep up the amazing work. :)
This is so cool 😎 One of my favorite projects!
Finally something amazing on TH-cam
This is absolutely awesome. I bet if you glued small ingots behind the PCB at the center of each coil you'll increase the torque significantly. If you can get thin slivers of iron in 0.1-0.3mm thicknesses and you sandwich the slivers with some spray paint as the dielectric you'll cut out a lot of the Eddy currents.
This is so awesome. You need to make the pcb press fit for a bearing and you need the magnet housing press fit for a shaft. Bonus points if you add a second pcb board that mounts to the first with another press fit bearing and mounting holes so you can mate this to a planetary (or other) gear box. That will allow this to easily drive your 2wd robot while getting more consistant readings/results. If this comes across as rude or intrusive I appologize.
Amazing innovation and engineering skills.
Amazing work.
nice improvements man !
Awesome always great to improve
I was thinking about if this was possible - great job
This is genius in action! Thanks for sharing!
Nice work sir! Keep this motor idea in mind. The number 1 killer of all motors is heat. When ever you can, find the best way dissipate heat.
Air flow will provide the cooling for sure. Your current changes will change the temp. Also as you found. Hint - Standard AC motors use their rotor (on an ODP motor) as the fan for air flow. Using both rotor and stator design, in harmony, you can create the optimum airflow you may need.
Very nice new version! If you ever need some dynamic balancing of rotors, I've got the machine.
You could add a small 3d printed fin structure to the PCB-facing side of the rotor which acts as a small ventilator for cooling the PCB actively.
Well done. Great idea. Maybe your high temperature issue can be improved by some design changes of the moving part to shape it more like a fan in order to push air to the pcb to cool it.
This is exactly what I’ve been looking for FOR YEARS
Inspiring ! Thanks a lot for your videos !
Need transformer steel laminate. Also use cnc to make a fixture plate that mounts into the base plate. Cool stuff!🙂
Yep, sandwiched steel plates is the traditional way to avoid eddy currents in motors . A very traditional brushless design is to have a rotor made of stacked plates separated by oxidation, dipped in conductive aluminium to form shorted rotor windings that make the rotor respond to the 3 phase rotating magnetic field from the stator. Thus there are no permanent magnets and the rotor is formed from stacked plates cut/stamped into a designed shape, clamped and dipped.
As always high quality content ❤️👌
You really deserve these sponsors !
Excelente trabajo Carl.
Your videos are amazing 😍
Keep going
Cool little project
I had no idea a pcb axial motor was possible! This is great!
Great video! Perhaps with thinner magnets you could layer the pcbs to get more torque!
Great engineering / trial and error 😎
PCBWay doing great job sponsoring this video!
I think you could further increase the torque by building a rotor with magnets on either side of the PCB. Besides the additional flux from the added magnets such a arrangement is likely decreasing flux leakage by a lot
Awesome project!
I wonder what would happen if it had 2 stators (180° out of phase) sandwiching the rotor?
You really did hardwork brother 👏
A cool project idea for these struck me today - a quad copter.
If you could write custom controller code for a small microcontroller and add a battery, you might be able to have a quad copter milled and assembled and sent to you save for the propellers.
PCBs are pretty lightweight but also decently strong, so being able to just order some quad copter a cheaply from a PCB mill would be pretty cool.
Heck, maybe the military could put electronically-activated explosives on them and get a cheap source of controllable missiles lol
A few more successes like this and you might be able to incorporate this design into a PCB quad-copter!
Mill vanes in the iron rotor that move air to cool the stator and rotor
Those motors would be a great upgrade for people who have models of the TOS Enterprise. The could use them for the rotating bussard lighting on the front of the nacelles. The motors which are being used at the moment are very loud. The rotating engine lights of the TOS Jupiter 2 also pops to mind. I am sure there are other model kits which use conventional motors for lighting effects. It's something to think about.
Imagine a quad where it's arms are the ESC and motors and they all connect into a central flight controller...
A quad where the whole thing can be made from a single etched PCB and some 3D printed props? Talk about driving the cost through the floor.
🤯🤯🤯
@@MurcuryEntertainment using a flexible pcb ...
A quad copter made completely out of one single PCB exists already.
@@maxhouseman3129 I was thinking a central brain that you could plug these arms into. Interchange them... Maybe even plug in other things.. sensors, wheels, whatever.. could be a terrific teaching tool.
Absolutely amazing.
The "real" temperature test is quite hard to do in a home lab.... Ideally you want to measure the motor structure heat - per watt of mechanical energy delivered. For steady state RPM's this is only really doable if you have a reference motor bench with constant load RPM and a torque meter on the coupling shaft.
Good work, really interesting to see :)
subscribed, this is really interesting for micro sized drone motors
This is great! Keep up the good work.
pcb drone? possible idea
Would be fragile, but awesome. 🤞
I mean, I have a small quadcopter that's basically a PCB with 4 motors on it. The main difference here would just be that the motors would be built in to the PCB itself.
Why don't you use a powerful magnets on the other side to keep the iron stable while drilling.
Good job. Thank you for sharing your innovations with us
That is awesome O_O Really well done, cheers for sharing!
You guys are really good and smart. Im very impressed. Keep going. Success to you.
lovely stuff
awesome project!
I think there could be another reason, why temps improve with higher speeds: the faster magnets move over the coils, the more they themselfs will actually induce a voltage, opposing the driving voltage. that will result in lower currents, and is usually the reason why typical electric motors have the highest torque at start up (higher current- stronger magnetic field - more torque). you could easily test that by measuring the amps drawn at different speeds.
Dude... wow... Thank you!
Awesome as always! I wonder if its worth replacing the screw and ball bearing with a bronze bearing and some parts created on a lathe - more space for magnets and less wobbling? :)
Very cool idea!
You could make the rotor with a drill press, it's not a complicated part.
And about the thermals:
Keep in mind that your motors have different efficiency at different loads (aka aluminium vs. iron rotor)
Mechanics usually heat the housing and cool down the bearings when they need a thight fit. It makes it much easier to mount
I just discovered your channel, and you're amazing
for iron milling, you could make a support with the aluminum to hold it!
You need magnetic chuck for such devours. Or you need to have custom designed work holding, soft jaw vice.
This was cool to watch
There are many new improvements with PCB Motor. For starters, the unboxing experience is much better. The initial testing yielded good results, and the iron rotor provides significantly more torque. Additionally, the shaft mount makes it much easier to install and use. Finally, the torque testing showed that this motor is extremely powerful. Overall, these new features make PCB Motor a much better product.
very very impressive research.
You should create a tiny wave (harmonic) reducer to fit on your motor. It will give huge torque
will you try to make an autofocus actuator for small camera sensors? It would be very exciting.
One word incredible
this is amazing
Stators are made of a special iron alloy with silicon, silicon steel. The addition helps keep temperatures lower with changing magnetic fields, and helps the stator "drop" the field faster, for less hysteresis. I have a few plates left. I originally sent to oshcut for some custom stator lamination. They were interested in attempting the material
what is the alloy called? where can I buy it / get it machined?
@@gedr7664 its called Silicon steel..
Carl, love the build and dedication. I have stumbled across this video and chewed on it for a day - have you considered an N+1 approach to the magnet/coil ratio? I suspect this will help torque. Would it be a different driver?
Could the added size and material of the thermal pads be used to make a larger motor instead which would heat up less in the first place?
This is a very important invention. combined with 3d printers and PCB milling machines you can create A LOT!
If you slow down your feeds and also depth of cut , you can cut that with just superglue holding it to the table. I do it all the time
nice man, so cool
It looks pretty cool Dude 👍
You are the real deal my guy.
How about an iron heasink at the back? Would improve thermal issues, but also increase magnetic field strength and improve torque.
The magnets will probably attract to the iron heatsink and based on why previous test I think the coil's magnetic field wouldn't be strong enough to overcome that attraction and rotate the rotor
@@CarlBugeja Didn't PCB way just release an aluminum PCB substrate? Would love to see the motor on that. It would, in effect, have a built in heatsink. Right?
@@Scott_C aluminium will still react to eddy currents
@@zoeyk.6338 true, however the thermal performance may be worth it for the dip in efficiency.
@@Scott_C quite possibly, perhaps one could design the rotor such that it has a fan coupled/built into it, in the name of reducing part count.
Like the shaft passes through the pcb to drive a small fan on the back etc.
Next time you can use a machinist trick. Glue it onto some stock with super glue and then use heat to release the blank when you are done milling.
JohnnyQ90 does this a lot for small parts that are difficult to hold.
I cant wait for the micro 1 board quad copter!
Use super glue to hold the parts you want to machine, at end try breaking the glue bond or dissolve the glue. Never use double sided tape for small contact area parts.