@@Kevely978 try it for yourself... Write out a script and talk into a microphone. Now just say whatever comes into your head. I bet you don't sound the same. Reading from a script always alters the way you sound, especially if it is a different microphone. You will notice the speed he talks at is even different. When you read from a script you have a tendency to enunciate.
Hello. Sound engineer here. First of all I am really amazed at how you made this thing yourself! And the range of frequencies you're getting is pretty impressive considering it has been manufactured by you. But it is not the high frequency that is counted as the parameter for a good sound. Even a tiny buzzer can deliver 20kHz+ sound easily. And also a normal adult human cannot hear above 14-15kHz (so we apply a low pass filter to sounds often) It is the dynamic range and how flat(transparent) the frequencies are across the audible band that decides how good a speaker will sound. Audible Bass in songs is considered to be above 50Hz. You are getting significant mid low frequencies around 600Hz in your test. I would say play with the PCB thickness you can get more bass in the sound. But again congratulations my friend for this project. I didn't even know sound can be produced this way!! 💐💐😊😊😊
@@AliYassinToma Nope, people can hear up to 20kHz. That means 20Hz-20kHz range covers 3σ of normal distribution for whole population (I may be wrong). My personal range is 24Hz-17kHz.
This is a cool project and i loved it. I made one as part of project work couple of years back for flexible electronics course. Try increasing the membrane thickness(you can have multiple silkscreen layers applied) for better response on lower frequencies and increase the drive voltage as your impedance is pretty high which would improve loudness, An O ring to secure pcb on the 3d printed base would uniformly tesnsion the pcb.
You need to find a way to suspend and tension the membrane between the magnets without touching anything. Usually planar transducers use very thin Kapton which is lighter and you can tension it better. Overall it's a very interesting project you did an excellent job.
I think the main limitation of this design is the stiffness of the PCB material. In fact, that's a major factor when it comes to designing microphones too and it's what's limited the production of low-cost and reliable ribbon mics since pretty much the invention of aluminium-coated plastics. The peak near the centre may have something to do with that but I expect the stiffness is what's reducing the bass response to be so poor. High frequencies give us directional clues but a lot of information is in the mid and bass. Not a problem I'd like to try to solve, I'll admit!
You’re correct about the pcb, it can’t vibrate like mylar to create the smoother curve, for its design. I like the attempt and idea. Planars are simple in design, but not to build for sure. There’s a reason only one company successfully builds a full speaker.
@@DodgyBrothersEngineering Magneplanar are the only planar quasi ribbon maker on earth. Martin logan electrostatics are made by many Quad being first full range in 1957. Apogee are the only full range true ribbon maker on earth (were).
My RAAL ribbon tweeters have an element in them that is a very very thin piece of aluminium. After you install a new element (if the old one is damaged) you need to wave heat over the new element to tighten it up, so it plays. You may benefit from a similar design. Make it so the membrane is tight on your location pegs, and have a raise on one side, and a groove on the other, so as you screw it together it tightens up the membrane.
That's a pretty cool project. I use a audeze Penrose headset since April last year and I will never go back to ordinary drivers. The sound is just one of a kind
Very cool project! I've just completed my own 4. diy headset with 50mm dynamic driver and AMT tweeter in a coax config. I used 1. order xo at about 2-3kHz. I used a 2.8uF cap on tweeter and through hole inductor on the bass. Very simple and highly tuneable. You would have huge issues with SPL because of the small pcb area. I would highly suggest that you mix a tweeter and a cheap dynamic bass driver. I would sugest you design the tweeter horisontally in front next to the dynamic driver (towards your face.) That would help with imaging and depth in the soundstage. I think this setup is the only way to beat a 3-500usd headphone. All dynamic drivers have brakeup above 5-6kHz so thats where the planer tweeter comes in. With external resistor 10-20ohm on the tweeter and xo in 3kHz you can easily create the Harman target hump you need to get the tonality right. My dual driver approach seems more complex than a single driver but its so more tuneable and faster in the end than a mechanical all in one membrane. Most planars also have to fast decay in the bass making it sound unatural. A dyncamic driver for bass and fast decaying tweeter its much easier to get the timbre just sounding right. Using an online crossover calculator 1. order you quickly get cap and inductor sizes. The tweeter resistor is only there to spl match the tweeter and could probably be SMD soldered on the membrane. I've been building headphones for a few years now so thrust me when I say its impossible to find or make a single driver that does it all perfect. If you found a good bass driver with low even distributed brakup you could probably skip the slightly expensive inductor aswell. Then the xo filter will only be resistor and tiny cap. The sensitivity of this setup will be normal/high so it can easily be driven too :) I really look forward to see where you take this pcb planar tech in the future!
The Hiel Air Motion Transformer, is the best use of this I've heard put into actual use. It's amazing the amount of air (dB's) those things could create. Yeah, it took a Thousand Pounds of Magnet (maybe a bit less 😂) but they put out. And in two directions by design. ... cool stuff from long ago
Years ago (early 90's) I bought old USSR made Elektronika isodynamic headphones, later sold as Amfiton TDS15. Great sound, construction almost the same :)
watching this while wearing my hifiman he-4xx cans. looks a lot like that circuit path also: it's totally ok that your headphones have a higher impedance - 27 is not at all abnormal for headphone drivers. most are much higher than the typical 4/8ohms for a non-headphone speaker
This topic is very new to me... i haven't heard such a speaker design... you explained it clearly as always... i would like to see more such things... this video is a good start for the year... 😉👍
I’d like to see a book shelf sized speaker, planar speakers aren’t great at low bass when it comes to larger speakers, so can do hybrid, planar for treble and woofers for low end . Very ideal for headphones I must say. Love this project.
Impressive results!!! I can see many different tests with different combinations of magnets and PCB materials & thickness. After spitting out my drink when I discovered the price of the Planar Headphones, I came here when researching the technology to justify the 50X price increase. Thank you for the demo and explanation.
I have an iPhone 13 Pro Max, and when you held the speaker above the microphone I heard the bass perfectly. It gave me a good idea of what it would really sound like and wow. My neighbor offered me a set he had with a frayed wire and YES. This made me want it more. Haha
Thank you, you are filling my curiosity. I've been searching for the animation of planar driver works and no one uploaded it. Thanks for the yt algorithm take me here. The project and the explanation was super lit, though for the noob like me.😂
Andrei, another great video! Thank you for educating us on this. This explains why when I took apart old headphones I didn't see any traditional speakers, just a small disc shaped circuit. I believe these were planar speakers. Very good to know. I look forward to seeing these in headphones.
Dynamic diaphragms can be quite small in headphones, and even smaller with in ear monitors (down to 3.6mm). They still operate using a central voice coil, and a "cone". They're just really small.
One thing to keep in mind when taking sound measurements is that you microphone will contribute a LOT to the recording, which is why you want a microphone not only designed for measurements, but also that your software knows how to compensate for. This is why a lot of people use the umik-1, its cheap, known by REW, and a decent mic for measurements. In regards to frequency response, the ideal chart will look pretty flat, and should have good response volume from about 20hz to about 15-20khz. If you can get it to be 25hz you wont pick up the deepest of base by any measure but the sound would be considered decent by a lot of normal folks.
Great project. I have 2 suggestions/observations: 1. The way you showed your magnet setups, the norths and souths seem to be pointing at each other and cancelling each other out. I would have expected that North on the lower half points to South on upper half so that the magnetic field is made stronger. 2. Since you are using a 2 sided PCB, if you split the coils into parallel sets on the top and bottom side, your total resistance should cone down as they are in parallel. This is basis my RUclips research on axial flux motors and generators, not sure if speakers work the same way. Cheers
yeah the magnets going N-N and S-S did not seem on first glance like it would yield the best result to me, I would have gone N-S S-N N-S S-N - but then applying left hand motor law it seems like this is correct how he has done it - en.wikipedia.org/wiki/Fleming%27s_left-hand_rule_for_motors
Whoahow!!! DIY planar speaker!!! Amazing awesome!!! Try to add thin foam between magnets and membrane. Would be muffled but at least won't hit magnets on overdrive
That's not a bad response curve. It could probably be tuned with different bell caps (the backpiece on headphones) or perhaps varying the track distances in some areas.
Having been in the audio industry for 40 years I must say this (The pcb) was beautifully realised. The planar speaker in a quasi ribbon. Not a true ribbon, but has a unique sound free from boxiness and colouration, well until you put it in a box! I have a pair of Magneplanar 2.5s with a planar bass and mid and a true ribbon tweeter. There has only been one speaker made to reproduce bass with a ribbon, Apogee (Scintilla was first, I believe). They are amazing things but need a power station to drive them. The other full range panels are electrostatics, and they are all different with different strengths and weaknesses. The flexible generator must be at least firmly held round its edge(glued) if not actually in tension. Is this the lightest weight material available? The weight and flexibility of the surround will determine a lot of the performance and maybe even consider a rubber surround? I am not a speaker maker, (obviously).
No, his design used a 0.2mm thickness of the membrane. PCBWay can do 0.025mm thickness. Plus he put even more additional layers on his for aesthetics. So no, you can do a lot better if you wanted. But the material it self is great. 😊
You may wish to look into thicker traces for less i² loss. Alternately, maybe you could have two flexible pcb glued together to drop the copper resistance to one half. Secondly, the magnets are powerfull enough that you could double the space between them giving more room for the pcb to flex without touching the magnets. Lastly, get your self a remote temperature sensor and plot how the the pcb temperature changes with current flow to get an idea what is the maximum cobtinouse power it can safely handle. And, i would hypothesize that the pcb becomes more stretchy the warmer it is, and that will effect freqency responce.
Great work by the maker of the foil speakers and the video! [1] Thicker traces is a good idea, for lowering the total resistance [2] using 2 flexible PCBs could be avoided by just putting coils on the bottom of the already 2 layer PCB. less coil loops with thicker traces necessary on each side. [3] By the way I think the center of the magnets should be on the slits between the coils, not on the coils itself for stronger force. [4] at 2:35 in the video it shows 2 North poles opposite of the foil (and 2 South poles), for more magnetic force they should be North opposite South, they attract as well keeping the 2 halves better together and make the force stronger. I could be wrong but I love to see more experimenting for optimization done. Keep up the great creative work!
Awesome! I love this one. I'm always looking for different ways to have audio output. What do you think caused the impedance in the PCBs you got to be different from what you had in your calculations? Thanks for sharing this one. 🙂
Congratulations! Nice project. As a suggestion, you could use a Hallbach array in the magnets and use them only in one side, so that the diaphragm would have more space for vibration. This should improve low frequency response.
I had this same thought. And thinking about this a little more, and I think it might impair sound quality to have a full halbach array, because it'll have a closed back and there will probably have weird resonances. I think using a bunch of small elements for the longitudinal magnets might help, if they're arranged with spaces between them. From the front, it'd look like a square grid.
It's an interesting idea, using a Hallbach design, but I believe with planars the idea is to have as uniform a static field across the diaphragm as possible. But it sure would be nice for minimizing acoustic reflections.
Would it improve anything if you could use 3 stacked PCBs for each frequency range? the one with the largest surface at the bottom for low frequencies. A smaller one for medium frequencies. And 1 or 2 smaller ones for treble. Is there any chance that it could generate something like surround sound? thanks.
I have some tweeters from JBL and Yamaha from the late '80s using striped coils like yours. They need a 12 dB crossover cutting at 2500 Hz, and a fuse. They didn't succeed in the market because they have a marked tendency to blow up. The printed track break with great ease above 5-10 Watt RMS, this is why they require a fuse in series. Your response is not too bad, but there are resonances caused by excessive spacing between magnet. The ceramic tweeters are the best compromise between cost and efficiency though...
Really interesting project, im not any kind of electrician or anything like that, it just happens that i own one set of planars and one sided planars (with the magnets only on one side) are a thing now, I suggest you may want to check how it works and if it improves the sound in some way or something before making headphones out of these speakers, can't wait for an update.
Fabulous! According to Oersted's right hand rule, the lines of force are perpendicular to the intensity, so why do you put the tracks parallel to the magnets?
Cool! I love the frequency response measurement. More flat than I expected considering that it's homemade. Not much bass of course but still super cool.
The power you need from the amplifier to get a certain acceleration of the membran is indepentend on the impedance. For higher impedances you need a higher voltage, but a lower current. I guess this floppy membrane tends to partial vibrations which can compromize the sound a lot. You get a better control over that if you take a voice coil on a rigid but thin plate suspended by use of two foil rings. The foil should have large enough inner damping. The material can be rubber or silicone rubber. There should also be a small radial tension on the outer rim of the rigit plate in order to define with the mass of the rigid plate a low frequency roll-off, somewhere between 50 and 100 Hz.
I got a laugh from when you said "impedance". Because of your accent it sounded like you said "impotence", which is when a guy can't get, uhh, happy down stairs 🤣 I actually do love hearing ppl from around the world in their native accents (not trying to 'copy' American accent) when they do their videos. Helps to convey the various differences in cultural backgrounds.
Hi, would you have time and interest in making a video showing possibilities of how to eliminate source noise? I mean the audible noises! There is a horde of people on the forums who, like me, change power supplies for LED lighting and even cell phone chargers because of the damn noise of the sources.
Now comes the real struggle, which is tuning the damn thing! It's flatter than I thought it would be besides that bass peak so it's kind of a good start, but that doesn't mean it'll be up to par with tuned professional equipment (at least not in this stage) because flat isn't innately good, usually tunings go more side by side with the Fletcher munson curve on how we perceive frequencies, so they caress the tuning into that, for our biology. Maybe a different mic to have a better reference could also help, Behringer has a measuring mic (Ecm8000) that could be better but it's more money just for the project, I guess... Also I've seen people use a way thinner material for the diaphragm, it almost looks like sandwich plastic wrap 🤷🏻♂️ so there's that, here in RUclips a guy has a small series on his planar headphone build and uses really thin see through material (can't remember what it was)... All this depends on how big or long you really want to go with this audio rabbit whole, which mind me, it's a whole sea of stuff by itself. Good luck!
I'm really surprised, they sound really good for something DIY. The uneven frequency response is 12.5db (excluding bass boost, because why not), which is not bad for factory budget headphones. And here is the result of homemade. And notice how good they recorded on 10:41, i think, you shouldn't put it on the table when recording.
*This project is absolutely perfect! We are waiting in the near future for headphones (even sound systems) with such a planar type of speakers to come out. I speak like an advanced sounder and audiophile*
Lots of ohms is not a Problem. You just need more voltage swing on the amplifier. High impedance is actually easier to drive. Could you space the magnets somewhat more appart? I would go with active speaker design. A small amp with equalizer to correct the frequency response integrated in an the headphones. Sure would make some decent wireless headphones.
What about glueing down the pcb with some silicone? It seems like use could deform the pcb or mounting pins over time. Not sure if this would affect sound in anyway but idbsay worth a shot.
small soft layer maybe 1mm or thinner to each side of the membrane can reduce the magnet rattle sound and also provide some better (imo) free air resonance characteristics, it may also help with smoothing out the peaks from hitting the resonance of the pcb base material. but the desktop speakers are a great idea. ill have to look into that.
That looks really cool; i bet it sounds better than most Hifiman headphones. . . . Joking aside, most planar structure does have a really high capacitance (unfortunately) making some audio IC's having hard time to adjust for it's load , you can circumvent these by coupling it with 1:1 Trafo or 12:1 DI Trafo (as transfer function) like a really ancient Bertagni planar speakers made in the 70's, thus making it really easy to drive. Also, don't get fooled by the frequency response like a typical audiophile on the internet; you do need to measure it with a waterfall test / spectrogram to see how the phase response actually looks like because a badly made planar does sound really thin, like as if almost the planar spent most of it's energy just to "punch" the magnets to it's case lol.
Wow, with this method you could create different speakers on one membrane, bass, mid and tweeter and maybe a surround speaker. Or what about a Helmut containing 7.1.4 speakers.
I almost thought this was a Carl Bugeja video. He does a lot of stuff with electromagnets on flexible PCBs, so making a planar speaker felt about right for him.
Wow This is Really amazing !! Great Idea !! For your bass buzzing - Yo ucould try cutting some finish nails in half, and then drill or awl a separate circle of tensioners for the diaphragm, then when you close the speaker, the diaphragm is already tight. Really Great Video !! I have never seen a small planar speaker like this before!
I think that if you make a type of suspension for the membrane at the edges and center, expand the magnet bars a little more, you can reproduce bass frequencies at a higher volume without damaging the membrane. And use it like headphones
Could you replace the magnets by creating an electromagnetic field using other PCB membranes? If this was possible it could greatly reduce the weight and thickness of the speaker. Cool Build Though!
Wow this is really cool. You should collaborate with DankPods, he talks a lot about headphones. I need to do some research I want to mod my old Kenwood on ear headphones with these.
Do you happen to know the specs of the magnets, like the grade and such, or maybe have a link to where you got them, I wanna try my hand at this but I'm finding it difficult to find the right magnets. Thanks!
try making the PCB spring loaded so it will be stretched from all sides equally and it would be perfectly centered in the middle of the 3D Printed material
Could have compared that dBm curve to a quality speaker hooked onto the exact same amplifier maybe? Would have given a better side by side comparison of what is shown to you.
30 ohms might be very high for free air speakers, but is not a bad value for headphones... A headphone amp might be able to drive them better. Nice project!
Higher impedance actually requires less power to drive, not the other way around. This is why powerful amplifiers use 2 ohm speakers instead of the usual 4 or 8 ohms so they don't get burned.
I'm wondering if I could make some of the PCBs for this and then modify the housing of a pair of AKG K240s to use Planar Magnetic coils, because I've got some fairly nice new production K240s but would like to see if I could install the ultimate upgrade
Imagine this. A thin, also flexible OLED panel in a (rigid) screen with a planar magnetic coil behind it. And what you get is the best speaker in any laptop, probably the best speaker in your house or even neighborhood.
Can you make it so the top band can I have a question or maybe that it could be printed out of flexible material with a solid core or maybe it slips into another core and a soft 3D printable material can be split over it maybe almost like a bubble with loose supports an inside hollowed out maybe even part of the earmuffs could be 3D printed in a similar fashion and then you put a fabric cover over them
Is the idea behind this design that it is an improvement in sound quality? Does this design also work well for freestanding speakers? And is this a recently invented or engineered design? In what way does this improve sound quality?
I think you can get more volume and better quality if you tension more the pcb, a quick solution that comes to my mind is to make the rods skewed so when you close with the one with the holes you force the pcb down and stretch a bit more...
hey this is just like mems... it looks like. hey would you please build a mems type cooler. like those upcoming computer coolers like solid state coolers. love your stuff. its easy to see your going to be very rich in life
![Lil Uzi Vert - Chill Bae [Official Music Video]](http://i.ytimg.com/vi/D7F42F_JM3U/mqdefault.jpg)
Join my Arduino Course (Spanish): bit.ly/2JY8icE
My Tools: bit.ly/3uv8bc3
Help me on Patreon: www.patreon.com/ELECTRONOOBS
it's strange that at the end of the video you can speak normally, The speech during the video is very annoying
@@Kevely978 try it for yourself... Write out a script and talk into a microphone. Now just say whatever comes into your head. I bet you don't sound the same. Reading from a script always alters the way you sound, especially if it is a different microphone. You will notice the speed he talks at is even different. When you read from a script you have a tendency to enunciate.
@@DodgyBrothersEngineering good job on it
@@DodgyBrothersEngineering exactly. It takes a lot of practice to sound natural reading from a script.
You scavenged the bluetooth pcb from the $2 bt adapters ?
Hello. Sound engineer here. First of all I am really amazed at how you made this thing yourself! And the range of frequencies you're getting is pretty impressive considering it has been manufactured by you. But it is not the high frequency that is counted as the parameter for a good sound. Even a tiny buzzer can deliver 20kHz+ sound easily. And also a normal adult human cannot hear above 14-15kHz (so we apply a low pass filter to sounds often) It is the dynamic range and how flat(transparent) the frequencies are across the audible band that decides how good a speaker will sound. Audible Bass in songs is considered to be above 50Hz. You are getting significant mid low frequencies around 600Hz in your test. I would say play with the PCB thickness you can get more bass in the sound.
But again congratulations my friend for this project. I didn't even know sound can be produced this way!! 💐💐😊😊😊
audeze and hifiman are the two most prolific planar headphone manufacturers if you are interested in the topic.
👍
We can hear up to 20khz.. not 15
@@AliYassinToma Nope, people can hear up to 20kHz. That means 20Hz-20kHz range covers 3σ of normal distribution for whole population (I may be wrong). My personal range is 24Hz-17kHz.
Remember the Hiel Air Motion Transformers ???
This is a cool project and i loved it. I made one as part of project work couple of years back for flexible electronics course. Try increasing the membrane thickness(you can have multiple silkscreen layers applied) for better response on lower frequencies and increase the drive voltage as your impedance is pretty high which would improve loudness, An O ring to secure pcb on the 3d printed base would uniformly tesnsion the pcb.
Oh, interesting thank you so much
Dope info all around
This one is already really thick compared to basically any designs
But did you find out why store bought headphones work on low voltage from what i understand from your video????
So... with a thin/thick layer you could emulate a low/high pass filter? Daaam science of materials and its aplications never ceases to amaze me.
Nice project. Small tip = do not place any silkscreen on the PCB, because even its very small added resistance to movement will effect the sound.
You need to find a way to suspend and tension the membrane between the magnets without touching anything. Usually planar transducers use very thin Kapton which is lighter and you can tension it better.
Overall it's a very interesting project you did an excellent job.
Maybe preform and tension it over a rim, like drum heads.
@@EversonBernardes This is probably the way to go for DIY !
Or just ditch the headphone idea and use real speakers (you'll need the 70s for this)
I think the main limitation of this design is the stiffness of the PCB material. In fact, that's a major factor when it comes to designing microphones too and it's what's limited the production of low-cost and reliable ribbon mics since pretty much the invention of aluminium-coated plastics. The peak near the centre may have something to do with that but I expect the stiffness is what's reducing the bass response to be so poor.
High frequencies give us directional clues but a lot of information is in the mid and bass. Not a problem I'd like to try to solve, I'll admit!
You’re correct about the pcb, it can’t vibrate like mylar to create the smoother curve, for its design. I like the attempt and idea. Planars are simple in design, but not to build for sure. There’s a reason only one company successfully builds a full speaker.
@@RectifiedMetals I am going to piss myself laughing if you say Martin Logan.
@@DodgyBrothersEngineering martin logan
@@DodgyBrothersEngineering Magneplanar are the only planar quasi ribbon maker on earth. Martin logan electrostatics are made by many Quad being first full range in 1957. Apogee are the only full range true ribbon maker on earth (were).
My RAAL ribbon tweeters have an element in them that is a very very thin piece of aluminium. After you install a new element (if the old one is damaged) you need to wave heat over the new element to tighten it up, so it plays. You may benefit from a similar design. Make it so the membrane is tight on your location pegs, and have a raise on one side, and a groove on the other, so as you screw it together it tightens up the membrane.
^
Man, your recent videos are getting much more interesting and advanced. Your PCB skills are going through the roof.
That's a pretty cool project. I use a audeze Penrose headset since April last year and I will never go back to ordinary drivers. The sound is just one of a kind
Very cool project! I've just completed my own 4. diy headset with 50mm dynamic driver and AMT tweeter in a coax config. I used 1. order xo at about 2-3kHz. I used a 2.8uF cap on tweeter and through hole inductor on the bass. Very simple and highly tuneable. You would have huge issues with SPL because of the small pcb area. I would highly suggest that you mix a tweeter and a cheap dynamic bass driver. I would sugest you design the tweeter horisontally in front next to the dynamic driver (towards your face.) That would help with imaging and depth in the soundstage. I think this setup is the only way to beat a 3-500usd headphone.
All dynamic drivers have brakeup above 5-6kHz so thats where the planer tweeter comes in. With external resistor 10-20ohm on the tweeter and xo in 3kHz you can easily create the Harman target hump you need to get the tonality right. My dual driver approach seems more complex than a single driver but its so more tuneable and faster in the end than a mechanical all in one membrane. Most planars also have to fast decay in the bass making it sound unatural. A dyncamic driver for bass and fast decaying tweeter its much easier to get the timbre just sounding right.
Using an online crossover calculator 1. order you quickly get cap and inductor sizes. The tweeter resistor is only there to spl match the tweeter and could probably be SMD soldered on the membrane. I've been building headphones for a few years now so thrust me when I say its impossible to find or make a single driver that does it all perfect. If you found a good bass driver with low even distributed brakup you could probably skip the slightly expensive inductor aswell. Then the xo filter will only be resistor and tiny cap. The sensitivity of this setup will be normal/high so it can easily be driven too :)
I really look forward to see where you take this pcb planar tech in the future!
I’ve always wanted to build a 2 way headphone driver like that. Why dont you upload a video on it my guy. I would be sooo grateful to you. 😁🙏
The Hiel Air Motion Transformer, is the best use of this I've heard put into actual use.
It's amazing the amount of air (dB's) those things could create.
Yeah, it took a Thousand Pounds of Magnet (maybe a bit less 😂) but they put out. And in two directions by design.
... cool stuff from long ago
Years ago (early 90's) I bought old USSR made Elektronika isodynamic headphones, later sold as Amfiton TDS15. Great sound, construction almost the same :)
The new video comes! EXCELLENT as always! We are eager to see more flexible PCB projects from you in the future.
Better versions are on the way... :)
@@ELECTRONOOBS Can't wait to see them
watching this while wearing my hifiman he-4xx cans. looks a lot like that circuit path
also: it's totally ok that your headphones have a higher impedance - 27 is not at all abnormal for headphone drivers. most are much higher than the typical 4/8ohms for a non-headphone speaker
This topic is very new to me... i haven't heard such a speaker design... you explained it clearly as always... i would like to see more such things... this video is a good start for the year... 😉👍
Thank you so much
I’d like to see a book shelf sized speaker, planar speakers aren’t great at low bass when it comes to larger speakers, so can do hybrid, planar for treble and woofers for low end . Very ideal for headphones I must say.
Love this project.
Impressive results!!! I can see many different tests with different combinations of magnets and PCB materials & thickness. After spitting out my drink when I discovered the price of the Planar Headphones, I came here when researching the technology to justify the 50X price increase. Thank you for the demo and explanation.
Great Video! It's always an unmatched experience listening to music on a homemade speaker/Audio System after putting in hours of work!
You might need a thin layer of foam on each side of the membrane to dampen it a little. It will also prevent selfresonance and stuff.
Good idea! I might try it :)
This channel is just what I was looking for! And these headphone elements are incredible. Thank you for sharing your work and knowledge!
I had never heard of Planar Speaker before, thanks for teaching me something new!
Try to test it irl. I’ve got an audeze lcd-1 and they a way better than any of classical dynamic headphones
I have an iPhone 13 Pro Max, and when you held the speaker above the microphone I heard the bass perfectly. It gave me a good idea of what it would really sound like and wow. My neighbor offered me a set he had with a frayed wire and YES. This made me want it more. Haha
Thank you, you are filling my curiosity. I've been searching for the animation of planar driver works and no one uploaded it. Thanks for the yt algorithm take me here. The project and the explanation was super lit, though for the noob like me.😂
Andrei, another great video! Thank you for educating us on this. This explains why when I took apart old headphones I didn't see any traditional speakers, just a small disc shaped circuit. I believe these were planar speakers. Very good to know. I look forward to seeing these in headphones.
Dynamic diaphragms can be quite small in headphones, and even smaller with in ear monitors (down to 3.6mm). They still operate using a central voice coil, and a "cone". They're just really small.
Whenever you picked the planar from the desk there was a moment of good base. These are kool
One thing to keep in mind when taking sound measurements is that you microphone will contribute a LOT to the recording, which is why you want a microphone not only designed for measurements, but also that your software knows how to compensate for. This is why a lot of people use the umik-1, its cheap, known by REW, and a decent mic for measurements.
In regards to frequency response, the ideal chart will look pretty flat, and should have good response volume from about 20hz to about 15-20khz. If you can get it to be 25hz you wont pick up the deepest of base by any measure but the sound would be considered decent by a lot of normal folks.
Great project. I have 2 suggestions/observations:
1. The way you showed your magnet setups, the norths and souths seem to be pointing at each other and cancelling each other out. I would have expected that North on the lower half points to South on upper half so that the magnetic field is made stronger.
2. Since you are using a 2 sided PCB, if you split the coils into parallel sets on the top and bottom side, your total resistance should cone down as they are in parallel.
This is basis my RUclips research on axial flux motors and generators, not sure if speakers work the same way.
Cheers
yeah the magnets going N-N and S-S did not seem on first glance like it would yield the best result to me, I would have gone N-S S-N N-S S-N - but then applying left hand motor law it seems like this is correct how he has done it - en.wikipedia.org/wiki/Fleming%27s_left-hand_rule_for_motors
Whoahow!!! DIY planar speaker!!! Amazing awesome!!!
Try to add thin foam between magnets and membrane. Would be muffled but at least won't hit magnets on overdrive
That's not a bad response curve. It could probably be tuned with different bell caps (the backpiece on headphones) or perhaps varying the track distances in some areas.
Even virtually EQing it flat will probably result in something impressively listenable for the price, assuming the dac and amp are up to the task
Ooh can't wait for the headphone enclosure
Having been in the audio industry for 40 years I must say this (The pcb) was beautifully realised. The planar speaker in a quasi ribbon. Not a true ribbon, but has a unique sound free from boxiness and colouration, well until you put it in a box! I have a pair of Magneplanar 2.5s with a planar bass and mid and a true ribbon tweeter. There has only been one speaker made to reproduce bass with a ribbon, Apogee (Scintilla was first, I believe). They are amazing things but need a power station to drive them. The other full range panels are electrostatics, and they are all different with different strengths and weaknesses.
The flexible generator must be at least firmly held round its edge(glued) if not actually in tension. Is this the lightest weight material available? The weight and flexibility of the surround will determine a lot of the performance and maybe even consider a rubber surround? I am not a speaker maker, (obviously).
No, his design used a 0.2mm thickness of the membrane. PCBWay can do 0.025mm thickness. Plus he put even more additional layers on his for aesthetics. So no, you can do a lot better if you wanted. But the material it self is great. 😊
@@INeedsMoneys Thank you sir.
You may wish to look into thicker traces for less i² loss. Alternately, maybe you could have two flexible pcb glued together to drop the copper resistance to one half.
Secondly, the magnets are powerfull enough that you could double the space between them giving more room for the pcb to flex without touching the magnets.
Lastly, get your self a remote temperature sensor and plot how the the pcb temperature changes with current flow to get an idea what is the maximum cobtinouse power it can safely handle. And, i would hypothesize that the pcb becomes more stretchy the warmer it is, and that will effect freqency responce.
Maybe instead of thicker traces he could try no enamel coating but gold plated?
Great work by the maker of the foil speakers and the video! [1] Thicker traces is a good idea, for lowering the total resistance [2] using 2 flexible PCBs could be avoided by just putting coils on the bottom of the already 2 layer PCB. less coil loops with thicker traces necessary on each side. [3] By the way I think the center of the magnets should be on the slits between the coils, not on the coils itself for stronger force. [4] at 2:35 in the video it shows 2 North poles opposite of the foil (and 2 South poles), for more magnetic force they should be North opposite South, they attract as well keeping the 2 halves better together and make the force stronger. I could be wrong but I love to see more experimenting for optimization done. Keep up the great creative work!
Awesome! I love this one. I'm always looking for different ways to have audio output. What do you think caused the impedance in the PCBs you got to be different from what you had in your calculations? Thanks for sharing this one. 🙂
I would love to see this project be finished. :)
The response looks good. Add fiberglass stuffing to the box to reduce the box resonance at 500Hz.
20 ohms is about the same impedance of KZ in-ear monitors, so not far away from already available products.
Congratulations! Nice project. As a suggestion, you could use a Hallbach array in the magnets and use them only in one side, so that the diaphragm would have more space for vibration. This should improve low frequency response.
I had this same thought. And thinking about this a little more, and I think it might impair sound quality to have a full halbach array, because it'll have a closed back and there will probably have weird resonances. I think using a bunch of small elements for the longitudinal magnets might help, if they're arranged with spaces between them. From the front, it'd look like a square grid.
It's an interesting idea, using a Hallbach design, but I believe with planars the idea is to have as uniform a static field across the diaphragm as possible. But it sure would be nice for minimizing acoustic reflections.
The amount of valuable information you provide is appreciated 👍🏿
Would it improve anything if you could use 3 stacked PCBs for each frequency range? the one with the largest surface at the bottom for low frequencies. A smaller one for medium frequencies. And 1 or 2 smaller ones for treble. Is there any chance that it could generate something like surround sound? thanks.
I have some tweeters from JBL and Yamaha from the late '80s using striped coils like yours.
They need a 12 dB crossover cutting at 2500 Hz, and a fuse. They didn't succeed in the market because they have a marked tendency to blow up. The printed track break with great ease above 5-10 Watt RMS, this is why they require a fuse in series.
Your response is not too bad, but there are resonances caused by excessive spacing between magnet.
The ceramic tweeters are the best compromise between cost and efficiency though...
Really interesting project, im not any kind of electrician or anything like that, it just happens that i own one set of planars and one sided planars (with the magnets only on one side) are a thing now, I suggest you may want to check how it works and if it improves the sound in some way or something before making headphones out of these speakers, can't wait for an update.
Very cool!! Can't wait to see what the headphones turn out like!
Не думал что снова увижу подобное спустя много лет, были у меня наушники еще сделанные в эпоху СССР, но там была проблема с кабелем и я их разобрал.
Fabulous!
According to Oersted's right hand rule, the lines of force are perpendicular to the intensity, so why do you put the tracks parallel to the magnets?
Cool! I love the frequency response measurement. More flat than I expected considering that it's homemade. Not much bass of course but still super cool.
The power you need from the amplifier to get a certain acceleration of the membran is indepentend on the impedance. For higher impedances you need a higher voltage, but a lower current.
I guess this floppy membrane tends to partial vibrations which can compromize the sound a lot. You get a better control over that if you take a voice coil on a rigid but thin plate suspended by use of two foil rings. The foil should have large enough inner damping. The material can be rubber or silicone rubber. There should also be a small radial tension on the outer rim of the rigit plate in order to define with the mass of the rigid plate a low frequency roll-off, somewhere between 50 and 100 Hz.
I have a set of Magnepan Magnplanar speakers. Bought in 1977, still work. Powered by a Phase Linear amp.
Amazing, you think of things mere mortals don't.
I got a laugh from when you said "impedance". Because of your accent it sounded like you said "impotence", which is when a guy can't get, uhh, happy down stairs 🤣
I actually do love hearing ppl from around the world in their native accents (not trying to 'copy' American accent) when they do their videos. Helps to convey the various differences in cultural backgrounds.
Hi, would you have time and interest in making a video showing possibilities of how to eliminate source noise? I mean the audible noises!
There is a horde of people on the forums who, like me, change power supplies for LED lighting and even cell phone chargers because of the damn noise of the sources.
Now comes the real struggle, which is tuning the damn thing! It's flatter than I thought it would be besides that bass peak so it's kind of a good start, but that doesn't mean it'll be up to par with tuned professional equipment (at least not in this stage) because flat isn't innately good, usually tunings go more side by side with the Fletcher munson curve on how we perceive frequencies, so they caress the tuning into that, for our biology. Maybe a different mic to have a better reference could also help, Behringer has a measuring mic (Ecm8000) that could be better but it's more money just for the project, I guess... Also I've seen people use a way thinner material for the diaphragm, it almost looks like sandwich plastic wrap 🤷🏻♂️ so there's that, here in RUclips a guy has a small series on his planar headphone build and uses really thin see through material (can't remember what it was)... All this depends on how big or long you really want to go with this audio rabbit whole, which mind me, it's a whole sea of stuff by itself. Good luck!
Some wool pads make miracles in such small enclosures as a stuffing.
Fantastic! I look forward to seeing your headphones come to life.
The curve looks like that because it has resonance, something you want to prevent it would be cool seeing how you could get around it though!
Today I learned something new from you.
I'm really surprised, they sound really good for something DIY. The uneven frequency response is 12.5db (excluding bass boost, because why not), which is not bad for factory budget headphones. And here is the result of homemade. And notice how good they recorded on 10:41, i think, you shouldn't put it on the table when recording.
*This project is absolutely perfect! We are waiting in the near future for headphones (even sound systems) with such a planar type of speakers to come out. I speak like an advanced sounder and audiophile*
Lots of ohms is not a Problem. You just need more voltage swing on the amplifier. High impedance is actually easier to drive. Could you space the magnets somewhat more appart? I would go with active speaker design. A small amp with equalizer to correct the frequency response integrated in an the headphones. Sure would make some decent wireless headphones.
What about glueing down the pcb with some silicone? It seems like use could deform the pcb or mounting pins over time. Not sure if this would affect sound in anyway but idbsay worth a shot.
small soft layer maybe 1mm or thinner to each side of the membrane can reduce the magnet rattle sound and also provide some better (imo) free air resonance characteristics, it may also help with smoothing out the peaks from hitting the resonance of the pcb base material. but the desktop speakers are a great idea. ill have to look into that.
That looks really cool; i bet it sounds better than most Hifiman headphones.
.
.
.
Joking aside, most planar structure does have a really high capacitance (unfortunately) making some audio IC's having hard time to adjust for it's load , you can circumvent these by coupling it with 1:1 Trafo or 12:1 DI Trafo (as transfer function) like a really ancient Bertagni planar speakers made in the 70's, thus making it really easy to drive.
Also, don't get fooled by the frequency response like a typical audiophile on the internet; you do need to measure it with a waterfall test / spectrogram to see how the phase response actually looks like because a badly made planar does sound really thin, like as if almost the planar spent most of it's energy just to "punch" the magnets to it's case lol.
Please make a video of the headphone with this plannar speaker.
You can add some cuts to membrane to improve low frequency response, decreasing it's stiffness.
tbh… can’t wait to see DankPods and Crinacle getting interested in this project and then make a review on the headphones
Wow, with this method you could create different speakers on one membrane, bass, mid and tweeter and maybe a surround speaker. Or what about a Helmut containing 7.1.4 speakers.
I almost thought this was a Carl Bugeja video. He does a lot of stuff with electromagnets on flexible PCBs, so making a planar speaker felt about right for him.
Well done, very nice. Lots of hard work for an amazing modern design.
Wow This is Really amazing !! Great Idea !! For your bass buzzing - Yo ucould try cutting some finish nails in half, and then drill or awl a separate circle of tensioners for the diaphragm, then when you close the speaker, the diaphragm is already tight. Really Great Video !! I have never seen a small planar speaker like this before!
Where to buy the same magnets you used for this project? Thanks!
That frequency response is good bro. There needs to be a bit more gap between the magnets and coils
Amazing project Bro 👍
I think that if you make a type of suspension for the membrane at the edges and center, expand the magnet bars a little more, you can reproduce bass frequencies at a higher volume without damaging the membrane. And use it like headphones
Could you replace the magnets by creating an electromagnetic field using other PCB membranes? If this was possible it could greatly reduce the weight and thickness of the speaker. Cool Build Though!
What would you power the static electromagnets from?
Wow this is really cool. You should collaborate with DankPods, he talks a lot about headphones.
I need to do some research I want to mod my old Kenwood on ear headphones with these.
Thanks for making this video! I've been wondering how planar magnetic speakers work
This is super slick, nice work!
Do you happen to know the specs of the magnets, like the grade and such, or maybe have a link to where you got them, I wanna try my hand at this but I'm finding it difficult to find the right magnets.
Thanks!
try making the PCB spring loaded so it will be stretched from all sides equally and it would be perfectly centered in the middle of the 3D Printed material
Could have compared that dBm curve to a quality speaker hooked onto the exact same amplifier maybe? Would have given a better side by side comparison of what is shown to you.
30 ohms might be very high for free air speakers, but is not a bad value for headphones...
A headphone amp might be able to drive them better.
Nice project!
Thank you! Glad to know that!
Are flexible PCBs made for handing such vibrations? Will they leach out chemicals over time?
I don't know allot about acoustics, but would a lite foam help with the sound you got from the large amp, or would that mess up the audio?
Higher impedance actually requires less power to drive, not the other way around. This is why powerful amplifiers use 2 ohm speakers instead of the usual 4 or 8 ohms so they don't get burned.
Man i love your channel i wish i have the same equipments you have plus you have cool ideas
Beautiful project ❤🙌 thanks for this 👍
I'm wondering if I could make some of the PCBs for this and then modify the housing of a pair of AKG K240s to use Planar Magnetic coils, because I've got some fairly nice new production K240s but would like to see if I could install the ultimate upgrade
It is possible to place thin soft dampers between the membrane and magnets. Silk fabric, or nylon tights?
Imagine this. A thin, also flexible OLED panel in a (rigid) screen with a planar magnetic coil behind it. And what you get is the best speaker in any laptop, probably the best speaker in your house or even neighborhood.
Can you make it so the top band can I have a question or maybe that it could be printed out of flexible material with a solid core or maybe it slips into another core and a soft 3D printable material can be split over it maybe almost like a bubble with loose supports an inside hollowed out maybe even part of the earmuffs could be 3D printed in a similar fashion and then you put a fabric cover over them
Can you make a diaphragm mold to heat up and press the pcb to get a sawtooth shape? I think that way it will have some space for cone excursion?
Is the idea behind this design that it is an improvement in sound quality? Does this design also work well for freestanding speakers? And is this a recently invented or engineered design? In what way does this improve sound quality?
Your using the tbone microphone? Maybe can you share your record settings? Sound better than mine
You should've tested it with scarlet fire
That is cool. only the flex PCBs are so pricey compared to FR* PCBs for one-time hobby projects and experimenting.
I think you can get more volume and better quality if you tension more the pcb, a quick solution that comes to my mind is to make the rods skewed so when you close with the one with the holes you force the pcb down and stretch a bit more...
This is fantastic video and info !! Do you have something like this for dynamic drivers?
Would this work if the coil part of the pcb would have sort of suspensions at edges? I just thought that maybe then movement would be more uniform.
interesting what can be done with divided windings and 7.1 input, could it improve volume effect from typical sorround technics.
hey this is just like mems... it looks like. hey would you please build a mems type cooler. like those upcoming computer coolers like solid state coolers. love your stuff. its easy to see your going to be very rich in life
Amazing build