Part 2: Hi-Res Pressure Sensor Matrix Mat finished
ฝัง
- เผยแพร่เมื่อ 25 ต.ค. 2017
- 2$ PCB service: jlcpcb.com/
Copper tape: www.banggood.com/custlink/DGK...
Lattepanda: www.dfrobot.com/product-1404....
Patreon: / marcoreps
Part 1: • Hi-Res Pressure Sensor...
PC software and arduino sketch: drive.google.com/file/d/0B6S_...
Interface board schematic (very messy): imgur.com/QWzrSu2
In this video we'll finish the large pressure sensor matrix mat. The tedious manual work of applying copper tape to vinyl foil isn't really worth showing, so instead I'll put a few more details in this video. Z-Axis tape, a Chinese 2$ PCB service, Arduino ADC clock manipulation, three-state shift register outputs, dog. - วิทยาศาสตร์และเทคโนโลยี
Please upload the "falling off the table and swearing profusely" video.
i guess Im asking the wrong place but does someone know a way to get back into an Instagram account..?
I was dumb lost my password. I would appreciate any assistance you can offer me!
@Harper Genesis instablaster =)
@Iker Nickolas i really appreciate your reply. I got to the site through google and im waiting for the hacking stuff now.
Takes quite some time so I will get back to you later when my account password hopefully is recovered.
Marco, your "Oh Shit" moment almost made me blow my coffee all over my screen. Impressive that you can even see your individual toes. Nice work.
5:33 Jesus christ, what are you doing to your router?
... the router had enough
@@ThePowerofElectricity more like hang itself amirite...
all tables should come whit a smaller sub shelf for cables and small devices
@@holetoanotheruniverse4690 hahahahah
R. I. P
listening to this i suddenly realized just how amazingly complex our skin is. it's like a super-high res pressure sensor matrix. now.. imagine the wiring schematics :D
and the functional block diagram of all the feedback mechanisms and interplays....
Pretty much a rats nest of wires that are all twisted and guttered into a 'spinal column' if you will. Imagine keeping the memories of when you were trying to figure out what wire goes where when you were an infant :P
Well, high res in some places. Fingertips, toes, lips, tongue, genitals. On your back, for example, points less than about an inch apart are indistinguishable. Even the brain doesn't have enough channels for high-resolution input across the whole body.
Serial or parallel?
@@AgentOffice I'd guess a weird, quasi-random mix of both that constantly changes.
"I have grossly underestimated the time it takes to do this..." I have a prototyping and manufacturing company. This statement should be the companies moto. We have a rule. If we have never done it before, estimate the time, with the usual +30% buffer and then times it by 5. That might get us close to what it actually takes.
Hofstadter's Law: It always takes longer than you expect, even when you take into account Hofstadter's Law. _from Gödel, Escher, Bach by Douglas Hofstaedter
@@frozenfrogz 😂
people like you is the kind of people that make science interesting
This will get on trending in the year 2020
Take me there with you.
Well done
how to make an iron pressure plate in real life
That magic 3M Z Axis stuff looks amazing. Thanks for the tip!
Surrealistic Pillow yeah i have seen it in the paperduino video but i didnt know what it was
I didn't know about that stuff until now. It's beautifully simple and elegant.
Man, you deserve a medal for such patience. Good job
Many hours in the service of can I get this to work! Great application of engineering knowledge and practical abilities. Thanks for taking the time to make this video Marco.
Mark
5:33
“Dad why does the wifi keep hanging and going down”
The fact that these was done 4 years ago is so impressive. I got a future project boss I want to do but I don’t have quiet the funds yet to hire a full team. But if your still alive in 2025. I’m definitely hiring you!
Great project!
The worst! (:
@@reps Great effort man.
@@reps
what about using a pseudo pressurized air mattress ? The inside top surface impregnated with reflective dots 1 mm spacing
an array of moveable mirrors on the bottom to direct a laser beam or light pulse which scans the entire ceiling of the inside surface of the top of the mattress at different angles for a certain desired resolution
the same mirrors would be able to direct the reflection of the light pulse back to a receiver
with the known pulse / beam location and the known ' static ' location of each reflective dot - it would be possible to calculate downward deflection of the ceiling by measuring the increased distance between neighboring dots which have been depressed due to the material being distorted
Couldn't you use cheaper aluminium tape, or would the oxide layer be an issue? ( copper will oxidise over time ).
Something else that may be worth a look is Bare Conductive paint - a low-cost carbon based paint that can be screen-printed
Instead of HC595's+diodes, check out the NPIC6C596 , which is like a 595 but with open-drain MOSFET outputs.
The copper tape seems to covered in something that protects it against oxidation, because when touching in with bare fingers, I am leaving very well-defined, oxidized fingerprints behind imgur.com/sZOilJH Aluminium tape or even better grocery store aluminium foil didn't even occur to me. Should I ever do this again, I'd ask for 2 huge flex PCBs ...
mikeselectricstuff 6
isn't Aluminum Oxide more restive then Copper oxide. I know we stopped using Aluminum in mains wiring for that same reason.
Shawn White the guy who knows little about materials science (me) thinks it's because copper is a better conductor.
Copper oxide is conductive, more or less.
You're going to be a very successful TH-camr, I'm sure of it! Well done! :) I'm very impressed by your skill-set. I'm 48 and am very grateful to people like you that are willing to share your projects and experience.
Love your videos mate. Fantastic sense of humour. My favorite electrical you tube channel
Wow that's unbelievable, it is even capturing the foot shape, amazing
Macro reps relased new video :3 that will be highlight of my day
Awesome!!!
Glad you're OK!!!
I've been waiting for this video, you're amazing man. huge thumbs up.
Very very cool project no doubt! Congratulations Marcos!
"I am a proffesional, i know what I'm doing."
I have to admit I really love your videos. Great projects an great style :)
Aw man, you had me at the feet shuffling with paper bits and of course the door, but the dog at the end was just perfect.
The sensor matrix is such a cool device. Like wow that is truely a useful too for so many applications :D great work!
I feel like this could be a helpful edition to home security, albeit done at a lower level with lower resolution. You could have pressure sensitive rugs that detect human footsteps, and can distinguish them from your dog or cat.
Hilarious video!! Good on you for persevering
This is very similar to my degree project where we made two insoles for measuring pressure underfoot at 7 by 14 points at 20Hz.
One main difference between this and ours is that our row and column conductors were conductive steel thread that we sewed into fabric on either side of the velostat.
We then used our small NRF microcontroller which was mounted on the heel to send the data to a nearby cellphone using Bluetooth LE for storage and a windows app for displaying the pressure.
Sadly it wasn't in real time, and the velostat wore out after around 14 participants so we weren't able to get a large enough sample size to train a classifier to detect different foot issues.
But it was relatively cheap, only around $100 per foot including microcontroller and enclosure.
Wow, nice work!
Samy!
This was an interesting build, thank you for sharing! Also, love the comic relief. Hope you didn't get hurt too bad from the fall
This is a very cool project! Great job!
Thanks! Well done. Great project, big fun - to watch.
You say you'd rather never see the thing again. If you're ever gonna burn it to bits, I've love to take it and add it to my garage gym / experimental physio lab where I'm using a bunch of depth cameras to gauge the imbalances in my movements. :)
I've always wanted to make a pressure mat for that data point but I'm sadly more of a computer vision kinda guy than hardware. This project is amazing by the way and I'm hoping I can try my hand at recreating this in the future!
Liked and subscribed, this project is awesome! Also, I love your workspace
that poor fritzbox at 9:34
masiboss exactly my thoughts...
That is fritzbox abuse and should be outlawed ! Jawoll
Why? Looks fine. It's just hanging around. ;)
5:34 btw. not 9:34
This is the perfect router mounting technique. Also works on other electronics that have a ethernet plug.
but not a server tho.
Interesting project, an you did professionally indeed.
Interesting project, it's helped give me an idea for one I was thinking of. Never knew shift registers could be used in such a way
Love your dry humour. Good on you.
Something like this makes the Sensel Morph seem all that much more amazing and cost effective.
The to be continued meme just killed me great videos keep it up
7:27 I just died because you almost did with this JoJo meme.
Brutus B. Martins I would agree with that statement, but not with your choice of avatar.
WOW, excellent Video and project
Wow, this is an amazing project! Well done! Please post a follow up video!
The fritz box is having a good time hanging out there
This reminds me of the Madrona Labs Soundplane instrument. Very nice work, mate!
To the point beyond reason! And beyond!
hnnng dat roundabout playing just for a second. such a good song :3 Awesome project and video of it, thank you very much.
yes
yes
yes
I once tried to make a pressure sensor based on capacitance instead of conductivity, though much smaller (palm sized). It was just a PCB with individual (non multiplexed) pads, a thin layer of some sort of rubber sheet that was lying around, a copper foil and a final top protective layer of transparent plastic from some bag of cement or something.
The measurement was by setting an output high then making it high impedance and measuring how long it takes to go low, which would be a few hundred nS up to maybe a couple of micro seconds. It worked, sort of. Not very accurate (about 20 steps) and very different sensitivity between the 'pixels' but I think with some more time and effort the principle could be refined to work very well. It needs many IO pins and I used several 40 pin dip PICs as I didn't have the skill or meams to work with high pin count SMDs.
Wow. This is brilliant!
only a german- amazing! I've been wanting to build something like this for years, but all those shift registers. time to get some of this velostat stuff. dog at the end was perfect. subbed
Fantastic work, man!
What English idiot said Germans don't have a sense of humour? LOL You remind me of a guy named Reinhardt I knew from Frankfurt. He was hilarious too! Keep up the videos ... and the sarcastic humour - love it. Bis später!
Amazing project!
7:27 It was at this moment Marco knew.... He f'd up
very informative! May I ask what was the software you used to draw up interface board? thanks
Led strips have tiny controllers which read a data line to set the indiviual leds according to the data. Its a shame there isn't a universal controller which you could use for each preassure element, in place of that matrix. This way each controller would send its current preassure reading together with it location on the matrix through to a single data line, kind of like I2C. I have a project in mind which would need preassure sensors wired in serial and would benifit from having controllers of this kind, instead of a mass of wires.
Each control chip in the serial chain would be connected to ground and a VCC, plus an analog read. Each controller would send to the data line its digital analog value, or zero to signal it was the last in the chain and had no values to pass through.
With each control chip responcible for reading it own analogue and converting to digital, This would vastly speed up that conversion, instead of having a single controller cycling through a matrix of analog values.
Hello Marco, I like your project and learned a lot from it! One question, do you know how the velostat actually works? is the resistance under pressure lower because the electrodes make better contact with it? or is it in the material itself? that would make a huge difference in designing the pressure sensor i think
Very interesting project and build. I'm getting a back ache just watching you lay that tape :)
Man I love watch your videos. Super funny stuff. I know nothing about electronics but sometimes I fall asleep to this shit. Two thumbs up
i ordered on the jlcpcb thing too, took 3 weeks. quality is fine. but its +10$ shipping. still unbeatable price.
gamerpaddy the "average" price for a PCB of that size is about 4.90, $2 is incredibly low. However as you said, shipping usually overshadows the actual PCB price, unless you are ordering alot of PCBs
They can do this because export shipping from China is heavily subsidised. This tactic undercuts your local manufacturer who actually has to PAY for shipping. You work out whether that's good or bad..
yeah. some shops actually "sell" their stuff for "free" and then do 4$ shipping and have a 400% margin or something(guessing the percentage since you can find it for less than 1$ on ebay with free shipping).
Shipping from china is actually less than 50 cents for most items. shipping on those pcb's is likely less than 1$(if its registered maybe 2 or 3). so you are defintly paying with the shipping.
Shipping is that cheap because china is still considered a third world country in that particular treaty because it was at the time the contract was written.
Another great video. You're an inspiration.
Awesome build! What was the adhesive used to keep the top/bottom copper tape + velostat sandwich together?
@Marco Reps Excuse me sir, how could you finish this project without using mosfets? Everytime you out a "1" in a Q output of the register the 5V are not enough to work with so many colums at the time. Can you please explain me how its done?
Great job, man!
Love this video. Earned a sub and notifications!
That is very very impressive!! That could be used with composing different rhythms to support gait retraining! Oh a Music Therapist could do the coolest stuff with that!
I love your channel so much
That SMD sticky tape is really neat.
Impressive work!
I've made something very similar. A 32x32 printed pressure sensor. I used a single voltage input and 3 4067 16 channel analog multiplexers. The 32 sensors connect directly two of the multiplexers, and the third is used to select between the two (it has 14 outputs unused so you could relatively easily scale it to 256x246 arrays by just adding more multiplexers). That way you don't need diodes. The multiplexers need a 4 bit address, so each axis is addressed with a shift register (first 4 bits go to the upper level multiplexers, second 4 are connected to all the lower level multiplexers).
The output is basically a mirror of the input going into the analog in via a potential divider (with adjustable gain via another 4067 multiplexer - selecting 16 resistances between 200 Ohm and 1 MOhm). Your code is certainly a lot faster than mine at sending the data to a computer I send both the address and resistance in plain text and get
Hi Tim,
Your work seems cool! I wonder what use do you make of an adjustable gain?
Also, can you share your code so I can inspire myself for a similar project?
Thanks!
Amazing project :-) How did you choose the reading frequency? What would be the maximum you could attain, and what would be the bottleneck? Is is the multiplexing, the arduino input, ... ? Thanks!
what about using a pseudo pressurized air mattress ? The inside top surface impregnated with reflective dots 1 mm spacing
an array of moveable mirrors on the bottom to direct a laser beam or light pulse which scans the entire ceiling of the inside surface of the top of the mattress at different angles for a certain desired resolution
the same mirrors would be able to direct the reflection of the light pulse back to a receiver
with the known pulse / beam location and the known ' static ' location of each reflective dot - it would be possible to calculate downward deflection of the ceiling by measuring the increased distance between neighboring dots which have been depressed due to the material being distorted
Possibly useful for gait analysis (weight bearing) in patients with arthritis or other lower limb issues (asymmetrical limb length for example leads to hip/spinal problems). Slight differences often go un-noticed by sufferers until they turn up at their GP with back-ache. Gait analysis is a technique commonly used in disease modeling where pain reduces the weight placed on one limb. More complex analysis compares actual x-y positioning of foot imprint and sub-areas where weight bearing varies (e.g., fallen arches, in/out-turned ankle joints and so on). So yes, far from being simply a bit of fun, I'm sure the medical department of your local university would be interested in something like this.
Great video
You honestly made this whole thing really entertaining to watch - Good Work!
Awesome project!
Great project! Maybe you can improve image quality by identifying the blur function and using deconvolution. It is much simpler than it sounds.
Lucas Hartmann that's a great idea, and it should be pretty easy to identify the blur by applying a point load.
Ooh! That's clever. I like it.
I adore the hanging FritzBox
Very interesting project
Huge project!
Is this the technology involved in pressure sensitive drawing tables?
Impressive project.
Marco's bizarre electronic adventures?!!
Very cool! I'd love to make an art project with it. Making an audiovisual interactive experience would be awesome with this :)
How would you get faster readings with a project like this? Are all external ADCs just too slow to get a super fast reading?
In the future, printable ink could be a good solution to replace the copper strips. You might just have to deal with lower sensor voltages.
Very cool series.
Do you know anything about printing such a circuit at home? I really wonder how hard this is. I read about Velostat and that it's inventor basically just mixed nickel shavings with some elastomer. I wonder if conductive plastic can be produced by varying the amount of metal you put in. This might be 3D-printable. How else could you print something like this?
Gude Marco!
First of all great great and very interesting project. I just wonder what it is exactly used for? I got the point with computerassisted physiotherapy, but could you please give me some more background info? Like, if a patient is lying down on there and you can see spots of higher pressure than usual you can conclude that they have a suboptimal posture?
I think this will also make a great device to analyze sleep behaviour and movement, and even check for REM phses and other stuff.
Whats the state of the project? I am really interested in this approach.
Keep up that good work and nice videos!
Greeting from middle germany ;)
That was amazing
Excellent. Well done
So fascinating!
Hi dear i am your new subscriber thanks for the beautiful Electronic art
Will be the best security system.
You're a genious !!... With a German sense of humor. I like it !
Hey Marco Reps! Can you show how to build a capacitive pressure sensor matrix?
OMFG that's a LOT of work!! If you had to do it all over again would you stick with this approach, or simply use an HX711 and four load cells (one under each leg of the bed)?
Hi, how did you deal with the velostat "creep" behavior? The reducing resistance without adding pressure?
This would make a good addition to a security system
amazing. well done
Awesome project. I can think of a million uses for an input device like that! I’m new to the channel but have been loving everything I’ve seen so far. Keep up the great videos man! 👍
I would've used a large wooden stick + right angle or a laser level from Ikea to lay out those strips instead of having those "separators". But I understand trying to get something quickly "out of the door".
The results look much better than I expected, would be interested to see if slowing down the ADC or using some faster ADC would yield better resolution. Even so, I guess it outpaces many commercial solutions that cost [tens of] thousands.
Not sure on the application, it might look like radiation therapy/imaging, but localized weight doesn't really translate to energy density. Would be great for developing mattresses though...