ฝัง
- เผยแพร่เมื่อ 14 พ.ค. 2024
- Radar is a valuable technology. Because of its unique features, it not only helped to win world war II. It also can solve many problems in our projects.
Sensors included: RCWL-0516 (3GHz), HB100 (10GHz), CDM324 (24GHz), FM24-NP100 FMCW Radar (24GHz)
I am a proud Patreon of @greatscottlab, @ElectroBOOM , @ELECTRONOOBS , @EEVblog , and others. No Docker, No Microsoft Teams, Zoom
These are the questions answered in this video:
- Which Radar sensors exist, and how do they have to be connected to our boards?
- How do they compare?
- How can we use them in our projects?
- Which other technologies exist to solve similar problems?
- And finally, we should have a comparison chart, also including PIR and Lidars
Links:
RCWL-0516 (Presence): s.click.aliexpress.com/e/_9y4W4C or amzn.to/3d82zvG or bit.ly/31mvXsD
HB100 (Speed): s.click.aliexpress.com/e/_ADfbUc or amzn.to/3d6Stem
CDM324 (Speed): s.click.aliexpress.com/e/_AZNnGU or amzn.to/3vZuGWx
FM24-NP100 24GHz (Distance): s.click.aliexpress.com/e/_AmT5cG
Diagram RCWL-0516: www.snapeda.com/parts/rcwl-91...
Info FM24-NP100: ebay.to/3swsfs4
Info speed sensors: bit.ly/3fhxFnd
Info HB100 presence detector: bit.ly/31qcGXg
ESP32 FFT sketch: github.com/G6EJD/ESP32-8266-A...
The links above usually are affiliate links that support the channel (no additional cost for you).
Supporting Material and Blog Page: www.sensorsiot.org
Github: www.github.com/sensorsiot
My Patreon Page: / andreasspiess
Discord: / discord
If you want to support the channel, please use the links below to start your shopping. No additional charges for you, but I get a commission of your purchases to buy new stuff for the channel
My Amazon.com shop: www.amazon.com/shop/andreassp...
For Banggood bit.ly/2jAQEf4
For AliExpress: bit.ly/2B0yTLL
For Amazon.de: amzn.to/2r0ZCYI
For Amazon UK: amzn.to/2mxBaJf
For ebay.com: ebay.to/2DuYXBp
profile.php?...
/ spiessa
www.instructables.com/member/...
Please do not try to email me. This communication channel is reserved for my primary job
As an Amazon Associate, I earn from qualifying purchases
#no#midroll#ads - วิทยาศาสตร์และเทคโนโลยี
Love your videos. Educational, practical, concise, perfectly structured, with subtle humour woven into the fabric.
Thanks. Glad you enjoy them!
I find your argument for preferring beer to water very interesting and original. Never considered that factor before. 😄🍺
You see, there is always something to learn ;-)
@@AndreasSpiess I think we need more -beer- data!
This needs extensive testing with different beers and volume.
Are you looking for a volunteer.
Beer is also proven to make you smarter, in line with Darwin's theories.
A herd of buffalo can only move as fast as the slowest buffalo. When the herd is hunted, the slowest and the weakest ones at the back are killed first. This natural selection is good for the herd as a whole, because the speed and health of the group keeps improving from the regular killing of the weakest members.
In a similar way, the human brain can only operate as fast as the slowest brain cells. Now, it's already proven that excessive intake of alcohol kills brain cells, and naturally it attacks the slowest and weakest brain cells first.
Therefore, regular consumption of beer eliminates the weaker brain cells, making the whole brain a faster and more efficient machine. This is why you always feel smarter after drinking a few beers. 🍺🍺
@John: Never heard this one, but sounds absolutely logical to me! I have to keep it in mind for future discussions ;-)
Thank you very informative. Short and to the point thank you for not adding unnecessary commentary.
Thank you. The comment about the beer was probably not absolutely necessary ;-)
I have been using the A111 from acconeer for some projects ( measure foiling height of a racing solar boat ). Fun to use and if you pick the right development board also very affordable. One could even use plastic lenses to change the opening angle vs distance .
Interesting. The datasheet says: Range 2m. What is your experience?
I love your Chanel. Thank you for inspiring a new generation of makers.
Glad you like it! So it is worth the time I invest.
I bought a TFMini lidar module when you showed it in your LIDAR video. It reached 15 m range with 1 cm accuracy, outdoors, with strong sunlight (the target was 1.5 m, white). It seems better compared to the radar sensor. Also it works indoors and was cheaper, around 40$. A good buy.
Truely a good buy! Thank you for sharing your experience.
Great video as always 👍😀
Thanks for sharing your experience with all of us👍😀
Glad you liked it! And thank you for your continued support!
Love this video, like your other videos you are a wealth of information! Thank you so much Andreas!
My pleasure! I hope everything is ok on your side.
Thanks for this very nice video!
Knowledge combined with humor, top mixture!
Glad you enjoyed it!
Excellent, this was helpful. I never could get the RCWL modules to work, my understanding of them was wrong. But now I have two CDM324's, this should be fun. Your old stuff is still good and interesting. Cool.
Thank you!
Thanks for this great vid and overview! I just had to retire my CDM324 which I used as a raindrop sensor. After several years in service it started to produce a lot of false positive. I have to investigate the root cause.
In my drawer still sits a RSM2650 with quadrature output. I will see how to continue.
As PIR sensors, the aquara zigbee motion sensors are interesting. They are very small and contain also kind of luxmeter. So, they are easy to hide and to integrate in a room.
This is the "expensive CDM324" I mentioned. I also should have one in the drawer but never used it. I bought it for our bicycle project.
Concerning the aquara zigbee sensors: I have two of them in my lab and they work very good
Beer is much better than water. And Rivella is much better than water too. Oh, I miss Switzerland! I'm coming back to Graubunden right after the pandemic. -Frank
Beer and Rivella: True.
And we hope we can travel to other countries ;-)
@@AndreasSpiess please let me invite you to Silicon Valley. I would be glad to introduce you to my friends and host you for a great meal. Or meet me in Shenzhen, the Disneyland of Chinese cities! I plan to go back there for my Esp32 wristwatch project once the pandemic is better here.
Thank you for the invitation! I plan to come to the next Hackaday Supercon whenever it is. And I also plan to go to Shenzen to visit a few companies I work with. So we have three possibilities to meet: US, Switzerland, or China ;-)
@@AndreasSpiess Marvelous. I am Frank Cohen, fcohen@votsh.com, I am working on an ESP32 Internet connected wrist watch to display home videos of my kids on my wrist. My previous maker project: startlingwatch.com. Seems to me there is a gap when working with ESP and Arduino projects that need to connect to a Cloud service. I'm intending to fill the gap with github.com/frankcohen/ReflectionsOS. Please send me an email and I will send you my details. Merci.
Your videos are always inspirational
Thank you!
Hello Andreas, thanks for the great video as usual. It is very informative and gives a good overview over radar sensors. By the way it is also possible to use an FSK modulation to measure the distance to moving objects. This makes it easier to filter out the environment in comparison to an FMCW distance measurement. Further the accuracy is better in comparison to FMCW because with FMCW the distance resolution is limited by the allowed usable bandwidth what is 200MHz at 24GHz. This gives a distance resolution of only 75cm for FMCW in the 24GHz band. With FSK it is possible to reach a much better resolution down to app. 5cm or even more. The K-LD7 from the company RFbeam located in Switzerland uses this modulation to measure the distance of moving objects. It is fully digital and has already a powerful FFT signal processing included.
Thank you for the info. I did not know about FSK in this application. Reading the documents, it seems, FSK only measures distance for moving targets. Is this true?
The module is interesting. I will try to get one for some tests. If it really detects cars at 30m, it would be very good.
@@AndreasSpiess Hello Andreas. Yes with FSK you can only detect moving objects and no static ones. With FSK it is only possible to separate objects over their speed and direction and not over distance in comparison to FMCW. This means that if you move the sensor itself all static objects will become moving objects with the same speed and you can separate them only over the direction -> For example if you are moving forward and pointing behind you all static objects will become moving objects with a negative speed and if a car or something is driving towards you it will have a positive speed. Because you can separate the objects over the direction it is possible to filter out the so called clutter to only detect the car (simple blind spot detection). With maximum sensitivity it is even possible to detect cars up to 50m as long as you work with the RAW target list and not the tracked one. RFbeam has also a simple Doppler movement detector and speed measurement sensor in its portfolio what is cheaper and has a better sensitivity (K-LD2).
Are you going out again for drinking??
No, I just want to appear on the radar
Exactly! Always innovation on this channel ;-)
The beer does help - if you have several - then the problems just go away!
That sounds like a justification for a couple of extra beers!! 🍻
Good idea! Now it gets warmer here. Seems to be perfect timing.
One nice thing about the RCWL-0516 I just found out is, that you can use Pin 5 to get the direct analog signal and write your own detection algorithm. Especially useful if a esp8266 etc. generates false positives on the circuit.
You can use the analog signal even for sleep tracking...
Btw nice video, as always :)
You are right, you get the analog signal on a BISS pin. I never looked in the false positives. It seems they are reported with Wi-Fi modules. According my video the sensors work around 3GHz and Wi-Fi is at 2.4GHz. So the false alarms most probably come from the Wi-Fi signals. So a good algorithm could be to switch WiFi only on after an alarm is triggered.
@@AndreasSpiess Thanks for the clarification.
For my purpose the disabled Wifi was not an option because I wanted to implement a presents detection over multiple rooms to turn on and off the light accordingly (as fast as possible). With the one analog pin on the nodemcu and some threshold fine tuning it now works pretty well :)
Good to know, thanks! I had already written off the RCWL-0516 for my project. So I can just connect the pin labelled CDS to an ADC on my ESP? Or did you mean pin 5 on the IC?
@@EatRawGarlic The IC pin. On the BISS0001 (very similar to the one used on the RCWL-0516) it is labeled "2OUT" pin. So you have to solder a bit ;)
(It is actually the pin 12 in the datasheet. They start counting at the bottom left in a counter clockwise direction).
@@AndreasSpiess or 5G, depends however on if the lower part of the n77/78 spectrum is used. I guess these modules have a pretty wide detection range.
Great video. Comment, pir vs radar. In alarm systems radar based detect best targets moving towards or away from sensor. Pir is detecting movement crossing sideways. Pir works like a 1 point temperature meter. The fresnel lens in front creates small areas in the view field. So that hot object moving across will induce big changes. Without the lense it would only measure average heat on area, and not notice movement inside area. Still would notice big heat entering the meter area.
Thank you for your additional information!
Who needs a Swiss knife when you have Andreas in your life !! Danke von Kanada meine freund !
Thank you! Greetings back to Canada.
Radar is a very interesting subject, I enjoyed this video for my Sunday tech fix, ie brain stimulus.
This video had me reminded of some badly engineered radar systems which had recorded nearby buildings doing over 100 kilometers per hour, that was quite some time ago that I read that somewhere.
The story still makes me smile.
Modern radar where they look at the phase angles of the reflected waves is amazing, resolving one wave length at those frequencies to within a few degrees is precision stuff.
With triangulation one can position stuff and monitor movement with precision.
Anyhow, rambling on, but it's so interesting.
Thanks for the video, have lovely Sunday further.
Kind regards,
South Africa
Yes, technology moves fast in the direction of more precision and higher speeds. Cars use already 70GHz radars which seem to have ranges of a few 100 meters.
The small range finders also work with phase shift detection, BTW and are also very precise.
Gracias por los subtítulos en español. Your videos are really very high tech. I like it.
Thank you! And I will tell Luciana, the Argentinean translator that you like her subtitles.
Your videos are fantastic!! :D
Glad you like them!
Muy bueno!! Muchas gracias!(from Spain)
You are welcome!
Hi Andreas, thank you, you are great. Now I understand what it takes to measure presence, speed, distance and direction. The only thing I don't understand is how multi-target detection works.
They evaluate the reflected signal and see several peaks.
Hi Andreas, thanks for the great video. I love your know how! Thanks Andreas 😘😘😘😘😘😘😘😘
My pleasure!
Excellent mime act at 2:40. A strong 7/10.
:-) I will tell him.
Thank you for your effort and work on this topic! I’m looking for a sensor that I can use to track golf shots! So it has to accurately find a small object , preferably also detect the golf club and it’s Speed, and Launch angle of the Golf Ball. As well as speed.
Sorry I clicked post too quick! Any information you have would be greatly appreciated !
You have to go to the golf pages. There they sell golf trainers which measure all the club movements. I am not sure if you also can get a device to track the ball.
Great video and great overview. However for me the price is also relevant and, as mentioned in the video, there are huge differences.
Good point! The price range is huge between the cheap presence detectors and a long range radar device.
An advantage of radar over PIR is that it will continue to work when ambient temperature is close to human body temperature.
I discovered this in the heat of last summer, when our outside pir sensors stopped working, becoming blind to body heat, even during the middle of the night.
True. This is what I meant in my table with the last column. But you explained it better than me.
Thanks. Useful material.
Glad you think so!
Maybe one beer wasn't enough, or the alcohol level was too low 🤣 Great video 👍
Well possible. Maybe I have to increase the dose for the next experiment ;-)
Very useful in 11:16 ,sensor table for right chose.
Glad it is helpful!
Hello Andreas! Thanks for the great video! I am also interested in radar technology and would like to clarify that there are radar sensors that can measure the vital signs of people, such as pulse and respiration.
You are right, there are many applications of this useful technology.
This is really useful…… Brilliant as usual
Thank you!
Hi Andreas. Hope you are doing well. I read that you asked people not to email you and so I'll ask you here. My dad is turning 70 this friday, the 16th and you are his favorite youtuber. He is a doctor (nephrologist), but recently (about seven years ago) he discovered a big interest in electronics and programming. It really makes my heart melt to see him excited when receives the notification that you have uploaded a new video, when he explains something that he learnt from you or he says your catch phrase "the guy with the Swiss accent". He also buys the gadgets you suggest through your links. Well, I am meaning to ask you if you could wish him an happy birthday. Not even through a video or anything complicated. A simple message would surprise him a lot. With covid, we have just been reaching out to the rest of the family and his friends to gather some messages and I'm sure it would mean the world to him, to see something from you in addition. Thank you so very much for your time and attention and hopefully you'll see this in time. :) Carolina
Contact me on Twitter or Facebook messenger. We will see what we can do... you find the Links in the description.
@@AndreasSpiess I really appreciate it! I contacted you on facebook messenger :) !
As always, the best channel I subscribe to and I’ll drink to that!
Thank you and Cheers!
@@AndreasSpiess do you have any plans to use BLE beacons to send data to MCU’s? I am attempting such a project with the Ruuvi tag
So far I have no such plans. You can buy such beacons for no money.
I had this gun thing intended for golf, which could tell speed over really long distances as well as measure long distances. It used laser, but no laser was visible.
I no longer have it, and I never really worked out how it worked so well. It worked in daylight up to 400m!
Honestly it scared me because I kept looking into the wrong side and feared I would injure my eye :)
I once made a video about a probably similar device. It was able to measure distances to 1500 meters in daylight if I remember right.
But it is interesting that yours worked with a flying golf ball! A very small target.
@@AndreasSpiess oh no, I don't play golf. I was measuring car speeds. I don't think it would work with a golf ball :)
Aha. Mine was also able to measure speed of cars. This seems to be quite "normal" these days. But it is definitively good not to look into the beam...
Nice shoes, I have the same pair. Very comfortable.
Yes they are!
The SMR-3x3 (333/313, etc) series sensors from InnoSent not only measure speed, but also distance, direction and have various angles of detection and are half inch size and are under $20
Interesting module. I did not find the range. Do you know?
Your contenet are very awesome!!
Thank you!
Automotive FMCW radars use a fast series of different frequency gradients. By comparing different frequency gradients you can measure speed and distance. In addition they have usually several antennas and can detect also the direction of the signal without any moving parts.
Sounds like great devices! I assume they are quite expensive for now. But maybe in a few years we can get them cheaper.
@@AndreasSpiess I'm not allowed to tell you exact prices. The complete sensor including an ECU that evaluates the sgnals and doing all things like calculateing if an autonmous braking is needed is definetly below 100€ of hardware costs. What I observe is that the next sensor generations become significantly better but not necessarely much cheaper. It seems that classical FMCW isn't used for actual/next sensor generations. They use a method called "chirp sequence".
Thank you for your information. I can imagine they use newer technology. Also phased arrays seem to be used to get some directional information.
Maybe we will get them from old cars in a few years ...
In addition to the mentioned frequencies, there are some sensors that operate in the 2.4 GHz ISM band. Automotive radar often uses 77 or 79 GHz.
So far I never saw 2.4GHz sensors. So good to know. 60, 77 and 79GHz seem to be common these days in the car industry. And I saw that they sell collision avoidance radars with 150m range for less than 50 dollars. Now we just have to hack them...
@@AndreasSpiess I bought a Siemens KMY24 about 20 yrs ago to play around with doppler radar. It works at 2.45 GHz.
Thanks for the great content. Have you seen, played with any of Vayyar's products? It would be interesting to see one of these in your hands..
No, I did not know these products before. But I also did not find a price for the Vayyar home. I think I heard once about the Walabot (a few years ago when it was introduced)
Ooh, we finally get reasonably cheap FMCW radars for hobby projects!
The UART on the FM24-NP100 sensor has an Rx pin. The factory manual says it's reserved. Maybe it could be used for modifying settings like radar power or update frequency?
I did not find any information on the input specs. But maybe I did not dig long enough.
@@AndreasSpiess I also did not dig past the manual. In fact, I hadn't found much information about it at all. Google gives me various links to shops selling it and not a single Github or other DIY project. I am thinking about contacting the manufacturer for some info.
I think your videos will always be useful and interesting to me, no matter what you're teaching. I want to open source experiment with WiFi radar ideas on a boat at sea (won't use it at harbour, will limit range, and I'm in a relatively empty part of the world (greetings from the west coast of somewhat north Canada)). Do you happen to know if there's any projects for this sort of thing, or similar with SDR? I used to work on fixing CAT scanners decades ago, and I've been getting thoughts on how a pause/chirp&listen/rotate antenna could be a sort of inside-out CAT scanner using similar matrix math (must all be out of patent by now and well understood, though not by myself necessarily). I want very cheap radar (and sonar would be similar) that can be analyzed and reacted to by a Pi or Jetson Nano that is also tied into AIS and a special boat network. Any thoughts?
I am by no means a radar specialist! They start now to build variable beam antennas for WiFi (MiMo, phased arrays). Maybe this opens possibilities for other applications? They should be reasonably priced because they are also used in consumer devices.
@@AndreasSpiess Thank you Andreas, I shall look into them.
Thanks for your sunday morning update on sensors, as always very informative. However you may want to mention that you're a guy with a Swiss GERMAN accent, as other parts of Switzerland do have their own funny English accent 😉
You are right. But those details are probably more important for us than for guys which confuse us with Sweden ;-)
Very useful video, thanks! Particularly the comparison chart of the various sensors. For the project I want to do, though, I need to connect an esp32 to a sensor that will trigger within a few milliseconds (5 max), so I can calculate speed, using either one or two sensors (with a distance to 10cm, or so, apart). So far, I haven't found anywhere that shows this information and I wondered if you had come across it in your research.
Ideally it would be directional and cheap. The RCWL-0516, HB100 and the CDM324 seem the obvious candidates but all of the sensors I have tried so far seem to have a sampling rate or refresh rate, which means that they cab be too slow to react to an object passing the sensor. most recently, I tried the GP2Y0A21YK0F but, again, it was too slow.
Any suggestions? 🤔
These sensors are not precise enough for this purpose. You either use the CDM324 to directly measure speed (I once made a video), or you use IR sensors to measure time between two points.
Thanks - I have been looking into the CDM324 but I think the IR sensors are too prone to light fluctuations for reliable use.@@AndreasSpiess
@@connalmcilwraith4367 Watch my IR video...
Great video!
Which one would YOU use for a driveway sensor to detect cars/humans walking up a long driveway surrounded be trees?
You can try a couple of the first category. But people report false alarms with ESP boards in their vicinity. Maybe the HB100 with electronics could be a choice. But I never tested its range.
@@AndreasSpiess I had some luck separating the ESP and the sensor. I wire a 3.5 mm audio socket into my enclosure and connect the IO pin and power to that. I have a matching socket on the enclosure for the sensor and they connect together with a standard 3.5 mm M to M cable. I also include a capacitor in the box with the sensor to level out power issues.In this way I can swap out sensors easily. Just make sure to power off the device prior to plugging or unplugging the sensor or cables.
Best Professor!
Wow, thanks!
Very nice video, thank you for sharing. In your expert opinion, what kind of sensor would measure distance through the glass of a car window?
I do not know if sensors using laser light travel through your glass or not and if radar has problems with coatings. So you have to try. Or find a project which did what you want to do.
Great tutorial. Did you look at new 60 GHz chips? From Acconeer, XM122? pulse coherent radar, and Infineon doppler, aim to compete with PIR so very cheap in quantity but hard to integrate, short range unless use dielectric lenses, very low power.
Interesting technology. So far I did not look at it. I am more interested in the longer range sensors. But it seems that the development towards highest frequencies is fast. So I will have an eye on it.
I didn’t read the previous comments about your shoes but I am assuming your fans highlighted how fit for purpose they are, thanks for a very informative video.🙂
You are welcome! There were different opinions about my shoes ;-)
Very interesting. I was hoping you'd eventually have more luck in measuring the speed of cars. It almost seems like it might be better to just buy a $99 bushnell and interface with it using an arduino
True. I once had a laser range finder but did not find a way to interface it to an Arduino
@@AndreasSpiess there's another guy on TH-cam who worked out how to interface with the Bushnell Velocity speed gun using an Arduino.
I'm probably going to try do the same thing, but with an ESP32
FMCW is not new. In the late 70's it was called chirp-FM radar (because the short FM sweep sounds like the chirp a bird makes). I know, it was part of my Master's thesis. The problem back then was the you needed to do the Fourier Transform to get the range to the target. The computers that could do back then were to big and heavy to fit on aircraft. My thesis was on an alternative technique that used hardware (multipliers and a CCD transversal filter) to do the Fourier Transform. I recall that one of the problem I had was the the frequency difference introduced by return response delay (caused by the range to target) was very similar in frequency to the doppler shift. This became an major problem when the transceiver was on something that was moving (like an airplane). The ground was usually much closer than the target so the return response from the ground would be doppler shifted and much stronger than the response from the target. I suspect that narrowing the beam using something like a phased array and/or fancy processing (like comparing the response time) could help discriminate between the two now.
You are right. Most of the basic technologies are old. What is mindboggling is how cheap we can build them today. We did the FFT with an ESP32 and it was faster than needed ( we only wanted to detect 25km/h, not airplanes ;-)
I would assume that the doppler effect is constant over a few chirps and therefore probably can be eliminated. But I am not an expert like you. My thesis was to increase the listening speed for news without increasing the frequency. Also a feature which is built-in in today's smartphones...
I'm thinking perhaps one beer just wasn't enough to get the results you were anticipating. I'd go for 3 to 9 beers and recheck the results. If that doesn't work then drink 3 more and turn on some good music and relax for a while. Also the shoes looked fine to me. If anyone complains. They can buy you some Hoka running shoes and resolve that problem. As always good video. .
Thank you for your wise advice ;-)
Hi
thank you for efforts and the information.
I have made a project to my daughter with microwave motion sensor to detect human behind a wall ,working well, but is there anyway to measure the distance for the human movement from the wall.
Appreciate ideas.
These sensors should give you distance information.
Great explanations. There is a guy in the USA developing some new lightweight instruments for experimental aircraft. One of them he showed a video of is a new LIDAR one that measures and counts with audio the distance to ground. The one he showed was only from 30ft but he says he will have a version that will do 500ft. If LIDAR is affected by light that makes it interesting as I thought it must have been some sort of RF based signal originally like you explained but if you have different lighting conditions wouldnt that affect the accuracy of the LIDAR measurement? I believe LIDAR is a laser based measurement so depending on the wavelength of the laser it would depend on what the lighting conditions were. Do you know if there is a specific wavelength they are using for LIDAR or is it a bit of a mixture depending on the manufacturer. One of his 500ft units would be fantastic for any aircraft...including mine :)
I made once a video where I had a device which used light to measure up to 1500m in daylight (LIDAR). But only if the target was bright and reflected back.
I assume the guy looked at those devices and decided to use ultrasound sensors. Maybe because typical "ground" does not reflect very well. Airplanes use Radar altimeters on 4.3GHz for that purpose.
I wear exactly the same shoes watching this video!
Boy reading comments before the entire video is funny.
I like them. But it seems they also have haters...
It would be helpful to have a video for visualizing this data and, very importantly, how to use multiple radar sensors for triangulation, etc. I need to record the data from objects moving over time in a very visible way.
So you have a great project! I have no such needs...
I have somewhere in my notes that you have reviewed the HFS-DC06 presence detector radar sensors before, which operate on 5.8GHz ISM.
You didn't mention it again? Or did you referene it under a different name? I am confused 😅
This sensor uses the same principle as the rcwl-0516 and it was covered in this video. I did not remember that it works on a different frequency. So I forgot to mention it in this video :-(
I found that Seeed now has various 24 and 60 radar. Some of them can remotely measure breathing and heart frequency. Some have fall detection and other distinguish between active and non mobile presence (like someone sleeping).
Maybe this could be interesting to re-visit the topic.
I Agree there is more to test in the future.
@@AndreasSpiess i hope you revisit this topic, very cool tech.
I see DM-19 doppler speed sensor going for under 50$ in aliexpress. 25 m max. range for vehicles and measuring up to 150 kmh. There is also DM-20 sensor with 40 m vehicle range but cost is 120$. I have many times thought about constructing some kind of car mounted sensor that would tell me the speed difference between me and the car driving in front. I could then use this speed difference information to adjust the non-adaptive cruise control to match.
Did you check it out? Does it work on the long distance? For me, the chip is tiny for the power it should produce.
Mandatory comment about the shoes ;)
If not about shoes - what should we debate about ?
:-)
Not influenced by light well this would have become stupid me hand a few years ago :), when I was trying to detect of the blinds are closed with an infra-red sensor, forgetting that sunlight has also infra-red.
The sensor was always triggered :).
I still have the sensor but recently unsoldered everything.
One day I will take a go at it again, now knowing more about sensors and considering using libraries and datasheet :D.
IR remote controls use modulated IR (38kHz). Because sunlight is stable it is easy to distinguish between them. You even get IR diodes with 38kHz filters built in for cheap. I once made a video about it if I remember right.
your shoes are amazing!
:-)
Considered putting the sensor on an oscillating stepper or servo and trying to map the environment? Might also give a better idea of the directional characteristics. Have another beer or two. That will make it work, for sure.
Another viewer found the reading frequency of the distance radar: 10 Hz. Probably a little slow, but it would be cool to have an "air traffic radar" at home...
@@AndreasSpiess especially with all the drones zooming around inside the house.... DUCK!
Any guess what module Garmin might use for their Varia RTL516? The range is quite impressive. I tried to build a similar car approximation detector with a LIDAR Light v3, however this seems to be limited to 40m only.
I also wonder how they get the 150m. Maybe I should buy one…
Crocs with socks most comfortable shoes hands down.
:-)
Hello Mr Andreas Spiess,
i am looking now for some time for an alternative for the common rainsensors that work by water conducting on a plated board. But there is problems with corrosion and these sensors are not very precise when it goes to tell when it _stopped_ to rain because water stays on them for some time. Would these Radar Sensors be an alternative to that? I have looked up a lot and could only find very expensive equipment for professional meteorologists. Maybe you have an idea.
My rain sensor works mechanically. And I know that some Makers experimented with radar. ( github.com/chunter1/precipitationSensorESP32 )
Greetings and thank you very much for this video.
I have a question, I would like to measure the movement of a piano key, it is a very short distance in the order of mm, what sensors could I consider taking into account an accuracy/price ratio?
Thank u so much in advance!
I do not know what parameters you want to measure. Just contact. Or speed? I would search for similar DIY projects.
@@AndreasSpiess Thank you very much for your reply. I would like to measure the speed. They use a CNY70 in this project, and I was wondering if you can think of any sensors that might work better. th-cam.com/video/PLwpEU9NLPA/w-d-xo.html&ab_channel=CyberGene
Maybe you have a look at the VL6180X. It has only a range up to 10cm.
hi Andreas, thanks for the video, do you think is possible to scrape off the copper part that connects the transmitter and receiver of the FM24-NP100 module at small antenna, solder an sma connector and attach a more gain MIMO antenna to it to increase the mesurement distance, I could also opt for a small power amplifier to increase transmittin power by a few milliwatts, let's give 3 or 4mw, the point is that I need 70 meters for my project.
having said that, do you think it can be done or is the limitation of the distance is a problem more related to the difference in measurable frequency than to the power?
Thanks in advance
These are very short wavelength and antennas are difficult to build. I would not try it myself. The whole setup of those sensors are not made for long range. Look at the long range sensors. Their antennas are much bigger and they use much more power. And they are very expensive :-(
@@AndreasSpiess hello andreas, thanks for the reply, I searched a bit on the net, and yes,they are actually expensive, you wouldn't have any links of the cheapest you could find, thanks again
I am not in such high frequencies (yet)...
An easy trick with the CDM320 / HB100 to know direction of movement and looking at the VOUT amplitude with time: if going up it's likely that the object is moving towards you, and vice versa
You are right. This is what we tried to do in our bicycle detector. But it was not as easy as we thought...
@@AndreasSpiess FYI I'm the guy behind limpkin.fr... I have a new board on my desk with a new amplification chain and a stm32 to do an FFT on a CDM324 amplified output :)
Cool! Your page is very good. Where do I find the info about this board? This looks very interesting! What is the range you get?
Maybe you send me a message to exchange the email addresses? BTW: I was not aware that you live in Switzerland...
@@AndreasSpiess I'll publish it all on github the moment I can check it actually works! I can send you one prototype if you'd like... I'll send you a message then!
@@AndreasSpiess Nearly 2 years ago... i finally finished it! want me to send you one?
Andreas, your radar sensor series inspired me to build a detector to know when my cat is at the door, and want to come in. Problem is that it works to good. I mounted the sensor inside a bean can and then placed it tightly mounted on the wall with the opening of the can facing outside direction, i was thinking it would make it directional and that no reflection could interfere, but no, still detecting me pasing from behind. Is the metal can now acting as an antenna? Any idea how to fix this? maybe a video?
As I mentioned: The cheap presence detectors are receivers and are very omnidirectional. I never experimented with them. Connecting the can to the ground pin is the only idea I would have.
@@AndreasSpiess Brilliant idea! It worked :-) Thank you so much Andreas. Happy easter to you and your family.
I need your help with some issue, I want to measure the clothes wet/dry using a microwave sensor just thinking about how to brake any sensor famous in the market to sender and receiver and I put the clothes in between
These sensors use integrated chips with transmitters and receivers. I do not think you get the field strength out of them because this is not needed for their typical application.
1:35 actually, this depends. The AirTag for example uses frequencies above 7 GHz as far as I’m aware, but the power must be strictly limited, usually to -35 - 0dBm @ 50 MHz bandwidth.
At least the BNetzA states power limits for UWB, but it‘s not in the ISM section. I wonder if the 3 GHz ones will cause trouble with 5G n78.
UWB is a very different topic because its bandwidth is extremely big (GHz) and it includes bands that are used by other services. With the coin cell of an Airtag you are not capable to bridge more than a few meters. So no big problem in my opinion.
I want to make a fast rotating Ka band radar detector with a super high polling rate so I can have really accurate 360 degree direction finding compared to the radar detectors on the market today that only have front or rear direction finding.
Interesting project!
Nice video and a very good explaination. But how about your shoes? 😄
:-))
I have a question
...could this doppler sensor, be used to detect infrared emitters pointed out towards it ?
Thanks, i hope to have asked correctly
Infrared is light and these radars use a completely different frequency. So they do not "see" each-other
nice video, but one question, shouldn't be received and transmitted signal graphs named another way around, in FMCW section, cause i think received signal is the one which is late on time axis which should be in front transmitted signal. thank you
The received signal arrives a bit after the transmitted signal. Maybe I made a mistake.
What would it take to detect the presence and positions of birds in a vineyard? I can do it with cameras and machine learning but I think radar would be way cooler. Cheers and thanks for another great video.
You probably would need an expensive sensor because of the range and the size of the birds.
Very interesting devices. Which is sampling frequency? I.e. could they be placed on a rotating base to obtain a radio lidar?
I mean the rate at which the measurement is given by the sensor (the last one)
FM24-NP100 no good. Only 10Hz
Thank you for checking it out. I did not know the speed. But guessing from serial output it produces a few results per seconds.
I was playing around with the little cheap radar devices with the goal of building another tool for ny daughter and niece who play at ghost busting. I am afraid I must, however take a break as I recently found that the pain in my back is not from the injury that has been paining me for over 20 years but a cancer tumor larger then a grapefruit on my right kidney. It must be removed as soon as the doctors can schedule the surgery, in early December, they say now. So I may well be out of it for some time, I am in hopes that this is the only cancer they find when they operate, they believe it is, so this is a good sign they tell me. I look forward to more projects when the healing is over and I can get back into electronics. So until then, please keep me in mind. Thanks
Bad luck! I hope that all will go well with your surgery, and you are ok till Christmas with your family!
Se puede dirigir (sensor 0516 ) a una zona en concreto de unos 60º.
Can anyone suggest the Lidar sensor for project purpose
Hi, Any video on sensors for detection of water in muncipal water pipeline. When muncipal releases water, passive sensor detect water in pipeline checks overhead water tank level, if empty starts water pump and fills the tank.
For water level many people use ultrasound. Radar also should work. For flow it is different. I never tried this one.
Thanks for your nices videos. Whicn device a need to mesure altitude under 1000 meters above the ground ?
I once made a video about a device which was able to measure up to 1500 meters in daylight. Unfortunately it had no interface. But maybe you find such devices with interfaces now. That should work for your use case.
@@AndreasSpiess Thanks for this informations but I need also by night alitude mesurement, so daylight is not appropriate.
The worked also during night. But it was hard to aim. Aiming would probably not be a problem for your use case.
Hello Mr Spiess,
Nice video.
I have a question please.
I wish to protect my fruit trees from air borne and ground attack animals at a distance over a wide area without great expense or use of poisons.
PRS radar will do this I beleive.
Question can your device upgrade to become a transceiver at 5 watts rf ,fmcw - pulse doppler FMCW with high power - low power mode .
The anntena could use your aerials, linked together maybe in a stacked array that is swr stable increasing radar abilities .
I guess the alarm output could be a tone or a strobe at the unit.
Programing the device would be I bit tricky.
Probably not you normal question Mr Speiss
Thanks for your shows.
Andrew Nz
These sensors do not have a long range and it cannot be increased. You have to buy the expensive stuff if you need more :-(
8:43 Only with stationary objects. Moving objects cause a change in frequency because of the Doppler effect.
How much does that influence the outcome?
You are right. I do not know how big the influence is because I do not know the rate of the frequency change.
Hi, I want to use a CDM324 to detect the presence of people in a room like the DFRobot mmWave Presence Detection Sensor does, so I'm not intrested in messuring speed at all. What would I have to change on the circuit around the sensor to just detect presence of humans?
Maybe you watch my today's video?
Thanks! It is interesting to know that the Aliexpress website and probably most similar websites only sell consumer products. When asked, Aliexpress said that to buy using a company, you must register as a Dropshipper, that is, you will only pass the product on to the consumer, there is no way for a company to buy for use on these websites, which is ridiculous.
I do not know their business processes. I only purchase as an end-consumer...
Would you consider letting us put up the funding for some longer distance sensors? I'd like to come up with one, ideally with speed detection that could cover 300-500 yards. I could also write software to make note of objects in the environment and then when something appears or disappears it could trigger an alarm. This would be in an outdoor desert environment and likely rotating.
What you described is how fully blown radar operates as for those distances you need decent transmiter say magnetron from microwave owen, build a crude antena and mount everything with a sensor on a pedestal that can rotate, range will be limited not so much by your sensor but more by your transmitter and antenna.
After making this video I researched a little into collision avoidance radars for cars. They claim a reach of up to 150m and cost less than 50 dollars. Which seems to be a good deal. Now we have to find if somebody reverse-engineered the CAN protocol.
@@AndreasSpiess Hi Andreas, Thank you for this video. I am interested in some solutions that have longer distance capabilities like what you have mentioned here. Could you please share a link to the product you are talking about? Thank you -Drew
@@drewbeathard3811 I do not have a product in mind, I just saw a few on Aliexpress (for car speed warning signals)
Do you know of a sensor that detects non-moving presence such as the mmWave sensor in the Aqara FP1? I’d like to build something similar to this so I can detect presence of non-moving humans in a room. Very useful for the living room as people may be present but watching TV and not moving. Cheers 😊
It detects conducting material like water.
Why is the FM24 indoors in brackets? I am looking for a good sensor to do home automations :)
As said in the video: Because of reflections.
Very interesting georadar the theme
Unfortunately, I do not understand what you mean
@@AndreasSpiess it would be interesting to build a georadar based on such sensors
Aha. I do not know how much power these devices need because they have to penetrate material. Maybe these sensors are not powerful enough.
8:26 Are the labels for transmitted and received signals the correct way round?
No. Too much beer ;-)
@@AndreasSpiess I had wondered if the graph was in anti-particle space, where time runs backwards.
Perfect, now I can ditect when missiles are targetting my house.
Probably too late (because of the small range)...
Oh those cheap RCWL sensors don't contain a transmitter? I honestly am very confused with that for years, I always thought they also work on some kind of doppler effect.
As you see in the diagram they only contain one active element. The BIS00001 has nothing to do with the radar frequencies.
@@AndreasSpiess Its basically a large range capacitive detection sensor then?
It is a resonator which is influenced by the environment. It is well possible that it the effect is capacitive, But I am no specialist in regenerative receivers. I just tell things I read.
10:29 Andreas Stealth! :D
:-))
Is it possible to calculate distance in millimetres with some radar module?
You mean measure mm distance? These modules do not do that.
@@AndreasSpiess would be very interesting to see some radio modules with triangulation capabilities
The problem that I (and many others) had with the sensors in the original video is the number of false positives that they generate - especially when connected to ESP32 etc boards. No amount of power supply filtering seems to make much difference. This makes them all but useless for most purposes except trivial ones such as turning on a light when someone enters a room, where a false positive isn't the end of the world. There are interesting articles on the web that integrate the analogue voltage output from pin 12 of BISS0001 - processing this varying voltage might provide useful information for makers' projects
That's disappointing. Do the false positives occur more than 3 in a row? I'm wondering if detecting 4 (or more) in a row would filter out false positives. I want to make a driveway sensor.
@@TheRainHarvester Usually no more than 2 'false triggers' in a row. BUT .. As long as there is actual movement then output will remain high - so you can't simply count 4 activations. I guess that you could check the line after 10 or more seconds and see if its still high. [The output is 'held high for 5 seconds after normal activation] but that adds a delay to activation. For just detecting room occupation I've found PIR modules much more reliable (plus they use less power). In fact, I have recently mostly moved to ZigBee based PIR sensors. The Aqara PIR devices cost less on Aliexpress than an ESP32/PIR module & power supply, are battery based and integrate well with Zigbe2MQTT.
I agree with Chris: The Aquara are great devices for indoors (I use them, too). I saw the discussions on the false alarms with the ESP32/radar sensors a few years ago. I did not follow those discussions. I do not know if the Aquaras work on a driveway because they are triggered by temperature differences.
@@AndreasSpiess I use the Aqaras outside. I have two. Both are still on their original batteries after 18 months, which is surprising. They are both protected from rain and direct sunlight (underneath a porch and a window top). I've never had a false alarm. I also have an Ikea PIR sensor outside that has worked perfectly for the past 3 months. I tried the very cheap xiaomi PIR sensor indoors but that does have various issues, including false triggers.
@Chris: Thanks for the info. The battery live is unbelievable!