I am looking to replce some of my motuon sensors with mmwave sensors but have a couple questions.
Thanks!
Hi Terri, I canāt answer the first question, bit I can add my 2c for the others.
The sensors I have ( Smart Life Zigbee Human Presence Detector Tuya Wifi Mmwave Radar Pir Montion Sensor With Luminance Detection For Alexa, Google - Smart Human Body Sensors - AliExpress) and Iām pretty sure most others, allow you to set the min/max range for detection. If you use the ceiling mounted ones, you could set the max range to be 1m or so above the ground so a dog would not set it off. Trial and error required to get it just so.
The sensors I have are Tuya devices. There are drivers by @iquix and @w35l3y that both work well with them and allow you to tweak the settings.
They can have problems with ceiling fans which are running because those can create air ripples which caused the sensor to detect it as a person breathing. But this varies from Model to Model and Sometimes you can solve the issue with zones. So sometimes itās just a matter of trying it to see how it goes for a specific sensor in a specific location.
Some models have a swivel base, which makes it much easier to adjust the area of detection.
The following is a very good technical review of some models, although not all of the ones that you mentioned:
At the present time, I believe that large dogs will trigger them in exactly the same way that people will. Aqara has promised a future update with a pet detection filter, but itās not here yet.
Iāve gotten ZY-M100-S to work with smartthings by modifying the Tuya Personal Driver. It works ok but you have to still move a little for it to detect presence. It technically uses microwave detection instead of mmwave. The distance it can detect is too noisy for any automation.
Most of the mmwave sensors require a hardwire to an outlet.
I liked reviews of the aqara fp2 because it supposedly supports zone detection. It isnāt compatible with smartthings I think. Fp1 might be ok, but read that you have to reset it manually of it gets stuck.
Iām interested in what you have found with this. I have a wall-mounted ZY-M100 set to monitor from 0.25-1.5m to detect when I am at my desk. It works flawlessly for my purposes. I havenāt measured exactly how accurate it is, but if I stand 2m from the wall, it does not trigger.
The limits work fine for binary presence detection. I set it to max for room detection. I was trying to use the detected distance value to emulate zones, but found the data to be too noisy to do anything useful with it.
The Aquara FP1 does not get stuck any more, all Aqara devices that I have now perform normally with Smartthings.
The fp1 is still about double the price of the ZY-M100. Does it report distance info reliably? i.e., trigger automations based on distance from the sensor rather than just present or not present?
No, you canāt set distance trigger on the FP1 for precense, you can only set target distance for precense status to Far (5m), medium (3m) or near (1m), what I can tell you is so accurate, I will receive this week a ZY-M100 from AliExpress to compare them.
For the main topic, I read it has issues with fans but no tested yet on any room with fans, I can tell you my Robot Vaccum triggered precense status any time it enters on the target space. And I believe the ZY-M100 is a very value for the price (if works as intended, and for the comments I bet it works)
I have an EP1.
Itās currently hooked up to my home assistant install.
I have a fan on the floor that runs pretty much constantly in the warmer weather.
Iāve noticed that overnight, the mmwave and occupancy sensors are constantly being triggered.
Iām assuming itās the fan that is causing the triggers.
I am using this over my tower fan which is pointing to my sofa and find it to set the status as present only we sit in the sofa so this is perfect but then it goes to āNot presentā after a few min even when we are still sitting in the sofa. Looks like it is expecting a small moment at least which I am not sure makes it a true mmWave? Any suggestions for this?
Hi Mike,
I am interested in trying out the presence sensor in the link you provided. However, I have never had any tuya device in the past. Presumably, this Tuya sensor be added to ST hub directly using Edge driver? No need to go via Smart Life etc?
Another question from the link, one can choose between the Tuya WiFi or ZigBee versions? Tuya wifi is the one you get it working?
Cheers,
You can purchase the ZigBee version and then use āPersonal Tuya Devicesā Edge driver, with that you can pair it directly to Smartthinngs hub.
For the wifi version I am not sure if they are available through the smart life cloud integration to Smartthings so I canāt tell you if will work or not (for sure will not work with Edge Drivers).
Try the last driver version (updated today July 8) from @w35l3y, set fading time to more than 90 and you should be fine, just tested and is working great.
Awesome and appreciate the explanation. Just ordered and canāt wait to try it out.
A bit late replying, but yes, I hand the zigbee version. You can use either @w35l3y Tuya personal driver or @iquix already has one that works.
I do have a Tuya bridge that I connect to initially just to check for firmware uodates. Then I remove them from the Tuya network and add speculum to SmartThings
Many thanks Almeyda and Wesley.
Almeyda - please can you confirm what the fading time is exactly for because I thought it was the time to go between present and not present? What value did you set it to and how did you observe the difference.
Regarding the detection delay, I think the value is in seconds right?
Also, is this change done in the beta driver?
Hi man,
Yes, fading time is the āno detection delayā, looks like actually the values are in seconds/10 for both detection delay and fading time, I observed the difference because I have one on my TV room, so before the update after less than a second that I leave the room the status changed to āno presentā but also that caused a lot of false negatives when I was just sitting on the couch for a while, so with this new change I just tested to set fading time to 90 initially, so I leave the couch and it takes around 20 seconds to confirm āno presentā status (I read in another forums that fading time will take a little bit longer than set values), that caused less false negatives, my actual set is 150, and it takes around 30 seconds to change status.
For your other question, this was done in the main driver, not sure if it was done first in the Betta channel.
Man, I didnāt find any driver for mmwave sensor from iquix, do you have the link?