Wyze Cam v3/PTZ Ignoring SD Card? Here's the Real Fix (After Wyze Said I Was SOL)

:wrench: [FIXED] Wyze Cam v3/PTZ Ignoring SD Card? Here’s the Real Fix (After Wyze Said I Was SOL)

I recently ran into an issue where my Wyze Cam Pan v3 refused to recognize any SD card I threw at it — even freshly formatted, name-brand ones. Wyze support reviewed my ticket and told me I was basically out of luck and needed new hardware.

Spoiler: They were wrong. Here’s how I diagnosed it and a PowerShell script that fixed it cold.


:test_tube: Symptoms:

  • Camera would read from the SD card to update or roll-back firmware
  • But showed “No SD Card Installed” in the Wyze app
  • No local recording options available
  • Multiple SD cards tested (including 32GB SanDisk Ultra)
  • SD not compatible or not present symptoms.

:magnifying_glass_tilted_left: What I Tried:

  1. Formatted cards as FAT32 with 32KB allocation unit (using GUIFormat)
  2. Tested with multiple cards (32GB and 128GB)
  3. Rolled back firmware to 4.50.4.9222
  4. Did full factory resets
  5. Confirmed card slot worked (because it read the firmware update)

Still, the Wyze app refused to recognize the card.


:light_bulb: The Fix:

Turns out, the camera’s firmware wasn’t creating the expected folder structure on the SD card. Once I manually created the folders and added a dummy file to /record/, the camera started working immediately.

To make this easy for others, I wrote a PowerShell script that:

  • Formats the SD card to FAT32 with the correct settings
  • Creates the folder structure Wyze cams expect
  • Adds a dummy file to help the firmware recognize the card

:warning: Warning: This script erases all data on the selected SD card. Be sure to pick the correct drive letter!


:toolbox: Wyze SD Card Prep Script for Windows (PowerShell)

# Wyze SD Card Prep Tool - PowerShell Edition
# Formats the SD card and creates required Wyze folders

Write-Host "=== Wyze Cam SD Card Preparer ===" -ForegroundColor Cyan
$drive = Read-Host "Enter the drive letter of your SD card (e.g., E:)"

if (-not (Test-Path "$drive\")) {
    Write-Host "Drive $drive does not exist. Aborting." -ForegroundColor Red
    exit
}

# Confirm with user
$confirm = Read-Host "WARNING: This will format $drive and erase all data. Type 'YES' to continue"
if ($confirm -ne "YES") {
    Write-Host "Aborted by user." -ForegroundColor Yellow
    exit
}

# Format the drive as FAT32 with 32K allocation
Write-Host "Formatting $drive as FAT32 with 32K allocation..."
Format-Volume -DriveLetter $drive.Replace(":", "") -FileSystem FAT32 -AllocationUnitSize 32768 -Confirm:$false

Start-Sleep -Seconds 3

# Create Wyze folder structure
$folders = @("record", "time_lapse", "log")
foreach ($folder in $folders) {
    $path = Join-Path "$drive\" $folder
    New-Item -ItemType Directory -Path $path -Force | Out-Null
}

# Create dummy file inside /record/
Set-Content -Path "$drive\record\camera_ready.txt" -Value "Wyze SD Card Ready - Created on $(Get-Date)"

Write-Host "Done! SD card is now Wyze-ready." -ForegroundColor Green

Note that FAT32 may start to have issues on larger cards as it gets really really slow. 32GB should be fine and maybe even 64. EXFAT should work fine with cards >32GB, all the modern Wyze cams support it.

I’m curious if you tried a total wipe of the SD card using something like the sd card formatter from sdcard.org? If so and it still would not create the file structure on the card, you seem to have found the mystery reason why some of these Panv3s won’t recognize SD cards no matter what. Of course why it affects some cams and not others is still a mystery, none of mine have ever had that problem so there is something software or server side that is glitching out.

If you’re interested in doing some more sleuthing:

  1. If you use the “format” function in the app (then reboot the cam for good measure) does it recognize your card still?
  2. Same, but power off the cam, pull the card, format in a PC, and see if the cam will recognize it after that.
  3. If it stops recognizing it again, try formatting with EXFAT and doing your dummy folder/file to see if it works with that file system too (no reason it shouldn’t).

Basically curious if this is a one time thing that the cam just needs to be “woken up” or if it is an ongoing thing (and also want to confirm that people don’t need to do special FAT32 formatting, which won’t work in windows with >32GB unless you use Rufus or another utility).

For those not familiar with powershell, simply creating the folder in windows explorer (and maybe a dummy file like you did) should do it. Not sure if the dummy file is needed or not.

@crease don’t you have a panv3 with the problem?

I do, yeah, and I saw this topic pop up earlier. I did pull the card that’s been in the problem child Cam Pan v3 and check it, and it’s missing some of the subdirectories mentioned in @snapshotdc’s post. I have not yet pulled the card from the “good” Cam Pan v3 for comparison, but I bookmarked this topic to check back at a later point in time.

I can’t recall with absolute certainty whether or not those subdirectories existed on the microSD card in the “bad” Cam Pan v3, but I think that they might have because one of the cards I tried in that camera—which worked as expected in a Cam v4—should’ve had those subdirectories already written on the card, including time_lapse, as I had used the Cam v4 with that card specifically for that purpose at one point. Since that time, though, as part of my troubleshooting, I’ve done the “Overwrite format” with the SD Memory Card Formatter, so that blew all of that data away.

Another thing that makes me skeptical about this proposed solution is @dmorris68’s description of an experience with microSD problems in Cam Pan v3, especially because the timing of the issue and the way recording seems to stop once a certain capacity has been reached tends to align with my own experience. The timing is especially suspicious to me, because my Cam Pan v3 had been chugging along and doing continuous recording for months until mid-December, when @dmorris68 said the problem started. I think what @snapshotdc is worth checking out, so hopefully I can try it at some point, but given all the other factors I wonder if the addition of the file and folders was just coincidental. I don’t feel like I have enough data to be satisfied that this is a definitive answer.

I would also modify this solution in any attempt because I’m using 256 GB cards and would format those exFAT, so if I try this I’ll probably just do a fresh format of a card in a PC and then write the subdirectories myself. I like scripts, but I do most of my work on a machine running Linux, and just doing mkdir manually makes more sense for my one-off application.

@dave27, I think the additional testing you suggested is interesting, and I’ll be eager to see what @snapshotdc does with that and what those results might demonstrate.

Welcome to the Forum, @snapshotdc! :wave: That’s an impressive first post! :+1:

It is certainly possible that we’re dealing with two unique issues here. We know there have been several different SD related symptoms, for some of us we see an error on one screen but not on another, for some that error goes away if you wait a few seconds, others it doesn’t, and others the cam totally doesn’t recognize the SD card. So we may be seeing a mix of failed SD card slots and software bugs, seeming like the same issue but not actually being the case.

OP seemed to imply a dummy file needed to be created in the folder (not sure if that is actually the case or not), if so you can just “touch” an empty file with that name too.

I’m going to do a couple tests of my own on my Panv3s that are working fine (other than the few second message in the v3 app that there is no recording, but then the recording starts playing) to see what directories are on there before and after a format.

Yeah, or more. :grin: That’s one of the reasons I mentioned “all the other factors” in my previous post.

Yup. Been there, done that. I had exactly that thought when I was reading through the PowerShell script. :wink:

:+1:

I look forward to reading about your results, too.

Well here is a fairly long list of observations.
3x PanV3, all the same results
All 3 running latest firmware 4.50.14.3339
All 3 running plugin version 3.5.0.9
All 3 running 128GB SD cards formatted to EXFAT (originally using sdcard.org formatter, but they remain EXFAT after using the internal formatter too).
Cam #1 - Activation date 7/21/2023 - Been running 24x7 recording since then. Card last formatted 8/17/2023 (I believe using the internal formatter but not 100% sure).
Cam #2 - Activation date 8/5/2023 - Was used for around 6 months 24x7 but is now a spare that doesn’t get used much. The card was formatted today for testing.
Cam #3 - Newer cam that came with the newer version mount. Activation date 3/5/2024 - Running 24x7 recording and the last format was the original sdcard.org one on that same day.

Worth noting, all 3 no longer have an error when trying to view SD Card footage from an event. That seems to have been fixed, probably with the recent interface update or maybe the firmware release that came around the same time. So at least that is one variable that is gone from this (at least for people with latest firmware/app).

File system on the SD card:

/.ota/ (note the leading dot)
This directory only seems to be created on reboot/boot. It contains 1-12 files with no extension. I’m guessing configuration/authentication related stuff. I inserted a card while the camera was on and let it go about 45 mins and this directory is not on that card. But upon rebooting the cam, it is created. Maybe it would be created after a longer time running, or if you changed settings/reset services, not sure.

/alarm/YYYYMMDD/YYYYMMDD_HH_MM_SS.jpg
One subfolder for each day, up to 14 days, the files are the thumbnails from events, I’m assuming the ones that display in notifications and in the events list. Seems to be created when the first event is logged. I’m guessing if you have a ton of events it might create more than 1 folder for a day, but that’s just a guess. It probably wouldn’t be until like 999 or something.

/cores/xxx.tar.gz
This seems to be the installer for the webrtc client from the latest firmware. Not recreated on format or reboot so just a temporary directory/file it seems

/log/various .txt files
Not created on boot/reboot/card insertion/format, seems to get created as needed. The file contents are encoded/encrypted. However it is not only related to submitting a log, after running a cam for 45 mins tonight there was a file in there.

/photo/YYYYMMDD/YYYYMMDD_HH_MM_SS.jpg
Seems to get created when you take a snapshot in the app I’m guessing. It is not created on format/boot/reboot, only on demand, but then seems to stay indefinitely. I have some dating back to 2023. None in 2025, can’t recall if I’ve taken any pictures since then, maybe no longer used.

/rebootlog/
Only see this on cam #1, even though #3 has been running over a year without format. Maybe related to a firmware update or the original firmware. Not created on boot/reboot/card insertion Folder is empty on that one cam. Maybe obsolete.

/record/YYYYMMDD/HH/MM.MP4
The actual 1 minute video clips. Not sure if the file structure would be different for “event only” recording, I’m guessing it would need to be more specific, maybe like MM_SS.MP4
On cameras that are hard rebooted, a 20250422 subdirectory is created with 1 MP4 file in it, which we know is what happens when the camera starts recording when the clock hasn’t been set by NTP yet. Soft reboot does not create this phantom subfolder. In this case that must be the date that this latest firmware was built.
As far as I can tell this folder is never overwritten, so it becomes unusable space on the card. Not a big deal for most, but those with long internet outages might lose space until they format. On one cam, it has not been hard rebooted in months but that directory is still there with the 1 file in it.
For the other files, when the card gets full it seems it deletes and entire hour directory (which is why the card never looks 100% full), then when those are all gone it deletes the empty day directory.
Assuming you have SD card recording enabled, this is created on boot/reboot/card insertion.

/system volume information/
Hidden directory that windows creates when you insert the card in a PC, can be ignored, unrelated to Wyze

/time_lapse/
Empty for me, I’ve never used the feature. It is created as soon as the card is inserted or camera booted/rebooted

/tutklog/one .txt file
This one is weird, it ONLY seems to get created if you insert the card while the camera is on. It is not created on boot/reboot/format. The text file is not encrypted and just has some generic connection info. I’m not sure if it gets deleted at the next reboot or not.

In summary, the folders that seem to be created immediately on a freshly formatted card are
/record/
/time_lapse/
(and that tutklog one if you put the card in while the camera is on but it does not seem to be a core/necessary directory)

Really only the /record/ seems absolutely necessary for functionality of the card. If the problematic cams can’t create this directory, that’s a bit of a mystery as to why (and why they can create subdirectories and files within it if you create it manually, as @snapshotdc observed).

I think at this point we may potentially be looking at two issues
Cams with failed SD card slots
Cams that for some reason have a glitch that they don’t create that /record/ directory and thus can’t record anything.

I guess the only way for people to find out is manually create that folder on problematic cams and see if they start working (and continue working/overwriting when the card gets full).

I would be :confounded_face: if this was different on any of them, because I believe that’s an app component (not specific to a particular device), which makes sense given the app’s seemingly modular nature (different UIs for different Cam models, etc.).

That’s interesting that this is created on reboot, and your guesses make sense. When I’ve poked around on cards and seen this, I’ve thought .ota was maybe a holding area for “over-the-air” firmware updates, like where they’re saved before being written into the camera, but I think your observation about authentication is probably a better answer.

I don’t think it does. I think it still records the individual video files with MM.mp4 names, but some of those files may not actually be full 60-second recordings when played. I believe I’ve actually observed this, but don’t hold me to that.

That one’s problematic on Cam OGs. I’ve never tried using Timelapse on a Cam Pan v3.

The rest of your findings align with my thinking.

I wonder if it might be even more than that, because I’m still baffled by all the complaints last year about microSD card issues on non-Pan Cam v3s—affecting some cameras but not others—that were apparently resolved by some server-side changes, not a new app or firmware push. I don’t know enough about how this all works together to understand how that could be the case, but the way some of these complaints about Cam Pan v3 read, the situation seems very similar to me, which is why I still wonder about a potential future fix on the back end. I’m really just speculating at this point, though, because my own experience with this is so puzzling.

I may still pull the card from the “good” Cam Pan v3 at some point for comparison when I don’t have so many irons in the fire. If there’s a difference in the subdirectory structure, then it’s probably worth trying to create @snapshotdc’s findings on a card and then see what happens when I put that into the “bad” Cam Pan v3.

So a quick update. I removed the Fixed moniker. It was fixed and recording to the 32gb SD card, but I went forward with the firmware update, and behold! afterward the SD is no longer recognized.

Card was originally formatted (full-format) with GUIFormat.

When it stop’s raining here I’ll pull the card to see what artifacts I end up with. it should be worth noting that this camera is also configured with Cam Plus.

I thought that seemed pretty bold and perhaps premature. I appreciate the update! :+1:

I still might try this at some point, because it certainly can’t hurt anything—or at least I wouldn’t expect it to. I’m still stumped by this problem and hoping that at some point we can get Fix-It Fridays again to try to focus some more attention on it.

Bold but correct. Much like getting a flat, changing the tire, and then driving over the same nail. The issue seems to be in the most recent firmware update. After I was able to confirm that it was able to see the SD and record to it (both in app and by checking the card in the laptop). In all told I probably only had about 15 minutes of continuous camera capture on the card, so yes, premature for those with the not overwriting when full issue, since I didn’t test that. I’ll probably just roll the firmware back again, and leave it alone to see if I end up the overwriting issue though. At this point I’m not out anything other than my neighbors thinking that I’m crazy climbing a ladder at all hours.

I’ll be interested to learn how that works. I think that I used to have an option to confirm (proceed) or cancel firmware updates when they were offered on my Cam Pan v3s, but I believe this most recent update was applied automatically, and I don’t see a way in the app to turn that off. I haven’t tried manually flashing an older version, but that might be something else worth trying (and then seeing if the camera updates without my intervention anyway).

I appreciate your taking the time to test and share your findings! :+1:

This is a problem because…? :man_shrugging:

Yeah I just figured it was worth mentioning in case it did happen to be different for some, might point to something.

I was thinking it might need to be more specific since you can have more than 2 motion events in a minute, but it could also just use _1 , _2 etc too. I have a couple OGs that are motion only that I could check, but given that they seem to operate totally differently from the other cams, not sure it would actually be the same as the Panv3.

Would be curious if you tried the same fix again if it would start working again.

The firmware update process does appear to write some directories to the card, maybe that messed up the file system you created.

If it is a 32GB card, you don’t need anything special to format it, in fact some of the utilities can end up causing more problems. If you do decide to get up and remove the card, I’d actually suggest formatting it in the app, waiting like 5 or 10 minutes, then doing a “eject micro SD card” and pulling it out. Would be curious to see what is on your card vs mine to see if that points to anything.

Actually you could even try formatting it from the app then doing a soft reboot from the app to see if it starts working with the new firmware potentially, and if not then remove it and see what is there.

sorry, tl;dr, if this was the case wouldn’t it work to install and format the sd card in another camera and let it run for a bit, then move that card to the problem camera? Wouldn’t it then have the file structure and everything? I guess I could try that and see. Certainly not putting my ‘good’ card into the bad camera, I’ve got an extra 32gb or 2 cards lying around here somewhere

Yeah, in theory that should work if you happen to have a working Panv3 and this is in fact the problem that some other Panv3s are having.

anybody tried this/these? any luck?

I have not. My time has been occupied by other things, and this has been low priority, though it’s still on my list. Trying to use a different camera to create the subdirectory structure seems like a reasonable idea if that structure is the actual root of the problem. I’m not convinced that it is.

I tried a similar fix yesterday. I took the card from the PV3 with the non-working SD Card and put it into the working PV3 cam. It setup and recorded successfully. I let it run about an hour. Then put the camera into privacy mode, removed the card from the working Pv3 and put it into the non-working Pv3. It did not work over there :frowning: So that camera’s SD Slot or firmware or something is broke :frowning: Oh well, It looks cute in the garage and can track motion and record based on the motion detector in my garage

holy moly after reading all this really makes me want to return panv3 b4 giving it a fair chance. I’ve owned an earlier version camv3 no pan & have never had a problem with that camera ever. Seems like panv3 is headed in the direction of cam outdoor as being 1 of the worst cameras ever sold by wyze.

I’ve got 3 Panv3s and love them. Two have been running for two years outside. There doesn’t seem to be any rhyme or reason to the SD card issues, I did have the app bug for a while where accessing the card in a certain way would give an error, then play the video fine (but it was a way I rarely used). I think there are a mix of issues, some of them just being bad SD cards, others being that a long USB cable is being used which causes various problems since the pan cam needs more power than the others.

If yours isn’t having any problems, I wouldn’t return it just because of what you read here.