Wyze Cam Pan V4 - Spotlight - Camera Factory Reset By Itself

Cameras
,
1

My rooftop Wyze Cam Pan V4 has been working for months. Last night, I was looking around the roof with the spotlight at all the newly fallen snow, then I panned over, and the camera image froze. The camera lost all network connection.

I can pull power inside my house to power cycle this rooftop camera, which has resolved this issue in the past. However, this time, the camera completely reset itself back to its factory defaults without a network connection. How do I know this? Because I can faintly hear it on the roof saying, “Ready to Connect” every minute or so.

Unfortunately, with all the snow on the roof, it’s not safe for me to get near this camera’s rooftop position to set up the camera’s network settings.

How does a Wyze camera factory reset itself on its own? I guess that the increased electrical current draw of the spotlight, combined with pan motor movement, caused some voltage sag brownout. Still, this should not cause a factory reset.

2

I’m sorry to hear this happened, especially given the current rooftop conditions.

A power cycle or voltage fluctuation should not normally trigger a factory reset, so we’d like to better understand what occurred on the device side. Could you please help us retrieve the device logs from the app? Enter the Pan v4 Device Settings → Wyze Support → Submit a Log and paste the log ID here please. The logs will allow us to review what happened right before the disconnect and reset.

Once we receive the logs, we’ll analyze them to determine what may have triggered the reset behavior.

We appreciate your patience, and please do not attempt to access the rooftop until it’s safe to do so.

3

Thanks for the reply. I was able to reconnect this rooftop camera by going to the bedroom closest to the camera and standing on the bed to get my phone as close as possible. Luckily, the reconnection signal was able to go through the metal roof and reconnect this camera to my WiFi. I’m very glad the newer Wyze cameras can perform wireless setups now without needing to optically scan a QR code!

I submitted Log ID: 2221230

Here’s some more background:

This WyzeCam Pan V4 can frequently hang or reboot itself if I pan or tilt when the spotlight is on. If it reboots, then I only need to wait a minute or two for the camera to become available again. If it hangs, then a power cycle is required to get it working again. However, this time it hung—and no amount of power cycling would reconnect it. As I discovered in the morning, the camera had factory reset itself! Because I could faintly hear the periodic “Ready to Connect” audio from the roof. I was able to relink the camera to my WiFi as I described above without having to climb onto the roof.

4

Your issue sounds like a power issue. How long is the USB cable to the camera? If longer than 15 feet, or if the wire gauge in the cable is too small, it can suffer voltage drop and panning with the spotlight on certainly needs a decent amount of power.

Sometimes you can overcome this by using a larger power supply (say 3000 or 3200MA instead of the stock 2000MA one) but generally this only helps a little bit and the preferred fix is to shorten the USB cable, or use a larger gauge one.

5

Thanks for the reply. Yes, this is a longer cable run at about 9 meters. I used a thicker 14AWG for the main run to reduce resistance and voltage drop at the far end. But it still could be an issue. A larger power output USB power adapter could help. But, if the problem is voltage drop at the far end, the best solution is to use a power supply with, say, a 2% increase in voltage to like 5.2V to help compensate for any voltage drop. I will need to make an actual measurement of the voltage at the roof to determine the best voltage on the power supply.

I had theorized that a voltage drop occurs when panning with the spotlight enabled. But that shouldn’t cause a factory reset of the camera’s settings. Will see what Wyze has to say about that.

6

I wouldn’t recommend increasing the voltage of the power supply as when the cam is idle it could be too high and eventually cause damage.

The 9M cable is too long, that’s almost definitely the reason the cam is having the problems you’re seeing. If you found a really heavy gauge USB cable it might reduce the drop enough, but I wouldn’t waste time and money on that when it might not work.

I’d recommend extending the AC power closer, or using POE (there are a couple threads on that, including one I just posted this morning).

7

If the no-load DC voltage at the end of the line is 4.9V, then it’s okay to increase the voltage at the source to get 5.0V—and that source voltage might be 5.1V.

The 9-meter length is not very long as cabling goes, even for low-voltage DC. That 9-meter length is very doable with a large enough gauge copper wire, along with increasing the source voltage to get the no-load 5.0V at the end of the line as necessary.

At worst, I would put a small conformal-coated weather-proofed buck voltage step-down converter at the end. Then send 12V (or even 24V) down the line, and have the step-down converter output exactly 5.0V for the camera, and pin it at that voltage. Could also place a buffering capacitor to help stabilize the voltage when the camera peaks the power demand. It likely won’t come to that, though.

8

If you add 0.1 or 0.2V to get the no load voltage up, that’s probably not going to be enough to account for the 0.5 to 1 that it is seeing under heavy load. Keep in mind it is not linear. Bumping enough to eliminate your issues will likely cause the no load voltage to ruin the cam or at the very least continue causing you odd behaviors and headaches.

USB spec allows 15 feet (less than 5M). You’re trying to do almost double that. Some have managed to get 20 or even 25 feet but the cam you’re using is one that draws double the current of other cams, and most people find around 25 feet is where problems begin even with the 1 amp cams. Some can’t even get it to go that far. Keep in mind the wire gauge in USB cables is very small, it is pretty difficult to find the heavier gauge ones these days and even those are going to be in the high 20s.

By all means if you want to get the largest gauge USB cable you can find and mess around with voltage converters, that’s your prerogative. But since there are safe, ready made solutions easily available that will likely cost less, not sure why you’d want to go that route.

The PoE solution will do exactly what you’re talking about, it sends 48V to the converter then steps (regulates) it down to a steady 5 the automatically adjusts for load, temperature, etc. And it will run over pretty much any ethernet cabling you have around or can find cheap.

9

Keep in mind that the USB 2.0 spec of 5 meters is limited mostly by the need for data transfer. But the WyzeCams only use the +V and GND pins without needing any data from their USB port since their data is all wireless.

Also, both the USB and Ethernet specs use tiny 28AWG conductors, which are over 25 times smaller than 14AWG conductors. This puts a 9-meter cable run within reach.

I’ll first add a capacitor on the roof end of the cable. Failing that, I’ll check the voltage on the roof and increase the source voltage as needed. Odds are that it will fix the problem. Failing that, I will go with a more complex buck step-down converter with a higher voltage cable run of at least 12V. All of these I have in my toolbox at the moment.

10

The reason data suffers beyond 15 feet is the same reason voltage does. Data is sent using DC effectively.

28AWG is not a standard for either USB or ethernet. You’ll find larger and smaller of each. In order to pass POE+++ you need 23awg ethernet. 24 is pretty common even prior to PoE being around.

The reason PoE can deal with it much better is because it sends a much higher voltage (48V) and the two ends actively negotiate with each other to adjust as needed. The original passive PoE doesn’t do this but the devices are designed with a decent operating voltage range to account for that. The fluctuations aren’t that much due to the higher voltage and lower current. Wire gauge impacts current capacity much more than voltage.

But as I said, approach it how you see fit. I definitely would not just add voltage at the supply, sending a high voltage with a 5v regulator at the other end is the way to go. A simple step down will probably handle it but having a regulator IC in there will keep things very consistent.

Regardless of math and formulas, you can just look at the people here who have had very similar issues to you with long USB cable runs.

I’m all for hacking things together and building stuff, but when an affordable pre-made solution exists, over the years I’ve learned it is just easier. As an example my 20 year old snowblower, I wanted more light, it outputs 12VAC to an incandescent bulb. I got an LED fixture for it and built a full wave rectifier with 4 big output capacitors. Buried the whole thing in hot glue. It lasted years but eventually the cold and vibrations broke the leads off the capacitors. A $15 AC to DC regulator from amazon took 5 mins to swap in and has been solid since, and since it uses a regulator instead of simple rectifier, it actually puts out more light (only have 1 amp to work with and the regulator is much more efficient) and less flicker.

11

Dave, thanks for your suggestions. Your PoE suggestion is definitely worth looking into as a plug-and-play solution that will probably work 100% of the time.

But, as it is, I already have my 14-gauge cable run through to the roof so will continue to use the advantages of a bigger cable. I already have the other items: capacitors, step-down voltage regulators, and a 24-volt power supply.

After doing some digging, I will actually use your first recommendation: get a bigger, higher-quality USB adapter. I’m reading that some of these already have a higher 5.2V no-load power output.

Then I’ll likely also add a cheap little 1000µF to 2200µF electrolytic capacitor near the webcam. I have a box full of these capacitors. When the motors on the webcam initially move from standing still, they can create a very brief dead-short voltage drop that a capacitor can really help smooth out.

Some other technical things…

That 14-gauge cable is still an order of magnitude larger in area than either 24-, 26-, or 28-gauge wire. Current affects voltage as more current is needed to overcome cable losses, resulting in more voltage drop. The 14-gauge wire has an order of magnitude less resistance and is a massive improvement over the thin cabling.

Data is also affected by stray EM since a longer cable run acts as an antenna. Data is far more finicky than power. A longer run can introduce clock timing errors outside the spec of USB. None of that is a problem with power-only USB connections.

My 9m run is not very long for power.

12

So you’re running the USB power over 14AWG? I thought you were running 14 or 12 for the AC part then a standard USB cable to the cam, so my mistake. That will certainly help minimize voltage drop, though when you’re starting with 5v there still isn’t much to play with, and that large cable will actually be harder for the small USB adapter to energize properly. For some with longer runs (in the 20 to 25 foot range) people have found that simply using a bigger power supply (3+ amps) helps, since the cam draws more current when the voltage is lower, and that can help compensate. But over 25 that doesn’t seem to help unless a really good USB cable is used too, and it varies based on camera model.

Some caps would certainly help, though I think you’ll find sustained moving of the camera, especially with the spotlight on, might need quite a bit of capacitance. I’m also curious how a full charge up of the caps would be handled by the power supply. I suppose it is a short enough period that it shouldn’t matter, but it will stress it some.

Yes I agree that data is far more sensitive. Shielded cables can help with that (which most non-flat USB cables are), but it also heavily depends on the ability to do pretty complex encoding which needs nice sharp square waves and the proper minimum and maximum voltage. You could have a perfectly shielded USB cable with 100% coverage and 0 interference and the signal will still be a mess at the other end. Some of it from crosstalk, but the rest just from the compressing down of the voltage range causing the different amplitudes to be too close to be differentiated, and thus those bits are dropped. It isn’t really timing, the latency of a 9M run is virtually 0, just that the amplitude modulation simply can’t get the number of distinct levels it needs anymore.

Another thing people have done is put a USB power bank in the path near the cam. It has to be one that supports charging and supplying power at the same time. But that’s an expensive solution, unless you’re also looking for some battery backup ability. But it also essentially acts as a big capacitor.

I know 9M doesn’t seem like a long run for power but for 5V DC on normal USB cables (usually 28awg, or if you get a good one, it will have 24 for the power and 28 for signal), 9M is actually really far. I suspect one of the ones with 24 for power, combined with a good quality AC adapter, might be able to do it. But it is all trial and error at that point.

Well be sure to report back your results, this is something that others run into from time to time. Though I think most do not have the ability to rig up their own fix and in their case the PoE solution is probably easiest and most reliable. I guess I have two categories of things, stuff I like to tinker and toy with and optimize over time, and others (like my cams) where I just want them to work and not have to spend time on them. If you’re running 14AWG to the cam, beefing up your power supply may be all that is needed.

13

Hi WiseCammer,

Thank you again for the detailed testing and for walking through your setup.

We reviewed the backend binding records for this specific device. Based on our system logs, we do not see any unbinding or new binding activity around 2/20. The only recorded bindings were on 9/27 and 11/3.

Normally, a true factory reset would clear the device binding and generate a new binding event in our system. Since we did not observe that, this suggests the device may not have performed a full factory reset at the account level.

Given your description (spotlight + pan/tilt under load), one possibility is a transient voltage sag condition. Under peak current draw, if input voltage briefly dips below operational threshold, the device may enter an abnormal boot or WiFi initialization state. In certain cases, this can trigger the “Ready to Connect” audio prompt without actually erasing NVRAM binding data.

Our team is currently attempting to reproduce the scenario under similar load conditions to better understand whether a brownout condition could lead to this behavior.

A couple of additional questions that may help our investigation:

  • What was the approximate outdoor temperature during the event?

  • Has the cable experienced prolonged freezing conditions?

  • Have you noticed whether the behavior is more frequent during colder nights?

Lower ambient temperatures can increase effective resistance and reduce adapter efficiency, potentially worsening voltage sag during peak load.

We appreciate your technical insight and the level of detail you’ve provided. We’ll follow up once we have reproduction results.

Stay safe accessing the rooftop — and thank you for your patience.

14

Ive had the same issue where cameras reset themselves while outside. Also they take forever to connect via live view when they worked great before. I suspect it’s one of those updates causing various problems. It seems Wyze services and equipment are getting worse instead of better. They expect the customer to compile all types of information to submit to them and then they can’t even talk on the phone. Everything is via email.

15

Hi everyone — thank you all for the detailed reports and for helping us dig into this.

We were able to reproduce a scenario internally that aligns with what some of you described.

Under certain edge conditions (for example during brief power instability), the device can enter an abnormal state during a storage operation. This may result in behavior that looks like a reset (such as hearing “Ready to Connect”), even though the device is not actually fully unbound from the account.

We haven’t seen this tied to a specific firmware version. Based on what we’ve observed so far, the behavior appears to be more related to power conditions during certain operations.

Based on this, we’ve made improvements to how the device handles critical data during these moments, making it more resilient to brief interruptions (such as voltage dips under load).

These changes are intended to improve stability under conditions like higher load or transient power fluctuations.

The fix is planned for an upcoming firmware release, and we’ll share updates once it becomes available.

Really appreciate everyone who contributed details here — this directly helped us identify and address the issue :folded_hands:

16

That’s good to hear you’ve found the problem, and I look forward to the updated firmware. :+1:

Becoming Unbound…

Becoming unbound (even without a factory reset) is a big problem if I’m away from home on vacation or a trip. I do have a WiFi-enabled power switch that lets me remotely power-cycle a hung Wyze camera, but a camera that becomes unbound (“Waiting to connect”) will not reconnect until I come home to bind it to the WiFi network again, no matter how many times I power-cycle it remotely.

USB-PD needed for upcoming Wyze cameras…

In the long term, I think the Wyze cameras are outgrowing their standard 5V USB power. Because a standard 5V USB port can provide at most 2A current, so 10W of power. Upcoming Wyze cameras would be better off to support USB-PD (power delivery), which can provide much more power by negotiating higher voltages, like 20V at 5A or 100W, which is well more than enough for the latest Wyze cameras. Even 20W is probably good enough for the latest Wyze 4k cameras with spotlight and motor pan/tilt.

17

Standard USB can go as high as 30W for power only (or power and data with certain quick charge technologies). I definitely would not want a camera that needs that kind of power though. 15W adapters are pretty commonly available and don’t require any special protocols to operate at that amount.

USB-C and PD add cost, so that has to be factored in. PD also has distance limitations so it isn’t a cure for the problem.

10W is more than enough to run the Panv4 and all of its functions, the problem comes when cable that is beyond acceptable lengths cuts that (especially when a sudden change in current draw happens).

18

Standard USB at 30W means 6 amps of current—that is dangerous because of overheating risks for the hair-thin power conductors in USB wires. Those conductors are typically 28 awg and may even be thinner for cheaper cables. The Wyze USB adapter is 2A maximum and conforms to the common USB standard.

USB-PD is ideal for long-distance power delivery because its auto-negotiated higher voltages of 9V, 12V, and up to 20V are exactly what’s needed with hair-thin USB conductors at longer distances. Customers can also plug-and-play with existing Wyze cameras. They can use existing adapters and cables they’re familiar with.

Yes, it will cost more, but an extra dollar or two per unit would be worth it. With more power, Wyze can add more features: faster/more processors, more/stronger spotlights, and faster motor movement. It also removes the flakiness. Because some of my Wyze Cam Pan V4 units use a stock USB adapter and a stock USB cable, yet still sometimes crash and reboot during motor movement with the spotlight on.

19

Yes if you want to use a 20+W charger you should ensure your cable is up to it. Flat cables are less likely to be rated for that than most standard USB ones, and is why anything over about 15 to 20W will require one of the quick charge protocols to operate (which also checks the cabling).

But 15W is totally normal and very common (3A adapters, and 3.2 and 3.4 are not uncommon either).

Since their most power hungry cam needs less than 10W and at 15 foot distances a 2A adapter can supply that fine, there simply is not a need for PD at this point.

I’m familiar with USB and USB-C PD, I’m not saying it doesn’t do what you say, I’m just saying there isn’t any reason to implement it and drive the price of the units up when 99.9% of people would not get any advantage from it. For the other 0.1%, there are solutions like PoE or extending the AC power source to the location that can solve the issue for them.

If/when they need to go beyond 15W (since 20W is pushing it) then they’ll need to look into PD, but since things are typically getting more efficient rather than less, I’m doubting they’ll ever hit that for a standard camera, even a PTZ one.

20

Sure, energy might be decreasing for the compute & storage stack. But the motors and lights could use power.

For instance, the single LED light on the Wyze Cam Pan V4 could be expanded to have 4 lights or a full annular light around the lens. The motors could be increased for speed and torque. All of these increase the peak power required, which may already approach USB power limits, leaving too narrow a power safety margin.

I’m already seeing hiccups on the Wyze Cam Pan V4 on the stock USB adapter and stock power cable.