Zenfox T3 Triple channel dash cams free test invitation, limited quantity

[HonestReview]

Went out for lunch and some shopping today and observed some seemingly heat related behavior. The dashcam had previously been on in the garage for 8 hours overnight until the parking timer expired and recorded all 3 cameras without any missing footage during that time.

Timeline:
11:07 trip 1 start
11:34 trip 1 end
11:39 parking mode start
11:59 rear camera stopped
12:05 parking mode end (unknown cause)
12:09 interior camera stopped
12:10 parking mode start
12:18 parking mode end
------> missing recording
12:20 trip 2 start (already moving)
12:23 trip 2 end
12:37 parking mode start (don't know why it took so long)
12:38 parking mode end (unknown cause)
12:45 parking mode start
12:46 parking mode end / trip 3 start
13:07 interior camera resumed recording
13:17 rear camera resumed recording
13:20 trip 3 end
13:25 parking mode start

The trip started mostly sunny and ended up mostly cloudy with very light rain and outside temperatures ranged from 85F to 90F according to the car sensor. A Samsung Evo Plus 128GB was used and the camera was powered by a portable DC power pack the entire time so it was not affected by the car's electrical system or triggered by ACC power. Unfortunately, I had a passenger so was not monitoring the camera and didn't have the IR thermometer. All 3 cameras recorded fine during the 30 minute drive to the destination and entered parking mode 5 minutes later. About 20 minutes after that, the rear camera stopped recording. The remaining 2 cameras continued to record for 6 more minutes, then exited parking mode for no reason that I could determine from the footage. Normal recording continued 4 more minutes, then the interior camera stopped and parking mode resumed shortly after with only the front camera. Parking mode continued for another 9 minutes at which time we returned to the car. However 2 minutes of footage were missing between the last parking mode file and normal mode resuming. Don't know if the camera had just shutdown or if it was a heat related issue transitioning back to normal mode. We then proceed to another location and the dashcam continued to record only the front camera and there was another transition from parking to normal mode with no obvious cause while we were shopping. On the return trip with the AC running, the interior camera started recording again after 20 minutes followed by the rear camera 10 minutes later.

I'm planning another overnight test in the garage with the same card after the DC power pack is recharged to see if any of the anomalies recur in the the lower temperatures.

KuoH

This is the exact behavior I described. First the rear camera disables, followed by the interior camera, leaving only the front recording. Supposedly, this is by design to keep the unit from overheating??

From my experience, a unit reboot is necessary to bring back the rear and interior cameras, although Kuoh's test seem to confirm they'll return own their own after temperatures fall and heat dissipate from the camera.

I agree with @SawMaster's concerns. Product isn't ready for post production or prime time until a solution can be devised to prevent the unit from overheating.

 

[HonestReview]

Other cams have failed (including plastics) at about that temp, so I wouldn't be surprised if it happens here given that this is a new manufacturer who may not grasp all the aspects of having your cam built by someone else. Time will tell and we will see.

I do think that dropping the 'extra' cams as the way of mitigating excess heat is a bad approach. Something like the Kenwood DRV-A601W does with lowering bitrates instead would be my preferred solution. Above all else, cams should never stop recording by themselves unless no other cam could record then either. The keyword is "Reliability" and any cam lacking that isn't acceptable.

Phil

As much as I'm taking sh*t for being critical, I am onboard with your statement 100%. I realize flaws are ever present in a new product, or there wouldn't be a need for beta testing. Unfortunately, when a major failure is experienced, that issue generally takes precedence over other concerns. At present, we need to await ZenFox's guidance on how to address this matter, because the current operating conditions of the camera are unacceptable.

 

[HonestReview]

Muthbusters and Overhaulin was some of my favorites at one time.
I like for Chip Foose to build me a Amazon, spiced up on the power and handling + stopping side of course.

Volvo Amazon? And yes Mythbusters made science fun. Who doesn't like explosions and products being incinerated in the name of science!

 

[kamkar]

Yeah my first car, shoehorn in a small block would be nice, maybe a minimal chop, massive disk brakes all around, fat tires, probably need a new rear end, the original Volvo Axle i dont think can handle 500 HP ( or there about )
Maybe a axle are not even good for handling, nice 5-6 speed stick shift

I like this one, but dont want boosted motor.

But damn nice sleeper wagon 😎

THis guy got the power / handling going alright, but the looks not so much IMO.

 

[HonestReview]

Yeah my first car, shoehorn in a small block would be nice, maybe a minimal chop, massive disk brakes all around, fat tires, probably need a new rear end, the original Volvo Axle i dont think can handle 500 HP ( or there about )
Maybe a axle are not even good for handling, nice 5-6 speed stick shift

I like this one, but dont want boosted motor.


But damn nice sleeper wagon 😎

THis guy got the power / handling going alright, but the looks not so much IMO.

Amazon's are cool looking cars. Man that first video had my jaw drop. That Amazon had some pep! Now that second video made me cry. What the hell did that guy do to the Amazon! He created a Frankencar!

 

[SawMaster]

Some thoughts:
Perhaps testers could place an outdoor or garden thermometer in view of the cabin cam to at least get an accurate interior ambient temp when parked in the heat. Plenty of charts online to correlate inside and outside temps in parked cars. No ambiguity such as you get with words like "very hot" which mean different things to different people.

Also remembering how the first Garmin cams shut down so early from their over-temp sensor and software programming. Their fix was software only in raising the temp limit; apparently they were being too conservative with determining how much heat their hardware could handle. Seems to have worked as I've seen no reports of any toasted Garmin cams after that update. Might also work here?

Does seem to be a bug in when the cams switch off due to heat. It seems that the processor is either being slowed down excessively by the asked-for actions, or there's a shut-down bug happening here too.

I propose trying a firmware with higher temp limits, and also seeing if there is indeed something buggy in the shutdown sequence/operation and trying a different approach there if necessary. Might want to do two firmwares individually to ensure each change does not affect anything else before trying both changes together. This must be dealt with before market release or this cam will be generating tons of warranty claims and returns.

Phil

 

[Jeremy Wong]

Never knew they literally threw these things into actual ovens, but it makes sense. Only way you're going to know their threshold is to cook them (literally). Bon Apetit!

Car Cam Central actually put them into a modified oven for heat testings

Car Cam Central

Welcome to CarCamCentral's Youtube channel. We make video and written reviews on dash cameras to help you decide on what to purchase. We go in-depth on all t...

www.youtube.com

 

[kuoh]

Yup, I did just that with the previous overnight tests. However even in a 90F+ garage with the car interior at 100F, the dashcam still seems to behave normally. I'll have to figure out a way to put in view of the interior camera for the next outdoor test. I just pulled the card out after recording in parking mode for about 5 hours in the garage and there didn't appear to be any missing cameras or lost footage. The dash was 100F, the top of the windshield where the camera is mounted was 110F, the bottom of the camera at the Zenfox logo was around 105F while the top of the camera at the T3 logo was almost 145F. I think displaying the temperature sensor readout in the recordings on the next firmware would be very helpful.

If it's not possible to have the camera continue recording all channels in high heat, then I would suggest an additional option in the parking mode recording. Instead of only continuous recording to the SD card, perhaps a new parking mode can be added to only record when the G sensor is triggered or significant motion is detected? This may help keep the heat down by lowering CPU and memory card usage. There should also be a different G sensitivity between the parking and normal modes as you would want it to be more sensitive when parked.

I also ran some power tests using the included USB cable and a USB power meter. The camera draws between 0.95A to 1.2A in both normal and parking modes and very little change with the screen on or off. There were a couple of very short dips to as low as 0.75A, but the average seems to be around 1A. I would think this is well within the capabilities of most parking kits, especially those that just switch 12V to a cigarette lighter socket.

EDIT:
After reviewing the garage recordings, it seems that some significant motion or lighting changes can trigger a transition from parking to normal mode. I opened the garage door about a hour after parking mode had started, but didn't touch the car and the recordings changed from parking to normal mode. It resumed parking mode 5 minutes later and continued to record all cameras normally for the next 4 hours until the memory card was removed.

KuoH

Perhaps testers could place an outdoor or garden thermometer in view of the cabin cam to at least get an accurate interior ambient temp when parked in the heat. Plenty of charts online to correlate inside and outside temps in parked cars. No ambiguity such as you get with words like "very hot" which mean different things to different people.

 

[kuoh]

My toaster oven temperature dial starts at 150F, but can actually be set lower. It it somewhat tempting to satisfy my scientific curiosity, but I'm not all that keen on taking the whole setup out now that I've gone through the trouble of installing and running the cables.

KuoH

Car Cam Central actually put them into a modified oven for heat testings

 

[HonestReview]

Yup, I did just that with the previous overnight tests. However even in a 90F+ garage with the car interior at 100F, the dashcam still seems to behave normally. I'll have to figure out a way to put in view of the interior camera for the next outdoor test. I just pulled the card out after recording in parking mode for about 5 hours in the garage and there didn't appear to be any missing cameras or lost footage. The dash was 100F, the top of the windshield where the camera is mounted was 110F, the bottom of the camera at the Zenfox logo was around 105F while the top of the camera at the T3 logo was almost 145F. I think displaying the temperature sensor readout in the recordings on the next firmware would be very helpful.

If it's not possible to have the camera continue recording all channels in high heat, then I would suggest an additional option in the parking mode recording. Instead of only continuous recording to the SD card, perhaps a new parking mode can be added to only record when the G sensor is triggered or significant motion is detected? This may help keep the heat down by lowering CPU and memory card usage. There should also be a different G sensitivity between the parking and normal modes as you would want it to be more sensitive when parked.

I also ran some power tests using the included USB cable and a USB power meter. The camera draws between 0.95A to 1.2A in both normal and parking modes and very little change with the screen on or off. There were a couple of very short dips to as low as 0.75A, but the average seems to be around 1A. I would think this is well within the capabilities of most parking kits, especially those that just switch 12V to a cigarette lighter socket.

EDIT:
After reviewing the garage recordings, it seems that some significant motion or lighting changes can trigger a transition from parking to normal mode. I opened the garage door about a hour after parking mode had started, but didn't touch the car and the recordings changed from parking to normal mode. It resumed parking mode 5 minutes later and continued to record all cameras normally for the next 4 hours until the memory card was removed.

KuoH

A suggestion: You need o break your tests into two different parameters. Considering a 90F in a shaded garage isn't comparable to 90F in direct sunlight. So your tests should include:

1. Measuring camera temperatures + vehicle temperatures in shaded garamge - Completed
2. Now measuring the car and camerasitting in direct sunlight on a similar day with the sun beating on it.

From my tests, it appears the camera experiences the failure of the internal and rear camera only in direct sunlight. Which signifies the camera is far exceeding the 145F experienced in a shaded garage with ambient temperatures of 90F.

I intend to invest in an analog thermometer, but the cheap ones only went up to 50C / 120F. So pretty useless. I'll have to see what I can come up with as I don't feel like shelling out a ton of money here. I've already spent around 90€ / 100 USD on a 256GB + 128GB Sandisk High Endurance MicroSD (was planning on installing a 2nd test camera in family member's vehicle + HK3 Hardwire kit. But I'll try to get one so I can verify the temperatures in my vehicle when the camera fails.

It'll probably have to wait until later this week as it's going to be cloudy and temperatures cooled some.

 

[HonestReview]

Some thoughts:
Perhaps testers could place an outdoor or garden thermometer in view of the cabin cam to at least get an accurate interior ambient temp when parked in the heat. Plenty of charts online to correlate inside and outside temps in parked cars. No ambiguity such as you get with words like "very hot" which mean different things to different people.

Going to try and find a cheap thermometer and check the vehicles enclosed temperature to see under what parameters the camera is failing. Unfortunately, it'll have to wait until later in the week. Cooled down and cloudy. Not conducive to the specs needed to cause the camera to overheat [/quote]

Also remembering how the first Garmin cams shut down so early from their over-temp sensor and software programming. Their fix was software only in raising the temp limit; apparently they were being too conservative with determining how much heat their hardware could handle. Seems to have worked as I've seen no reports of any toasted Garmin cams after that update. Might also work here?

There are three questions we need ZenFox to Answer:

1. Was this camera tested for heat and if so, at what temperature?

2. Is the camera shutting off by design to prevent the internal circuitry from being damaged? If so, what is the threshold the camera must reach before Internal and Rear camera are disabled?

3. Can you up the maximum cutoff temperature and stress test the camera beyond these levels

Does seem to be a bug in when the cams switch off due to heat. It seems that the processor is either being slowed down excessively by the asked-for actions, or there's a shut-down bug happening here too. [/quote

It depends if this shutoff is internal to preserve the camera's electronics or if what we're seeing is a product design flaw with inadqauate heat dispersement. I agree, upping the maximum temperature and possibly lowering the low bitrate record mode bitrate even lower, may resolve the issue. Unfortunately, we're at the mercy of Zenfox to give feedback here to initiate follow up testing. [/qupte]

I propose trying a firmware with higher temp limits, and also seeing if there is indeed something buggy in the shutdown sequence/operation and trying a different approach there if necessary. Might want to do two firmwares individually to ensure each change does not affect anything else before trying both changes together. This must be dealt with before market release or this cam will be generating tons of warranty claims and returns.

Phil

I know I may be coming off as an A55hole, but sugar coating problems isn't going to help this company resolve them. I'd rather be told directly my product has a problem, than release said product to market, and find out I've now alienated my customer base with a dysfunctional device. First impressions are lasting impressions.

So I fully agree with your assessment.

 

[HonestReview]

Came Across another "Quasi Bug":

So a few days ago, the camera set itself to 21 June 2020. Moving itself two days ahead for no reason. Others here had same issue. Well I didn't think to format the SD card. So come today, the camera was parked and recording, and I came out to find it beeping. The camera had stopped recording anything what so ever. What I discovered was a naming conflict. As soon as the camera reached the files recorded on the 19th of June erroneously listed as 21 June, it halted due to the naming conflict. Which you would expect, as two files can't have the same name.

 

[kamkar]

I have had several start / stop situations today, and the failure to start and record have not been there since i added a 5 second boot delay.
All other files are also just fine.

 

[kuoh]

I've had that 2 day ahead jump happen multiple times now as well, usually after I power down to swap cards. In fact it recorded about 10 hours again in the garage last night with the wrong date, but didn't lose any footage or cameras. I would prefer it to overwrite the previous file rather than stop recording when encountering a filename conflict, but that's a personal preference. Alternatively it could append additional characters and let the user sort it out later.

Went out for another drive today in 93F temps with partly sunny conditions and the camera shutdowns and recording gaps were consistent with yesterday's trip. Also noticed transitions from parking to normal mode generally loses around 5 seconds of footage. This is less than ideal if someone happens to encounter a parking lot incident right when backing out. With the current behavior, I'd probably set the parking timer to just a few minutes to an hour tops in all but the coldest climates. I would guess the dashcam is probably behaving as designed, but we just have slightly different expectations than the engineers.

Timeline:
11:58 trip 1 end
12:03 parking start
12:15 parking end / rear camera stop
12:22 parking start
12:25 interior camera stop
12:34 parking end / returned to car
-----> missing footage
12:36 trip 2 start / font only already moving
12:37 trip 2 end
12:42 parking start / end (unknown cause)
12:48 parking start
12:51 parking end / returned to car
12:52 trip 3 start
13:08 trip 3 end / still front only due to short drive

KuoH

As soon as the camera reached the files recorded on the 19th of June erroneously listed as 21 June, it halted due to the naming conflict. Which you would expect, as two files can't have the same name.

 

[HonestReview]

I've had that 2 day ahead jump happen multiple times now as well, usually after I power down to swap cards. In fact it recorded about 10 hours again in the garage last night with the wrong date, but didn't lose any footage or cameras. I would prefer it to overwrite the previous file rather than stop recording when encountering a filename conflict, but that's a personal preference. Alternatively it could append additional characters and let the user sort it out later.

Agreed, though conceivably, there should never be a naming conflict since two days are never suppose to be repeated. There will never be another 21 June 2020. So the bigger issue is the camera randomly changing to the incorrect time, thereby causing the conflict. However, I would prefer it overwrite the existing file too, in the event of said conflict.

Went out for another drive today in 93F temps with partly sunny conditions and the camera shutdowns and recording gaps were consistent with yesterday's trip. Also noticed transitions from parking to normal mode generally loses around 5 seconds of footage. This is less than ideal if someone happens to encounter a parking lot incident right when backing out. With the current behavior, I'd probably set the parking timer to just a few minutes to an hour tops in all but the coldest climates. I would guess the dashcam is probably behaving as designed, but we just have slightly different expectations than the engineers.

Same symptoms I've experienced on numerous occasions. Quickest way to make the Rear and Interior camera return is to reboot the unit. My only choice to measure interior temperature may be an oven thermometer. Not sure a digital one can withstand the heat of an enclosed vehicle.

Even if the front and interior cameras are disabling by design to prevent damage to circuitry, I see this as a major obstacle. The camera lacks a reliable parking mode, losing its edge over existing products on the market. The selling point is a 3 channel camera to capture around one's vehicle. While the T3 functions without issue during normal driving, the same isn't true while parked. Setting a timer is merely a guestimate, since a car's interior temperature rapidly increases on a hot and sunny day. Especially since we do not know the camera's designed threshold before shutting down the interior and rear camera.

Timeline:
11:58 trip 1 end
12:03 parking start
12:15 parking end / rear camera stop
12:22 parking start
12:25 interior camera stop
12:34 parking end / returned to car
-----> missing footage
12:36 trip 2 start / font only already moving
12:37 trip 2 end
12:42 parking start / end (unknown cause)
12:48 parking start
12:51 parking end / returned to car
12:52 trip 3 start
13:08 trip 3 end / still front only due to short drive

KuoH

 

[SawMaster]

I would prefer it to overwrite the previous file rather than stop recording when encountering a filename conflict,

I don't think this is possible, at least with a NT chipset. The behavior here is the same with several NT cams: B1W had a time-loss problem early on before they added a RTC battery and also the G1W-S. With those time would rreset back to the original firmware release date, not move forward. Viofo and CamDii couldn't fix this. I've seen it happen when I manually reset the time in NT cams but made a mistake and I also didn't format afterward 🙁

The GPS should be preventing this time-change problem but it obviously isn't. @kamkar1 did you see a RTC cell in there? Can't see one in your pics. If Zenfox is relying on the caps alone to hold time when shut down like the early B1W and G1W-S then that could be part of the problem.

Phil

 

[kuoh]

I don't think an RTC battery should be necessary to fix this as the GPS knows the correct time, otherwise it wouldn't be able to calculate the speed properly. It is likely a bug in how it handles the GPS data, as the error is always 2 days ahead with the correct GMT time instead of local time.

KuoH

The GPS should be preventing this time-change problem but it obviously isn't. @kamkar1 did you see a RTC cell in there? Can't see one in your pics. If Zenfox is relying on the caps alone to hold time when shut down like the early B1W and G1W-S then that could be part of the problem.

Phil

 

[kamkar]

I dident take it all apart, so a RCT battery might be on the other side of the PCB.

This time hiccup think i think i got that one time in the early days, but it is not something i have seen lately.
It might have occoured on day # 2 or was it 3 when i briefly checked if it would go into parking mode if the camera just sat there on in my idling car.

 

[EGS]

I don't think this is possible, at least with a NT chipset. The behavior here is the same with several NT cams: B1W had a time-loss problem early on before they added a RTC battery and also the G1W-S. With those time would rreset back to the original firmware release date, not move forward. Viofo and CamDii couldn't fix this. I've seen it happen when I manually reset the time in NT cams but made a mistake and I also didn't format afterward 🙁

The GPS should be preventing this time-change problem but it obviously isn't. @kamkar1 did you see a RTC cell in there? Can't see one in your pics. If Zenfox is relying on the caps alone to hold time when shut down like the early B1W and G1W-S then that could be part of the problem.

Phil

There is an RTC battery used in this camera, slightly larger than the one seen in the other cameras I have oppened in the past. By the way, I like the fact that it is not soldered to the circuit board and it will be very easy to replace whenever it fails as it is taped to the shell and wires plugging to the main board.

 

[EGS]

I just reviewed some footage from a short drive to walmart yesterday and I have fount that it has only recorded the first 2 minutes on all 3 cameras an nothing else after that, meaning all 3 cameras did stop recording and no warning whatsoever. I think an audible warning is critical when the system fails to record entirely. High temperature was not a factor in this case as temperature was around 80F and vehicle was parked in a cool garage. Things are quite unpredictable at this point.