The new 808 #18 DR32 Car Camera

[superpat]

There is a deficiency in the sound tests carried out above, using a mobile phone and the DR32.

The mobile phone was NOT connected to the car 12v supply during the tests. It was running off it's battery.

I would like to know the frequency of the noise shown in the plots. To me it looks quite high frequency, well above audio range.

I would hazard a guess that the problem car installations have very electrical noisy alternator output, with a missing or broken alternator noise suppression capacitor AND the DR32 has piss poor common mode noise rejection on the input power lines. There has to some filtering in the 12v to 5v car plug adapter, perhaps that is ng.

I am still waiting for my DR32, so cannot test myself, but a member has stated quite clearly, he gets NO noise when connected to a mains 5v supply and rubbish sound in the car. The DR32 HAS to be picking up car interference,

What is interesting is that there is no apparent loss of video quality, one would think that both video and audio would be affected.

Try a low pass filter on the 5v input to the camera.

regards


P

 

[Jazzist]

The noise is not getting in through the dc input.

One member has run it through a ups, I have run the camera from a 5v battery.

The cameras are faulty.

 

[gobbertinker]

I tried a Rocketfish brand car adapter, it seems to be a lot better but notice when hitting rough road it gets distorted. This adapter does seem better in lowering the noise when accelerating to higher speeds and traveling at those speeds. When I first installed the camera when I received it the mail, I noticed that the audio became more distorted as the car gained speed and also when riding on rough roads it would become worse. Does anyone know what the voltage is in cars in China. I forgot to add to this post that I did not put the ferrite collar on this adapter, maybe that would help also, when I put the ferrite collar on the included car adapter that came with the camera it did help with the distorted audio.

 

[jokiin]

Does anyone know what the voltage is in cars in China.

the same as the rest of the world, no different, it's very much a standard, 12v

 

[gobbertinker]

Guys this is what's happening and yes the camera is getting 5v (not the problem).

I did some more testing.

1. One of the questions was addressing that the microphone may be moving inside the chassis. With the windshield vibrating, it could be adding to the sound in a bad way. So, while it was running and there was road noise, I removed the unit and held it by the mount for a short time. In short, there was no difference.

2. I was curious to see what the audio profile was. So I simultaneously recorded a couple of clips with both my mobile phone camera and the DVR. The phone was placed right next to the DVR to get any additional audio reflections from the hard surface of the windshield. I took these video samples and stripped the audio out (and converted to .wav 48kHz stereo). I then viewed the waveform side by side in Audactiy. I did try to keep them the same scale, but there is some difference.

My results:
The first is with the car running, stopped, and very little ambient noise. You'll see 4 big sound pulses. This is my voice saying "Mark Mark Mark Mark" to try and line up the waveforms for analysis. On the right is my phone, on the left is the DVR.

Some observations:
1) The noise floor of the DVR is significantly higher.
2) The peaks are clipped earlier.
3) A lot of the extra detail is lost, although perfect audio fidelity is not required for this type of device. In other words, the lower dynamic range is fine.
4) So overall, the shape is there, but it quickly gets distorted.
5) Just listening to the audio though, I find nothing wrong with either. Both produce usable sound.


This second clip was performed the same way, but instead at 60mph (110kph) on the highway. My car isn't exactly the quietest, but it's far from loud. Windows were shut, no radio was on.

1) The mobile phone on the right, you can see that the ambient noise has significantly increased, which reflects real world sound.
2) You can clearly see the four pulses again on the right, although they are starting to get distorted. Again, the phone audio is usable.
3) The audio from the DVR, to be frank, is totally misleading. I couldn't find my 4 pulses at all. The entire waveform is saturated by ambient noise and there's no detail in the sound left. Additionally, there's significant clipping everywhere.


As suspected, it appears there is just too much audio gain in the unit. It needs to be lowered significantly.

I respect your testing and posting your results here, I'm by no means disrespecting your opinion. But! Is your phone being powered by the vehicle? and is it attached to the windshield by a mount of some sort. Vehicle charging systems can and will produce noise through the wiring system of vehicle, and also the voltage can and will become higher than 12volts when accelerating. I'm a ham radio operator and do get noise through my two-way radio in my truck when the engine is running, this noise does become higher in sound when accelerating. Your phone test really isn't a true test as in regards of the comparison between the two. If you had another camera that was using the vehicle power and not a battery for testing it would be more comparable, than again if you had another camera I guess you would not here trying to get the DR32 working. You posted that you also tried the camera with a 5 Volt battery, I guess you have to verify that the battery is in fact putting out a constant 5 volts and not any less, when powering something off a battery the voltage can drop considerably.

 

[gobbertinker]

This is a picture of the ferrite collar that I took off of another usb lead and put on the car adapter that came with the camera. Cables can act like antenna's and pickup other noise that is produced by the vehicles charging system. I'll put this ferrite collar on the Rocketfish adapter and see if helps in eliminating audio noise any further.

 

[gobbertinker]

People have already tried numerous adaptors - doesn't work!
It does work better indoors as you haven't got the noise of the car.
The fault is 100% the camera, way to much gain.

I don't know why there would be a sound difference in noise that is produced in the house verses' the noise produced in a car. Doesn't really make any sense, we can still produce the low frequency noise in the house that you say is in the vehicle. So to say it's the different frequency's that effect the camera would be wrong idea. On the other hand I guess this low frequency noise could be traveling through the mounting system of camera and be picked up through the case of the camera.


As anyone tried simply holding the camera instead of mounting it to the windshield? Maybe the vibrations are actually causing the over driving of the audio.

 

[Mikk36]

The car adapter I have is simply to small to have any voltage reducing in it.

Nonsense.
https://www.sparkfun.com/products/107

People have already tried numerous adaptors - doesn't work!
It does work better indoors as you haven't got the noise of the car.
The fault is 100% the camera, way to much gain.

I don't know why there would be a sound difference in noise that is produced in the house verses' the noise produced in a car. Doesn't really make any sense, we can still produce the low frequency noise in the house that you say is in the vehicle. So to say it's the different frequency's that effect the camera would be wrong idea. On the other hand I guess this low frequency noise could be traveling through the mounting system of camera and be picked up through the case of the camera.


As anyone tried simply holding the camera instead of mounting it to the windshield? Maybe the vibrations are actually causing the over driving of the audio.

It's not about difference of noise, it's about the amount of noise.
Have you really measured the noise level in a travelling car?

And yes, people have tried holding the camera in hand instead of having it mounted. It made no difference at all.

 

[gobbertinker]

The car adapter I have is simply to small to have any voltage reducing in it.

Nonsense.
https://www.sparkfun.com/products/107

Read my recent posts! I have verified that there are components in the adapter. Whether or not the adapter is working correctly is the question. Have you tested the adapter to see that it is putting out a constant 5 Volts. Also what is the maximum imput voltage that the adapter can handle?

 

[Mikk36]

The car adapter I have is simply to small to have any voltage reducing in it.

Nonsense.
https://www.sparkfun.com/products/107

Read my recent posts! I have verified that there are components in the adapter. Whether or not the adapter is working correctly is the question. Have you tested the adapter to see that it is putting out a constant 5 Volts. Also what is the maximum imput voltage that the adapter can handle?

I powered the camera from a different adapter, that has two outputs. I connected one output to the camera and the second to a multimeter. The voltage was a constant 5.18V, well within the spec limit.
Max. voltage
5.00±0.25 V (pre-3.0);
5.00+0.25-0.55 V (USB 3.0)

The max limit the adapter can handle should normally be at 14.4 V, which is the allowed limit for a car voltage.
This is not a problem since I have also checked that my car generally runs below the 14 V mark. Furthermore, I have tried with 3 different adapters, two of them having a limit of 1 A, one of them 2 A.
Noise distortion was not affected by using different adapters.

 

[gobbertinker]

The car adapter I have is simply to small to have any voltage reducing in it.

Nonsense.
https://www.sparkfun.com/products/107

I don't understand why you have a link to a voltage regulator, that's not what is in the car adapter.

 

[gobbertinker]

Read my recent posts! I have verified that there are components in the adapter. Whether or not the adapter is working correctly is the question. Have you tested the adapter to see that it is putting out a constant 5 Volts. Also what is the maximum imput voltage that the adapter can handle?[/quote]I powered the camera from a different adapter, that has two outputs. I connected one output to the camera and the second to a multimeter. The voltage was a constant 5.18V, well within the spec limit.
Max. voltage
5.00±0.25 V (pre-3.0);
5.00+0.25-0.55 V (USB 3.0)

The max limit the adapter can handle should normally be at 14.4 V, which is the allowed limit for a car voltage.
This is not a problem since I have also checked that my car generally runs below the 14 V mark. Furthermore, I have tried with 3 different adapters, two of them having a limit of 1 A, one of them 2 A.
Noise distortion was not affected by using different adapters.[/quote]


I'm simply trying to help with this problem and also trying to get mine fixed in the process. As I stated I tried a Rocketfish brand and it did help to an extent.

 

[gobbertinker]

With all the testing of the camera has anyone looked to see if there is software of some sort that would allow us to decrease the mic gain?

 

[Mikk36]

The car adapter I have is simply to small to have any voltage reducing in it.

Nonsense.
https://www.sparkfun.com/products/107

I don't understand why you have a link to a voltage regulator, that's not what is in the car adapter.

It was to show that the size of the adapter is not a limiting factor.
Different adapters can have different solutions.

Read my recent posts! I have verified that there are components in the adapter. Whether or not the adapter is working correctly is the question. Have you tested the adapter to see that it is putting out a constant 5 Volts. Also what is the maximum imput voltage that the adapter can handle?

I powered the camera from a different adapter, that has two outputs. I connected one output to the camera and the second to a multimeter. The voltage was a constant 5.18V, well within the spec limit.
Max. voltage
5.00±0.25 V (pre-3.0);
5.00+0.25-0.55 V (USB 3.0)

The max limit the adapter can handle should normally be at 14.4 V, which is the allowed limit for a car voltage.
This is not a problem since I have also checked that my car generally runs below the 14 V mark. Furthermore, I have tried with 3 different adapters, two of them having a limit of 1 A, one of them 2 A.
Noise distortion was not affected by using different adapters.

I'm simply trying to help with this problem and also trying to get mine fixed in the process. As I stated I tried a Rocketfish brand and it did help to an extent.

I understand, but when You'd read the topic from the beginning, we have already tried lots of things out.
I have also spoken to two people from the Hetai company and according to them, they're working on fixing the issue.
They haven't given any estimates though.

With all the testing of the camera has anyone looked to see if there is software of some sort that would allow us to decrease the mic gain?

Haven't checked but they haven't notified me to try one either. Last time I spoke to them was yesterday.

 

[NoAIm]

I've been playing with the camera some more. And I still really like it.

But I found something that I should have realized, but still caught me by surprise.

The camera is advertised as a 1080p 30fps recording. But it also has 720p @ 30fps and 60fps. So today I decided to try out the 60fps. My theory was that with moving objects, maybe higher temporal resolution would be better than visual. The thought intrigued me enough that I figured I should try it and compare.

What I didn't expect, but should have, is that in order to run at the 720p @ 60fps, that the recording is just a crop of the 1080p sensor. I've seen other cameras do this, namely some point and shoots where you can record at 120fps - 240fps for some slow motion.

Unfortunately, this has the effect of also lowering the field of view. If the FOV of the 1080p is 127 degrees, then the cropped 720p becomes more like a 85 degree FOV.

A comparison between the two. For a guide of what you can see, take a look at the reflected air vents in the glass.

1080p

720p

Again, not a problem, but just something interesting.

 

[andrew73]

720p is a bit irrelevant when 1080p mode is so good. I'd rather 1080p has a greater field of view than the other way round, like the GS1000! I hope the firmware update gives us a vertical flip option for mounting the camera upside down; I'm more interested in the visual functionality than sound, tbh.

I'm delighted with the DR32, the image quality is superb. Only slight negative is colour trails, most obvious with bright reds:

 

[gobbertinker]

I have updoaded a video from my camera if anyone is interest in this lousy audio. http://www.youtube.com/watch?v=Tq2XMb1D ... e=youtu.be

 

[Mikk36]

Damn your engine must be noisy, not to mention the very loud radio.

 

[Jazzist]

Haha

 

[sludgeguts]

If this camera is designed to run on 5v & a PC's USB port outputs 5v, how comes the camera doesn't work when I plug it into the PC?
It switches on then the screen goes black & it comes up on the PC as a removable drive. :?: