Discussion in 'SJ8 Series' started by nianeko2, Jul 10, 2018.
Excellent idea. Is it possible to make one cable POWER/MIC.?
Please watch the YT comment of this video.
Yes, both nianeko2 and I have made one, and when recording WITHOUT charging, it works fine, but when power is plugged in, an ugly hum appears in the audio. Check out the attached audio files to hear what the recording sounds like with and without power -- make sure to listen to them with volume up, so the hum can be heard. I have written about this to SJCAM Support, and they only replied with this: "USB adapter is not compatible with SJ8 Pro due to software limitation." This is total bull, of course, as audio IS being recorded under those conditions, so it's not like the feature is unsupported in firmware. Rather, what we have here is clearly just a fatal design flaw, whereby SJCAM's engineers failed to account for and eliminate interference from the power circuit. How sad, even heartbreaking, as this camera had a lot going for it, and I was so excited to get it. Now, I'll have to sell it. Unfortunately, being able to record with an external mic while charging is an essential requirement for me, since I tape long seminars, and so battery operation is out of the question. So, for now, bye-bye, SJCAM, and let's hope they'll do better next time...
Does the problem appear even if to use external battery as a power source?
Maybe it helps, if to use a quality AC/DC adapter to power the system from the wall socket, instead of a crappy charging adapter, that is not meant for powering sensitive audio devices?
The golden rule for noiseless audio recording is - don't let the noise into your system. By using a noisy power source you break the rule.
Thanks for replying! Actually, I have tried to record with several different adapters, including two original Samsung 2A chargers, which in my experience do not produce any audio interference (unlike the fake ones that are sold from China, which do!). I am not sure what else I could try. What kind of a quality AD/DC adapter did you have in mind? My result was the same as nianeko2's, by the way, so I would imagine that this issue is probably adapter-independent.
Did you hear the recording I attached to my previous message? It's not just a regular 60Hz hum that you would expect of a noisy transformer. There is modulation on the noise that suggests interference coming from some kind of an integrated circuit. My guess would be that it's the charging circuit inside the camera. What's your impression of it?
I mean something like this: https://www.aliexpress.com/item/HIF...ply-25VA-LPS-output-DC-5V-3A/32858772125.html
Or rather like this https://www.kempelektroniksshop.nl/keces-audio-p8-ultra-low-noise-linear-power-supply.html
I heared your recordings. Those crappy adaptors have very primitive and poor smoothing scheme, that can produce various types of noise. I checked your samples with my Wave Lab software and it seems, that there is a clear 30Hz rythm in the noise. As you use 60 Hz power grid, that noise is obviously related with adapting the grids sinus.
Just try to power the system from a battery bank. AFAIK, batteries are noiseless source. If the noise is still there, then blame SJCAM.
Oh, I see what you mean; that looks like a serious device... Can you imagine, though, lugging that puppy around to every presentation just to power a tiny camera?
Anyway, I just made another attempt to record, with power coming from a USB 3.0 port on my (very high-quality) laptop. The result was exactly the same noise. USB power is very highly regulated, so I don't see how that could be the source of the trouble. Everything points to the charging circuit of the camera as the probable source...
Are you joking?
Try battery power bank as a source!
I use this cheap power bank for my public recordings: https://www.dx.com/p/universal-1-x-18650-battery-box-power-bank-green-usb-5v-2032624#.XLL562c_OHs together with a 3500 mAh 18650 type Lithium battery. This simple solution can handle up to 4 hours of recording. And you can have as many spares as you need.
The noise pattern of your power source:
As you can see, it has regular rythm in the left channel. Selected 50 down peaks have total duration of 1.666 s, meaning, that it has exactly 30Hz frequency.
Dear Sulev. I tried dry cell battery, but the noise appeared.
Ok, then blame SJCAM!
BTW, if you use even battery pack (even this on, what I use and linked above), then there might be still crappy voltage regulator in it, that generates noise. Nowadays voltage regulation is often trusted to cheap voltage regulation chips, I can't quarantee, that they are noiseless. The best solution is to use 5V battery without any electronics components (especially those, containing transistors, capacitors etc) between the battery and the USB jack.
Thanks for the power bank recommendation. If 3200 mAh will last for 4 hours, it shouldn't be hard to cover a whole weekend's worth of seminars with a larger unit. Unfortunately, I don't have a power bank on hand, but I'll see if I can get my hands on one and try again.
nianeko2 said, though, "I tried dry cell battery, but the noise appeared." That's not a good sign...
You are right about the frequency of the louder beats in the noise --it's 30 Hz. In any case, from my standpoint, even if the source of the noise were external, no recording device has any business being THAT sensitive to it. After all, I was able to record with exactly the same chargers on my cell phone (Samsung Note 4) and there was no interference at all. You know you've got trouble when you have to dance on tiptoes around your recording device just to get it to work properly with standard, original, good-quality hardware. If I were an SJCAM product manager and got wind of this kind of problem that users are having, I'd take it very, very seriously...
Did you feed your Samsung Note 4 with external mic together with external power?
Maybe you can borrow a battery bank for testing? I don't want to be blamed because of useless expenses on gadgets, that actually does not solve the problem, in case the noise is not caused by the power adapter. Most people I know have a battery bank or even many, because they charge their smartphones with them.
Yes, I did, and I used exactly the same equipment -- charger, cable, mic. On the Note 4, the recording comes out crystal clear and free of noise, every time. I did notice that when I tried using fake Chinese copies of the Samsung charger, I did get a power hum, even on the Note 4, but with the original chargers, there was no problem at all. On SJCAM, however, no matter what power source I've used (except a power bank or a bare battery, which I haven't tried yet), the hum has always been there, and it was always exactly the same. If it were externally sourced, you'd expect to hear at least SOME difference between the different chargers and USB ports, wouldn't you?
Is the hum always on the left channel mostly?
I expect that if the noise is caused by external charger, then the noise varies, if you switch between 50 Hz and 60 Hz grids.
If you suspect, that the 30 Hz noise is related with the sensor (frame rate), then you may test, how switching to PAL modes (50/25 fps modes, let's forget about possible flickering caused by lamps at 60 Hz, for this debugging test) or higer frame rates affects the noise.
I have ripped apart several USB chargers from different brands and they all had generally very similar electronics scheme.
SJCAM improved power system at SJ8.
When using SJ7 with mobile battery, becomes empty & SJ7 shut down even if inner battery is full.
At the case of SJ8, inner battery works parallel, so after both battery become empty then shut down.
They are selling only external mic, and do not consider mic & power adapter.
According to my understanding, this automatic shut down, when external source gets discharged, was entirely FW problem. They thought, that the dashcam solution works for everybody - stop the car engine, and the camera switches off. I personally hate this behaviour, because I can't swap batteries, when recording several days or even weeks long videolapse videos.
The fact that the noise in my test recording only appeared on the left channel had to do with the fact that I had used a mono mic. When I use a stereo mic, the noise is exactly the same on both channels.
It's an interesting suggestion to try recording at 50 or 25 fps to see if the noise frequency shifts accordingly. I'll give it another try later (if I haven't sold the camera by then -- it's already advertised...).
In any case, I think you are correct, in assuming that the power input is the initial source of the noise. The point, though, is that it is being picked up and amplified excessively by the external mic circuit, and that should not be happening. I suppose someone might expect the user to jump through hoops to try and reduce the external noise by getting a super-duper-smooth power supply, but since SJCAM doesn't recommend such an investment for normal operation, I don't see it as a reasonable expectation. Besides, the internal microphone circuit is NOT picking up the power hum -- it only happens with the external mics.
Separate names with a comma.