Review of FPV-Japan DVR

[Coyote]

Hi Guys, ok, ill upload two clips, one from battery power, the other from 5v supply.

Yes supplied battery used on battery example Ren

The AC adaptors no good for UK mains sockets, so i chopped that and used the jack lead for the 5v supply on my GS

5 mins ill post the uploaded vids

[Mark Hitchman]

I had lines on my recordings for a start too, sounds different to yours but i'll tell you anyway

I powered my DVR from a 5v BEC and my VRX from a any volt in to 12v out thing and could not understand why the recordings were crap. I kept adding toroids and caps to no avail. Then I put my brain into gear and realised that I had not tied the grounds together for the 5v and 12v systems, worked sweet after that.

Mark

[Coyote]

This is with a very clean 5v supply via DC input


View this video on YouTube

This one is the internal battery result

View this video on YouTube

[FPV-JAPAN]

Guys,

The supplied adapters are 100-240 VAC type so rather than chopping the cables it would be better to use a travel adapter.

In the long run a better solution is to attach a UK mains 3 pin plug to a US 2 pin socket extension lead/box. That way you can purchase from all corners of the world (100-240 auto switching permitting) and not have to worry about spending extra on shipping (UK type psu are much heavier) or buying yet another travel adapter.
If you have more than one non UK psu it's an elegant solution.

I'll see how much an Ikea one is and how heavy they are. May offer them just for cost if you pay the shipping. It's a one time purchase that will last you.

Will be around for a few hours looking out for your videos Ian.

Thanks,

Ren
FPV-Japan

[Coyote]

They are in the post above your Ren

[FPV-JAPAN]

As Mark just said, check your ground (isolated or looped).

But you should all know better than to do funky things with earth returns near me

I'm just firing up the PC (it's not even 7am here yet lol).

Will check your videos and post right back.

Thanks,

Ren
FPV-Japan

[Coyote]

I had lines on my recordings for a start too, sounds different to yours but i'll tell you anyway

I powered my DVR from a 5v BEC and my VRX from a any volt in to 12v out thing and could not understand why the recordings were crap. I kept adding toroids and caps to no avail. Then I put my brain into gear and realised that I had not tied the grounds together for the 5v and 12v systems, worked sweet after that.

Mark

Cheers Mark ( I missed your reply there ) The first thing I did was go through my grounds looking for the dreaded loops, but to no avail

[FPV-JAPAN]

Typing a long reply Ian - stay tuned - will post in a few minutes

thanks,

Ren
FPV-Japan

[FPV-JAPAN]

This is with a very clean 5v supply via DC input


View this video on YouTube

This one is the internal battery result

View this video on YouTube

Ian,

good morning! a cuppa later and my thinking cap on - here we go.

I would say the first video looks like an intermittent connection.
The video isn't dropping out completely (as you would expect) because the intermittent connection is before any output capacitors on the PSU. So I would check connections going into the regulator you are using as opposed to connections coming out (but yes, do make sure the plug in the DVR's socket is not moving).
The diagonal noise is typical of a switching regulator. Best to use a normal 5v regulator with a 2 cell LiPo. If you work from a higher voltage then drop it down with a switching regulator first to conserve your waste/loss to heat and then after adequate filtering drive a 5v linear regulator. Again with adequate output filtering.

More importantly, the second video is definitely a levels issue which I will go into a bit further down.
You can see the video tearing to the side and the H-sync pulse part (black area on the left) moving onto the active video area mainly where the OSD white text is overlaid whereas it remains fine in areas without the overlay.
You can also notice the video gets darker as the tearing happens. This is a classic case of video levels going cold (i.e. moving towards ground) which means the H-sync pulse gets smaller and subsequently the device has trouble maintaining horizontal sync.
Edit: Actually the tearing is more pronounced than the OSD overlay when you have a very bright light source like the window at 00:10. This smells of long cables and cascaded devices to me but I stand to be corrected
Edit2: Ian appears to have short cables so the only thing I can suggest is to track down a device in the signal path that may have poor impedance matching (i.e. off of 75ohms) - check each device in turn to find the culprit - also do be sure to check ground loops (please also check the next few posts with more details on hot video and sync levels)

Video levels can go hot or cold for several reasons

• Over driving the white level in the signal source (many possible causes - would cause the video to go hot)
• Impedance mismatch at any point in the video signal path (most probable - can cause levels to go both hot and cold)
• Ground loop issues that result in the ground voltage rising above 0v (thereby reducing the amplitude of the H-sync and blanking pulses which would also cause the video to go darker)

A quick look at video signal levels

Sync pulses 0v-300mV
Video 300mV - 1v (black - white)

Looking at the sync pulse levels of 300mV we can see that a drop of only 50-100mV yes, only 0.05v-0.1v can cause a lot of trouble here.

A quick recap on ground loops for those not familiar

http://fpv-japan.com/technical-info/...-return-paths/

For video it can be confusing as the video signal path is really not a power return path. It's just for screening the signal from noise, be it coax or a twisted pair.

Key points to remember for ground loops on signal paths are as follows.

Ok to use coax or twisted lead grounded at both source and destination

Device 1 - independent power source (it's own battery etc)
Device 2 - independent power source (it's own battery etc)

NOT Ok to use coax or twisted lead grounded at both source and destination

Device 1 - shared power source (single battery direct or feeding one or more regulators/BEC)
Device 2 - shared power source (single battery direct or feeding one or more regulators/BEC)

In this case you would ground at source and only connect the signal at destination because the signal ground would be taken care of by power source ground

In both the above setups with ground loops taken care of, any video level issues would be due to impedance miss matching (overly long cables or cascading of several video devices in the signal path).

Towards a solution
First, I would want to remove the OSD (and any other gear - TX/RX etc) and test with a clean video input (TV/DVD output) to be happy the device is performing as you would expect, for your own peace of mind.

Then I would want to make sure that the ground loop rules are followed and signal paths are not overly long (a coiled 5m cable for a 1.5m run for example - remember that current flowing to ground in a longer cable will push the 0v rail higher as it gets longer).
Try to avoid cascading more than 2 devices.
Try to use video line driver (a splitter) with variable level control to correct any impedance mismatches that can happen.

If I have confused anyone then please do ask questions.
I may start a thread on these kind of topics if people are interested.

Hope this helps,

Ren
FPV-Japan

[Coyote]

Ok wow ! Thanks for such a detail response. That`s much appreciated.

Ill have a look, one thing Im sure its not is cable lengths. The maximum length of any one cable is one leg of a tripod



Runs to the DVR from the spitter is about 5", same going from DVR out to monitor

Thanks again Ren