PDA

View Full Version : Video filtering, Diode vs. inductor, what's the diff?



changosurf
31st December 2013, 05:46 AM
Aside from the slight voltage drop when using a diode in a video-power filter, what's the difference between doing a 'diode+cap' vs 'inductor/choke+cap'? I read on some archived thread that IBCrazy said that he also used a diode+cap as a filter, using the diode as an alternative to the choke. It seems like the diode would be preferable because of the weight savings, unless I'm missing something here? Can someone please fill me in on the differences/pros/cons of using a simple diode vs an inductor/choke (wound ferrite toroid) as a means of current regulation? feedback would be greatly appreciated...thanks...

rob.thomson
31st December 2013, 05:47 AM
Diode and cap sounds like a simple LC filter.

changosurf
31st December 2013, 06:26 AM
Diode and cap sounds like a simple LC filter.

ok, so what's the difference then between the diode & inductor? same thing? If so, wouldn't the diode make more sense? (saves weight and wire).

rob.thomson
31st December 2013, 06:33 AM
Google to the rescue :)

http://www.daenotes.com/electronics/devices-circuits/filter-circuits

rob.thomson
31st December 2013, 06:34 AM
Just posted a link to a page on the subject, but spam check is stopping it!

No doubt will be approved soon :)

Channel 1
31st December 2013, 07:27 AM
Aside from the slight voltage drop when using a diode in a video-power filter, what's the difference between doing a 'diode+cap' vs 'inductor/choke+cap'?

A diode and cap will help to isolate the load from sags in the supply voltage, the capacitor charges up while the voltage is higher and discharges as it sags and the diode serves to keep the capacitor from discharging back into the source which would negate its usefulness in smoothing out the power supplied to the load.


Adding an inductor to the mix can reduce the possibility of noise and hash from the source making it to the load. This is important when such noise could effect the operation of say a camera or other device that requires clean DC power to operate.

Wayne

Channel 1
31st December 2013, 07:29 AM
Diode and cap sounds like a simple LC filter.

An LC filter is a inductor (L) and a capacitor (C).

A diode and cap would be a "C" filter.

Wayne

IBCrazy
31st December 2013, 12:23 PM
A diode/Capacitor works well if you have very mild video scrambling due to power fluctuations. You need to use a Schotky barrier diode as silicon rectifier diode hold some capacitive charge that can actually amplify the problem. What the diode does is keep reverse power from flowing from the video system where the capacitor holds that voltage.

In my personal filters (and the ones I sell) I use a diode, capacitor and an inductor for maximum benefit. Most people need only a diode and capacitor or an inductor and a capacitor. I used all three just to be sure it works well.

-Alex

changosurf
3rd January 2014, 07:34 AM
Hi Alex, thanks alot for the feedback, I really appreciate it...

I went ahead and made the filter with a 1000uf/25V cap, a 1N4003 diode, and one of these ferrite rings from HK
http://www.hobbyking.com/hobbyking/store/__23207__soft_ferrite_rings_16x7x10_5pc_.html

Here are some pics of the filter+harness:
38252
38253

The diode is hidden underneath the heatshrink right before the choke.
I did 10 winds around the choke, which I figured would be more than enough.
I tested the filter out today on my tricopter and the results looked pretty bad.
Here's a short example video of the interference I saw:

http://www.youtube.com/watch?v=tls9wzujFyI

Anyway, I can't figure out what I did wrong; I would have thought that the filter was over-built & should therefore provide clean power. Obviously, something's not right :(

Up until now, I've been using one of the HK FPV filters, and it's far better than mine by comparison; it eliminates almost 100% of the noise. Since those are almost always on backorder, I figured I'd be ok with going the DIY route.

Any feedback/help on figuring out what I'd need to do to fix this would be greatly appreciated...

rob.thomson
3rd January 2014, 10:40 AM
Try to get the same result with then power cables moved away as far as possible from your osd and camera.

Often... ..it is just radiated interference causing the noise.

I always try keep my power feeds as far away as possible .

turdsurfer
3rd January 2014, 11:24 AM
@Channel 1
Thanks for your explanation because it makes most sense. It's so simple and I've heard of it before but never understood it until now.

@changosurf
It could be numerous other causes, such as tapping your power for the filter too far away from the battery because inductance induced spikes come more into play then (my 1st mistake in my 1st FPV plane that baffled me for months). Try different and no filters and tap from different locations to debug the problem.

Maximus
3rd January 2014, 11:34 AM
Try winding only the positive wire through the toroid, not the ground as well. Do as many turns as you can fit through. I read somewhere that if you use an osd you should only wrap the positive wire of your lc filter. With no osd in the system you wrap both positive and ground wires.

changosurf
3rd January 2014, 06:58 PM
Try to get the same result with then power cables moved away as far as possible from your osd and camera.

Often... ..it is just radiated interference causing the noise.

I always try keep my power feeds as far away as possible .
ok, I'll give that a shot; I'll try to lift the FPV cable runs away from the existing cabling on the frame and see what happens. I've used the same exact physical layout with the HK LC filter and there was no noise though. I'll have to grab the other filter off of my bixler and do a comparison with the new one.

@Channel 1
@changosurf
It could be numerous other causes, such as tapping your power for the filter too far away from the battery because inductance induced spikes come more into play then (my 1st mistake in my 1st FPV plane that baffled me for months). Try different and no filters and tap from different locations to debug the problem.
It's a tricopter, so the distance from the battery to the FPV harness is pretty short. The battery sits right below a 'multi-rotor power distribution PCB' like this one: http://www.hobbyking.com/hobbyking/store/__28255__Hobby_King_Quadcopter_Power_Distribution_ Board_USA_Warehouse_.html?strSearch=power%20distri bution

All three ESC power lines run into that PCB, and there's an additional JST connector tapped into the PCB for the video harness. But, ok, I'll give troubleshooting it a shot.

Try winding only the positive wire through the toroid, not the ground as well. Do as many turns as you can fit through. I read somewhere that if you use an osd you should only wrap the positive wire of your lc filter. With no osd in the system you wrap both positive and ground wires.
Thanks, if all else fails, I'll have to disassemble my filter tonight and try wrapping only the '+' lead.
However, I read somewhere (a post by sassen somewhere I think) that it's supposed to function as a 'common-mode inductor' which requires both positive and negative lines to be wrapped around the coil. But, at this point, I've read so much different, conflicting info on this subject that I'm going cross-eyed and don't know what's right and what's wrong :(

turdsurfer
3rd January 2014, 07:16 PM
I've read so much different, conflicting info on this subject that I'm going cross-eyed and don't know what's right and what's wrong :(
I know exactly what you mean. That's much how I started out.

On my Skywalker that I was referring to, my tap point was just behind the ESC which was just behind the motor, about 20cm away from the battery. That was a disaster in terms of interference due to inductance.
I was also using a common-mode LC filter then (which impedes the flow to ground - but not sure if this was a factor).

On my tricopter (and my other multicopters) I tap my FPV power from the battery balance port. That means distance between tap point to battery is effectively 0 on the motor power leads.
I don't even use a separate LC-filter anymore. I simply step the voltage down from 4S to 9V using a step-down that has sufficient filtering built in:
http://www.fpvflying.com/products/Step-down-PSU-3.3v-5v-9v-12v.html
Use google's search by image feature if you want to find it in other stores.

Good luck troubleshooting by elimination :)

Maximus
3rd January 2014, 11:29 PM
ok, I'll give that a shot; I'll try to lift the FPV cable runs away from the existing cabling on the frame and see what happens. I've used the same exact physical layout with the HK LC filter and there was no noise though. I'll have to grab the other filter off of my bixler and do a comparison with the new one.

It's a tricopter, so the distance from the battery to the FPV harness is pretty short. The battery sits right below a 'multi-rotor power distribution PCB' like this one: http://www.hobbyking.com/hobbyking/store/__28255__Hobby_King_Quadcopter_Power_Distribution_ Board_USA_Warehouse_.html?strSearch=power%20distri bution

All three ESC power lines run into that PCB, and there's an additional JST connector tapped into the PCB for the video harness. But, ok, I'll give troubleshooting it a shot.

Thanks, if all else fails, I'll have to disassemble my filter tonight and try wrapping only the '+' lead.
However, I read somewhere (a post by sassen somewhere I think) that it's supposed to function as a 'common-mode inductor' which requires both positive and negative lines to be wrapped around the coil. But, at this point, I've read so much different, conflicting info on this subject that I'm going cross-eyed and don't know what's right and what's wrong :(

I've been using homemade lc filters on a couple of planes and always wrap only the positive wire. My video is as clean as can be. I think it was actually Sander who mentioned on rcgroups to only use the positive wire. He's been known to argue with himself at times :p

changosurf
3rd January 2014, 11:55 PM
I know exactly what you mean. That's much how I started out.

On my Skywalker that I was referring to, my tap point was just behind the ESC which was just behind the motor, about 20cm away from the battery. That was a disaster in terms of interference due to inductance.
I was also using a common-mode LC filter then (which impedes the flow to ground - but not sure if this was a factor).

On my tricopter (and my other multicopters) I tap my FPV power from the battery balance port. That means distance between tap point to battery is effectively 0 on the motor power leads.
I don't even use a separate LC-filter anymore. I simply step the voltage down from 4S to 9V using a step-down that has sufficient filtering built in:
http://www.fpvflying.com/products/Step-down-PSU-3.3v-5v-9v-12v.html
Use google's search by image feature if you want to find it in other stores.

Good luck troubleshooting by elimination :)


I've been using homemade lc filters on a couple of planes and always wrap only the positive wire. My video is as clean as can be. I think it was actually Sander who mentioned on rcgroups to only use the positive wire. He's been known to argue with himself at times :p

I went for a test flight today. distancing the power leads didn't appear to have helped.
I plugged up the HK LC filter before mine, and it cleaned up things pretty well, just as before.
It almost feels as if my filter is doing absolutely nothing & I'm starting to wonder if maybe the capacitor even works.
Assuming that the cap (which was scavenged) was completely dead, would I be seeing that type of interference in the video?

Moelich, are you also tapping power from the balance plug?
If so, then, crap, that means I'll have to redo my power distribution. Ideally, what I wanted was to be able to plug the battery up only once/in one place and not have to worry about messing with the balance plug in order to power the accessories/FPV harness. But, I guess I can see how having it on the main power rail could cause problems due to the induction. Now, I'll have to have a beer while I redo my filter. We'll see how it goes...

Maximus
4th January 2014, 04:03 AM
No I'm getting power from my main power leads, not the balance leads.

rob.thomson
4th January 2014, 04:09 AM
These days, I always take my camera power from the video power output on the tx module.

I use this to power my osd and camera.

Since then..

Never has a power interference issue!

changosurf
4th January 2014, 04:55 AM
These days, I always take my camera power from the video power output on the tx module.

I use this to power my osd and camera.

Since then..

Never has a power interference issue!

not sure what you meant...you're able to get 12V from you tx? or are you running a 5V video system?


I've been using homemade lc filters on a couple of planes and always wrap only the positive wire. My video is as clean as can be. I think it was actually Sander who mentioned on rcgroups to only use the positive wire. He's been known to argue with himself at times :p

No I'm getting power from my main power leads, not the balance leads.

Moelich FTW!!!

I disassembled the filter, removed the diode, removed the ground wraps, and did as many wraps as possible of the positive lead. Results on the ground/in my garage look good enough for me.

Here's a vid of some full-throttle blasts on the strapped-down tricopter:


http://www.youtube.com/watch?v=jSkw70rHvvg

...and here are some pics of the tri strapped down and the modified filter...
https://www.dropbox.com/sh/3wtmuy2wm6pvp5b/US-2nQhujv

It looks like it would probably help to add another couple-few more wraps, but I ran out of wire on the existing harness, and I don't want to start from scratch since it looks good enough for now, so I'll just leave it as-is and see how it does in a real flight test. The next filter I build will most definitely have a zillion wraps on it...

I'll post another update as soon as I get an actual flight test with this filter, but I'm pretty confident that it will do well in the air.

Thanks again for the assistance!!! I'm soooo stoked, can't believe that it worked sooo well... :D

rob.thomson
4th January 2014, 04:57 AM
Glad you got it sorted.

To answer your question...

The Boscam av tx modules I use have a 12v input, and a 12v power output for the camera.

I suspect this supply is in some.way filtered in the module :)