Replies to This Discussion

Permalink Reply by Rick Eis on August 11, 2012 at 7:05pm

Hmm...very good. Thank you Vernon.

I will have to read this a few time :)

Let me ask you this. Lets say I provide a small battery source for each; The Radio receiver, The apm2, and the Video.  Should this eliminate most interference.  My thoughts are then I could eliminate one source at a time and hopefully find the source......

I do get a bit confused with the ground loop concept.  If I have a separate source for my radio receiver ...don't I need to take the ground wire to the APM2?  And if I connect an + and - from an ESC doesn't that create a ground loop ?

Maybe you could point me to a wiring diagram that might help me?

Thanks Vernon for your time.  Again I will be rereading your post several times when the sun goes down.  Right now I am having too much fun flying :)

Thanks again,

Rick

 

Permalink Reply by Vernon Barry on August 11, 2012 at 8:17pm

Ok, sorry, the posting tree is maxed, so I cannot reply in line.

Anyway.

Some notes: The minimOSD is supposed to isolate the 5 volt and 12 volt sides inlcuding ground. I even have one, but haven't yet gotten to this level of testing.

I say that because the mentioned ground loop. Basically, a ground loop happens when you have mutliple devices that have different physical ground paths and are also connected to each other. One example is that the ground for the APM goes through the ESC signal wires then to the battery from the ESC. That wire length has some voltage drop, but since it's a return path, in theory, ground at the APM is maybe a few 10th of volt above ground at the battery. Now say you attach the video power wires direct to the battery and that ground wire is shorter or heavier than the same path taken by the APM ground. Then you could have maybe a 10th of a volt difference between the two ground paths. The problem is that when the motors start drawing heavy current, now that short peice of wire between the ESC and the battery is carrying more current, and being return current, now the APM could be as high as 1/2 volt above battery ground, The other wire path direct to the battery ground for the video is only carrying the current required for the video stuff and thus is closer to ground (less voltage drop) and now there is a half volt different between the two grounds with all that motor noise as spikes. This is why we either serially ground or star ground, but even then, star ground can cause issues when adjacent devices are interconnected.

 

So say you followed this:

 

As long as you did not connect in any way connect grounds on the right hand side, to ground on the left hand side, or solder the optional jumpers on the board. You should not have a ground loop. That's good and probably the case. The problem is, I cannot vouch for how well the isolation factor is through the OSD chip. It might be great or it might not. We do know, if the battery shown in the above diagram is the same battery that powers the motors,then there is all kinds of noise on the power wires going to the minimOSD and camera and TX.

The plan to isolate it and just power via a different battery would be my first approach. This likely will solve the problem. When you get to different power sources for the APM, you are opening pandora's box of new problems and potential to just make things worse on many levels.

If powering the video side with a battery does work, but you don't want the extra weight of the battery, then we need to implement the filter I mentioned in the other post. It should work, but if you have the battery on hand, the isolation test will verify the theory and then we can work on the problem.

 

Permalink Reply by Rick Eis on August 11, 2012 at 9:58pm

Vernon,

Wow!!  yes !   I put the video system on its own power source.  It is as clean as when the motors are not running.  Tomorrow I will do a distance test on the ground with things running... but I bet it will be awesome!!  Thank you so much for taking the time to help.

I power my apm2 and my radio receiver off a shielded 3 amp BEC.  I'm currently using this one  ....  http://www.innov8tivedesigns.com/product_info.php?cPath=41&prod...  It is connected to the same battery source as my ESC's which are plugged into the APM2.  I do not have the jumper (j1) installed on the APM2.

Does this sound like a ground loop?

I think the BEC is filtering my APM right?  Your sentence scares me, "you could be losing the RC link too and just don't know it yet".  I want to find out before it causes problems... :)

You said, "I would place an inductor (say 100mH) in series on the positive lead and a cap in parallel to the ~12 V feed you are sending to the video setup (camera, OSD, and video TX).  

Your talking the setup as in the attached dwg...yes? Is this what you are calling full strength filtering?

Do you know any good online source to purchase the 100mH inductor and the cap?  

 

Thanks again Vernon for your time :)

 

Rick

Attachments:

• filter.gif, 4 KB

• ▶ Reply

Permalink Reply by Rick Eis on August 11, 2012 at 10:18pm

One reason I don't want to run with another power source is that I am not sure we can monitor more then one battery source in mission planner?  And I have a difficult enough time monitoring just that one. ha ha :)

• ▶ Reply

Permalink Reply by Vernon Barry on August 11, 2012 at 11:14pm

Yes, I totally understand not wanting to carry 2 batteries. So, just build the filter and put in in line for the power for the camera, OSD,and video TX.

The point is, the ground wire basically goes straight through, The power wire will have the inductor in series, and the capacitor should be a 25volt electrolitic of any value between say 150uF to 3300 uF. The diagram you linked was perfect. If you have a ferrite core, you can try making the inductor by just coiling the positive wire a couple of times (as many as will fit through the ferrite bead). In both cases, the larger the value, the lower the pass through frquency of noise would be. At some point, it won't matter how big, it just blocks anything not DC.

Old PC power supplies or parts are full of this stuff.

• ▶ Reply

ModeratorPermalink Reply by Graham Dyer on August 12, 2012 at 5:50am

Very informative, thanks Vernon. I have some range issues with a 0.9-1.3GHz video Tx/Rx and my 2.4GHz RC system (Hitec) where I lose RC at distances further than about 800m with the video running. Would this type of filter help me too?

• ▶ Reply

Permalink Reply by Vernon Barry on August 12, 2012 at 6:19am

There's no way to know for sure without testing , but the point is, it cannot hurt. Again, I know that the module design of most cheap video TXs they are jsut adding a powerful broadband amp, and may not be putting sufficient filtering on the input of the amp to reject anything outside of the exact 900MHz signal so noise can get amplified.

The question is, are you generating noise that covers 2.4GHz?

Further, go back and read the thread (Page 1) about radio loss calcs based on fequency.

http://diydrones.com/forum/topics/question-on-frequency-interferenc...

Your RC tx is likely 100mW or 20dBm @ 2.4GHz (the m indicates a power and is relative, see the chart). At a normalized 1.6 miles (so there is an even comparison between frequencies) 108dB loss. The same power and using 900Mhz would result in 99dB loss. That's 9 db difference, but explains why you aren't getting the range out of your 2.4 Ghz RC radio, but have no issue with 900Mhz. The guys pushing the 1500mw 900MHz video TXs are either going for really long range, or might just be a bit of overkill and hurting control channel performance.

You might also have an RC TX that is not 100mW so that would even further explain your limited range.

All I'm saying is that people blindly are choosing these TXs and equipment frequencies without ever even looking at a power or link loss chart. 5 Minutes with the chart and likely, many would change their entire setups. Anybody who bought 5.8GHz and expected range out of it is just pissed off right now. The point being, you should do a radio link budget and then if the actually testing shows you are out of budget for the range then you assume interference is the cause. The calcs work and help you balance the entire system since you have multiple channels. Blatantly buying the highest powered TX you can buy is not the best choice.

• ▶ Reply

ModeratorPermalink Reply by Graham Dyer on August 12, 2012 at 6:37am

Thanks, the VTx is 700mW and the RCTx is (AFAIK) 100mW, I deliberately went for the 0.9-1.3 to try get the most range from the video system but have read that there's some harmonic from that into the 2.4GHz band.

After I first put the system together I did a flight to 2.6km without mishap so am not sure why the shorter range now, apart from a little less separation of the VTx and RCRx, all the time using an inverted vee antenna for the VTx pointing forward. I'll try the filter, just dug up some old power supplies and they have a selection of inductors I can use.

• ▶ Reply

Permalink Reply by Vernon Barry on August 12, 2012 at 6:54am

The inductor size is not critical and anything out of a PC power supply is definitely going to work. By definition, they are switcher supplies in the kilohertz range, thus the filters inside are perfect for our use.

Also, head what I said about mismatched antennas. If the power doesn't couple from the antenna to the air, the energy must go somewhere so it travels back down the wires from the TX into the rest of your electronics. Thus if you changed antennas from the first flight, that could be the culprit. The filter can and will try to stop this too, but fix the problem as a mismatch is hard on the TX too.

• ▶ Reply

ModeratorPermalink Reply by Graham Dyer on August 12, 2012 at 7:58am

The antenna has been the same all the time on a 30cm extension cable, although very recently I tried moving it back directly onto the VTx to see if that helped but haven't had a chance to do a longer distance flight, a bit nerve-wracking too when you know you are fairly sure that you will fly out of RC range (at a point when you're out of LOS too).

• ▶ Reply

DeveloperPermalink Reply by Mark Colwell on August 12, 2012 at 3:10pm

I have a 200mw 5.8ghz set with IBCrazy antenna set works well. I think I.3ghz tx may cause GPS issues.

I use EagleEyes Diversity too.

You might get 910ghz to work, try this, turn on video tx before 3DR radio, so 3DR selects open channels outside video tx band.

I prefer to use a better video rx set and quality antenna vs high power tx on sUAS to save power.

• ▶ Reply

• ‹ Previous
• 1
• 2
• Next ›
• Page