Mini 3 Pro Antenna Mod (it works for me)

[Siam]

I posted this in another thread but that was actually under the Mavic Mod discussion forum. I thought I would put this here:

I just did a quick mod on my RC Controller. I use to be able to do 2300 ft at 200 agl in an urban environment.

After the mod I did 3800 ft. Same time of day, same flight profile, You guys do the math..but that is drastic. Tons of RF interference and obstructions. EDIT: I did 4800 ft after doing the mount.

I am not a DJI fanboy, nor am I not. DJI makes a spec assuming everything is PERFECT. I would assume that is true. Many fly in areas where foliage, buildings, and other obstructions affect things. Just telling you that 15 bucks and 30 minutes make a big difference for me.

I never removed the original antennas. It would take 5 minutes to plug them back in. I am designing an rf switch that will switch between the two.

The only thing you can see as far as physical changes to the remote are two small (1.5mm) holes to allow the cable to pass through.

I made a quick mount for the antennas with Kydex and it is attached with velcro (good velcro..not cheap stuff..it ain't going anywhere.

My goal is to make an RF switch which will on the back of the antenna mount.

Bottom line: I have done 4 flights with doing this mod. I used the existing patch antennas and the mod. The omni antennas with 8 dbi gain work very well and outperform the stock antennas.

 

[TwoSticks]

It's very interesting, thanks for sharing. So far as I have seen, the remote has not been the weak point it's been the mini 3. My Air 2S with the RC Pro has much more range than my mini 3 in the same area but if this helps I guess it's worth the mod. Those antennas have to be an improvement on what is found internally.

 

[PaJi]

Interesting. I vacationed recently in NH on a lake. I could fly my M3P out over the lake as far as I wanted with no signal problems. However when I flew up and over a large hill covered with trees I seemed to lose connection much sooner than I did with my Mini 2. Thx for sharing.

 

[Siam]

It's very interesting, thanks for sharing. So far as I have seen, the remote has not been the weak point it's been the mini 3. My Air 2S with the RC Pro has much more range than my mini 3 in the same area but if this helps I guess it's worth the mod. Those antennas have to be an improvement on what is found internally

Any system (Ham radio, satellite comms, drones, Mobile phones, VHF/UHF/comms) all have a uniqueness to them. Some rely on the power and receive capabilities of one unit while the other has limitations. For example, A person operating 5 Watts from Texas trying to communicate with a station in Germany needs lots of things going in their favor. One is that the TX transmitter is operating on the best possible conditions. Directional would be nice LOL. The station in Germany can pick up the signal if they have a good directional antenna. They can send a signal that can be received by the TX station if they have enough power.

Same thing with aircraft. A long time ago I worked on a system that used directional tracking dishes to follow recon planes. They produced a lot of transmit power and had a large dish to receive weak signals. The aircraft had a very small dish (1 foot or so) and had limited power. It worked.

I have not done any more testing but I think that an additional set of antennas on the Mini 3 Pro would be beneficial. (I am not exactly sure where they are now.

I am not so much concerned about the distance, but the ability of the drone and the controller to stay in contact with each other under more unfavorable conditions.

A long time ago we experimented with COFDM transmissions. They work great in urban environments and for weak signals and multi-path fading.
I don't know if DJI has looked into this or incorporated that into any transmission scheme that is being used.

 

[Siam]

Interesting. I vacationed recently in NH on a lake. I could fly my M3P out over the lake as far as I wanted with no signal problems. However when I flew up and over a large hill covered with trees I seemed to lose connection much sooner than I did with my Mini 2. Thx for sharing.

When I retired I moved to NH and I worked for a communications company. Trust me...I understand the challenges of that state well LOL

Lake Ossipee by any chance? I moved to Conway.

 

[YVRGUY]

Siam, why the RF switch? Can you streamline the design by tucking the leads inside with the antennas mounted on top?

I like the ingenuity of improving the signal’s strength and directional quality. I also like the compactness of the DJI RC controller and it’s portability…it seems the sacrifice is giving up the sleek design. I don’t fly too far with my M3P except, out in the wide open spaces away from civilization. Again, I like the ingenuity and experimentation of what you’re doing.

One last question. Would the RC PRO antennas be adaptable to the MP3? Just curious

Thanks for sharing and good luck in your R & D

 

[Siam]

Siam, why the RF switch? Can you streamline the design by tucking the leads inside with the antennas mounted on top?

I like the ingenuity of improving the signal’s strength and directional quality. I also like the compactness of the DJI RC controller and it’s portability…it seems the sacrifice is giving up the sleek design. I don’t fly too far with my M3P except, out in the wide open spaces away from civilization. Again, I like the ingenuity and experimentation of what you’re doing.

One last question. Would the RC PRO antennas be adaptable to the MP3? Just curious

Thanks for sharing and good luck in your R & D

1: As far as the cables: yes. I just was doing a test setup. They can easily be stored inside or tucked away neatly.

2: The RF switch would be so I could easily switch between the internal antennas and the external for quick testing. I plan on doing some other mods and this way I could always go back to the DJI original controller for a baseline. Another reason would be to keep the streamlined design if I just want to take it out of the case real quick. The mount is only held on by velcro

3: Any antenna designed for 2.5/5.8 GHZ should work.

4: Another reason I am doing this is because I like to fly low through canals, thorugh woods and stuff like that. I want the best signal possible when there are obstructions.

 

[tlyons]

I posted this in another thread but that was actually under the Mavic Mod discussion forum. I thought I would put this here:

I just did a quick mod on my RC Controller. I use to be able to do 2300 ft at 200 agl in an urban environment.

After the mod I did 3800 ft...

Link below to earlier thread with discussion, link to video instructions, link to antenna, etc. The mentioned Alfa APA-M25 further improve the range compared to omni. Both will readily bring you out to maximum reasonable VLOS in even the worst conditions.

DJI RC External Antennas

I’d rather buy from another drone company that cares for his customers. We really need another competitor in the market to really challenge the price and practices of dji That idea has been around for years and proclaimed by many before. Their sales have not suffered by it yet.

mavicpilots.com

Good luck!

 

[YVRGUY]

1: As far as the cables: yes. I just was doing a test setup. They can easily be stored inside or tucked away neatly.

2: The RF switch would be so I could easily switch between the internal antennas and the external for quick testing. I plan on doing some other mods and this way I could always go back to the DJI original controller for a baseline. Another reason would be to keep the streamlined design if I just want to take it out of the case real quick. The mount is only held on by velcro

3: Any antenna designed for 2.5/5.8 GHZ should work.

4: Another reason I am doing this is because I like to fly low through canals, thorugh woods and stuff like that. I want the best signal possible when there are obstructions.

Thank you

 

[Siam]

 

[Siam]

A cool thing happened today. Swapped out antennas with the Antennas from a Netgear modem I was getting rid of. need to check the specs on them..they could be 9dbi or 10...who knows? Just testing since I was bored. The range is now 6000 ft. I hooked up the stock RC Controller antenna's and it was still only 3600-3800 feet, I might just get 10 db gain antennas and see how it performs so I have a spec I can quote.

 

[AZDave]

It's very interesting, thanks for sharing. So far as I have seen, the remote has not been the weak point it's been the mini 3. My Air 2S with the RC Pro has much more range than my mini 3 in the same area but if this helps I guess it's worth the mod. Those antennas have to be an improvement on what is found internally.

When using a passive antenna mod, it improves the RF path in both directions. The remote transmits to the drone better, and the remote also receives FROM the drone better. Of course you never get something for nothing, and the cost for greater range is that you end up with a narrower beam width so you have to be more diligent about where you are pointing the remote.

 

[AZDave]

Interesting. I vacationed recently in NH on a lake. I could fly my M3P out over the lake as far as I wanted with no signal problems. However when I flew up and over a large hill covered with trees I seemed to lose connection much sooner than I did with my Mini 2. Thx for sharing.

RF over a lake (as opposed to through it) has very little loss, and the RF that hits the lake at such a shallow angle probably gets reflected and is still usable. RF that tries to bounce off trees gets mostly absorbed, and there is no such thing as RF reflecting off trees at a shallow angle because the foliage is large and rough compared to the wavelength of the RF. What you experienced fits the physics perfectly.

 

[AZDave]

Those larger external antennas are definitely going to give you better performance. Nice job!

When you make your RF switch to do the comparison, make sure that it is symmetrical. Any switch is going to have loss at these frequencies, so make sure that the path for the stock antennas and the path for the new antennas are as similar as you can make them. Once you finish your tests you probably will want to eliminate the switch.

Also, the pattern gain you get from two phased vertical antennas is a function of the spacing between the antennas. I'm not sure I would rely on the spacing for the stock antennas to be optimum since DJI probably had other factors to consider, like case size or distance from the antennas to the circuit board. If you really want to optimize the external antennas, and if you are willing to go through the hassle of more tests, make the spacing between the antennas adjustable. Theoretically, 5/8 of a wavelength should give the optimum broadside gain for two verticals fed in phase, and maybe you already set it to that.

Lastly, to make this even more complicated (sorry), the feedpoint impedance of two verticals side by side and fed in phase is at least partially a function of the spacing between them. We don't really know what input impedance the circuit board has (it probably has a stripline tuning network), so there could be a tradeoff between the spacing for optimum raw antenna gain and the spacing for the best match to the circuit board.

 

[Siam]

Those larger external antennas are definitely going to give you better performance. Nice job!

When you make your RF switch to do the comparison, make sure that it is symmetrical. Any switch is going to have loss at these frequencies, so make sure that the path for the stock antennas and the path for the new antennas are as similar as you can make them. Once you finish your tests you probably will want to eliminate the switch.

Also, the pattern gain you get from two phased vertical antennas is a function of the spacing between the antennas. I'm not sure I would rely on the spacing for the stock antennas to be optimum since DJI probably had other factors to consider, like case size or distance from the antennas to the circuit board. If you really want to optimize the external antennas, and if you are willing to go through the hassle of more tests, make the spacing between the antennas adjustable. Theoretically, 5/8 of a wavelength should give the optimum broadside gain for two verticals fed in phase, and maybe you already set it to that.

Lastly, to make this even more complicated (sorry), the feedpoint impedance of two verticals side by side and fed in phase is at least partially a function of the spacing between them. We don't really know what input impedance the circuit board has (it probably has a stripline tuning network), so there could be a tradeoff between the spacing for optimum raw antenna gain and the spacing for the best match to the circuit board.

Thanks for the reply. All great thoughts. I made the bracket it it is no wider than the controller itself. And unless my math is wrong...splitting the difference between bands.... 5/8 wavelength of a 3 GHz signal is about 2.5 inches.

I know I might get some loss with a switch, but that is why am designing a solid state switch, not a mechanical switch. Loss might be minimal in comparison. But who knows? Doing this more for the love of electronics LOL. And you bring up a good point about removing the switch when done: I designed it this way so I can easily switch back and forth between antennas since I did not remove the originals. One thing I could do is just use two jumper antenna cable so I can manually switch from the outside whenever I want without a switch...but what is the fun in that

Thanks for you insights! It does make me rethink a few things.

 

[tlyons]

A cool thing happened today. Swapped out antennas with the Antennas from a Netgear modem I was getting rid of. need to check the specs on them..they could be 9dbi or 10...who knows? Just testing since I was bored. The range is now 6000 ft. I hooked up the stock RC Controller antenna's and it was still only 3600-3800 feet, I might just get 10 db gain antennas and see how it performs so I have a spec I can quote.

Sounds like fun. But limit how many times you plug and unplug the cable assemblies from the controller PCB. That connector is likely only rated for 25 plug/unplug cycles. The SMA side you change as often as you like.

But you did discover the something that is often true: people selling branded assemblies take more care that their antennae actually work. In other projects, I have seen a 2dB stubby antenna from a name brand router out perform a 9dB generic from Ebay. I have a high quality Netgear router I am retiring and I plan to try out the dual band antenna from it.

 

[tlyons]

Those larger external antennas are definitely going to give you better performance. Nice job!

When you make your RF switch to do the comparison, make sure that it is symmetrical. Any switch is going to have loss at these frequencies, so make sure that the path for the stock antennas and the path for the new antennas are as similar as you can make them. Once you finish your tests you probably will want to eliminate the switch.

Also, the pattern gain you get from two phased vertical antennas is a function of the spacing between the antennas. I'm not sure I would rely on the spacing for the stock antennas to be optimum since DJI probably had other factors to consider, like case size or distance from the antennas to the circuit board. If you really want to optimize the external antennas, and if you are willing to go through the hassle of more tests, make the spacing between the antennas adjustable. Theoretically, 5/8 of a wavelength should give the optimum broadside gain for two verticals fed in phase, and maybe you already set it to that.

Lastly, to make this even more complicated (sorry), the feedpoint impedance of two verticals side by side and fed in phase is at least partially a function of the spacing between them. We don't really know what input impedance the circuit board has (it probably has a stripline tuning network), so there could be a tradeoff between the spacing for optimum raw antenna gain and the spacing for the best match to the circuit board.

Likely this is a diversity antenna arrangement and it is just picking the strongest signal unlike beam forming or some kind of MIMO. If not, I would have expected it to be difficult to improve the performance so easily. OTOH, maybe not. It is amazing such a low power radio works so well so maybe there is more cleverness in the design than I expect.

I would be curious if you know what they actually implemented.

Good luck!

 

[Siam]

Sounds like fun. But limit how many times you plug and unplug the cable assemblies from the controller PCB. That connector is likely only rated for 25 plug/unplug cycles. The SMA side you change as often as you like.

But you did discover the something that is often true: people selling branded assemblies take more care that their antennae actually work. In other projects, I have seen a 2dB stubby antenna from a name brand router out perform a 9dB generic from Ebay. I have a high quality Netgear router I am retiring and I plan to try out the dual band antenna from it.

Yuppers on the antenna connectors. Those pesky little U.FL IPX IPEX connectors are tiny and one bad press you are toast....again...the reason for a switch Though the lazy side of me now is just thinking about using two small RG-178 cables so from the outside I can switch back and forth and not worry about the connectors on the board.

 

[YVRGUY]

RF over a lake (as opposed to through it) has very little loss, and the RF that hits the lake at such a shallow angle probably gets reflected and is still usable. RF that tries to bounce off trees gets mostly absorbed, and there is no such thing as RF reflecting off trees at a shallow angle because the foliage is large and rough compared to the wavelength of the RF. What you experienced fits the physics perfectly.

 

[Siam]

RF over a lake (as opposed to through it) has very little loss, and the RF that hits the lake at such a shallow angle probably gets reflected and is still usable. RF that tries to bounce off trees gets mostly absorbed, and there is no such thing as RF reflecting off trees at a shallow angle because the foliage is large and rough compared to the wavelength of the RF. What you experienced fits the physics perfectly.

FYI: I saw the results of studies done by DoD and DARPA about how foliage affects RF signals. Very interesting info. Big difference in how things react to pine needles versus leaves from an Oak tree versus Triple canopy jungles. That takes into account the freq, modulation, and angles of the signal.