Typhoon H antenna upgrade - some (helpful?) points

[deluge2]

I now have several flights in after mounting my H antennas external. My H is pretty solid for the most part and I have had no major issues since the upgrade.

Standing near the H I can raise and lower the gear with zero glitches. I have noticed a couple times if I'm in my vehicle where the ST16 may not have the most direct link to the H and I operate the gear there can be a momentary glitch.

The glitch only seems to happen when the gear is in motion. This makes me believe it is some kind of RF or noise from the retract motors in action. Orientation does not seem to matter only the gear in operation.

Thanks for posting, I completely agree with your thoughts about what might be going on.

While I haven't noticed video link glitching specifically when the retract motors are in action, I haven't systematically looked for it yet either. Now that weather on the US East Coast is not so conducive to frequent flying, it might be time to do some safe, legal range testing (aircraft on but not actually flying). If I can figure out a way to have someone hold the YTH (with no props obviously) in a manner that permits gear actuation, it would be interesting to see if glitching occurs when there is substantial distance between the ST16 and the aircraft. I would think that susceptibility to RF interference would be greater when the SNR is reduced (by distance), and so this admittedly atypical setup could help determine whether operation of the retract motors can reliably produce glitching. I have not yet attempted any modifications to the aircraft antennae, but that may be on my list for winter upgrades.

 

[ArnhemAnt]

Keen to hear how this goes - please report back once you have had a chance to do some testing.

And, I hope the weather improves for you so you can get out and fly again.

 

[deluge2]

To ensure you have the best signal link(radio and video), you need to try and rule out any possible interference. Carbon fibre has the tendency to block/interfere with radio/video signal. With the landing gear down (not ideal) the legs are almost parallel with the antenna on the underside of the H and will have an impact on signal strength, etc. Depending, of course on where you/the ST16 are in relation to this. When the landing gear is raised (what most people will do whilst flying), this lessens the issue, however, some situations (banking turns, being above the H, etc) will increase the chances of the CF restricting the signal.
Distance also plays a critical part in this. While both the H and ST16 are close to each other, the chances of losing link quality are pretty low, however, this is amplified the further each are apart from the other.

Put simply, there are a lot of variables that may come into play in regards to signal strength and quality, and the landing gear (legs) are just one of them. The antenna mod for the original H and the new antenna's on the Pro will all help to alleviate things.

Thanks for the reply. I don't disagree with anything you've said, but my guess is that passive blocking of RF signals by the CF components of the LG is unlikely to be a significant factor, compared to the many other variables that, as you state, may come into play. This is certainly true when the gear are up, in which case the CF components of the LG are in essentially the same orientation relative to the link antennae as are the CF tubes for the 6 motor booms. And all of these members are well clear of the signal path for either the aircraft or the camera during normal flight (i.e. for most photo video work). I see your point that under more extreme attitudes, the retracted gear could be positioned differently relative to the transmitter, but keep two additional points in mind: 1) even at extreme attitudes, the relative positions of the CF components of the retracted LG and the antennae remain unchanged, and 2) the orientations of the CF components of the LG are not wholly dissimilar to those of the CF motor booms.

I do agree that the kinds of mods being made (by Yuneec, 3rd parties, and end users) are probably a good thing. So far I have only replaced the 2 external antennae on my ST16 and it seems to have improved link robustness. But I haven't done any controlled A/B comparisons. I look forward to conducting some more systematic testing and to learning more about the experiences of others in this and related threads. Still contemplating when/whether to do the mods to add a 3rd external antenna to the ST16 and to mount optimized external antennae to the aircraft itself.

 

[deluge2]

Keen to hear how this goes - please report back once you have had a chance to do some testing.

And, I hope the weather improves for you so you can get out and fly again.

I'll drink to that! I also need to push the limits a little more as regards flying in the winter months. We've all seen some wonderful photos and videos from winter flights. Just need to be extra conservative regarding battery temps and max energy use during flights.

 

[ArnhemAnt]

. Still contemplating when/whether to do the mods to add a 3rd external antenna to the ST16 and to mount optimized external antennae to the aircraft itself.

I've just completed the mods to both the ST16 (including the Itelite antenna) and the YTH. Not enough flights done yet to be able to provide any real feedback, but I'll be sure to do this a little later. Both mods were relatively easy to do. If you do go down this road, pay particular attention to the images and instructions from the Yuneec Typhoon EU FaceBook page (Carlos) as he covers some detail that is not included in the YT clips.

 

[Flying Egle]

purchased a used second H standard and did the pro antenna upgrade and there is still a 4th 5.8 antenna tapped to inside of case............ when I get a few more parts I am going to move that one to exterior also......just not sure where.... but want to keep it symmetrical...... dont understand why Yuneec had these antenna 1-2.4 and 1-5.8 inside......

See pics of Mod

 

[ArnhemAnt]

^^ No read need to bring that last 5.8 outside the ST16. It's for the WiFi connectivity only, and does a pretty good job where it is already.

 

[terrence davidson]

I wondering now if I could relocate the radio modular on the h480 to the location it is at on the h520. which is at the led area of the h480 at the rear I believe.

 

[homedoc]

I've recently upgraded the antenna's on my Typhoon H (Yuneec External Antenna Kit for Typhoon H, & Pro Upgrade Kit) (and ST16: ITELITE DBS Upgrade for Yuneec Typhoon H, Tornado H920L, ST16, ST24 Radio) and after the first few flights, I noticed that there was still some slight 'glitching' with the video, even at close distances of less than 50m. Nothing really bad, but I am fussy about this stuff.
After some thought, I decided to go back to the H and dismantle it again so I could get a look at the wiring. My idea was to put some copper shielding between the battery and the antenna cables, and to also re-route the antenna wire on the right side (facing the front of the H). This is also the same side as the radio module and this is where I think that people are having some issues when the landing gear is raised or dropped.

Finished product now has the copper shield in place which should negate any possible RF issues between the battery and the antenna wires and I have now re-routed the antenna wire on the right side (same side as the radio module) so that it is as far away as possible to the landing gear mechanism.

I plan to head out over the next few days and test this to see if there are any noticeable differences.

Hope this may help other members out there in some way.

Pics attached to hopefully show what I am talking about.

View attachment 4459
Bottom cover removed to show the existing array. Note the radio module on the left of the image, which is the right hand side when upright and facing forward.

View attachment 4460
With the bottom cover removed, you can see that the left side antenna wire goes across the base of the battery tray.

View attachment 4461
Copper shield attached.

View attachment 4462
Inside the battery tray, showing the shield.

The copper shielding must be grounded or it will do nothing. You need to solder a wire to it and then attach the other end to the ground terminal of the battery.

Also, you can use thin copper screen and it will work just as well but it's substantially less heavy.

If you really want to work on this, it would be better to actually find the source of the RF rather than trial and error. You need to get access to an oscilloscope. Then you make a little (1/4 inch diameter) coil of copper wire and attach it to the end of the scope probe. The coil serves as an antenna and will pick up all the noise near it. So you can scan around inside the H, moving it near to each component until you see a sudden jump in the scope signal. Whatever the probe is near when that happens, that's what you need to shield.

Last thing: usually the biggest source of noise is from power switching. The H controls its motor speed using pulse-width modulation. That means that the ESC is generating a square wave of a certain frequency and controlling the speed of the motor by varying the length of time that the square wave is positive. More positive time equals higher motor RPM.

The main source of noise occurs at the leading and falling edge of the square wave. This occurs because at the rising edge some capacitors charge up in the electronics and the motor winding acts as an inductor. At the falling edge, the magnetic field around the motor coil collapses and as it does so it generates a noise spike because the collapsing magnetic field is moving through the motor coil. So that is one source of noise from the square wave.

The second source of noise is the square wave itself. This explanation is a little bit technical but it is well known that a square wave of a given frequency is actually composed of a whole series of higher harmonic sine waves frequencies added together. I.e a sine wave at 2x the frequency, another at 4x, 8x, 16x etc. So when the power switches on or off in a square wave pattern, at every edge there will be a noise pulse containing the higher harmonics. The lower harmonics are not too much of a problem. However the higher frequencies are in the radio area of the spectrum, and they radiate out of the wire as RF noise.

So why, you may ask, do they use square wave pulse width modulation to power and control the motors? They use it because: (1) it is very easy to generate square waves, and (2) the power transistors in the ESC that are generating the power pulses are way way more efficient when they are either fully on or fully off. If we were to use sine waves or triangle waves instead, the transistors would get very hot and would fry the ESC. So to use something other than a square wave, the ESC would have to be re-designed using much bigger power transistors with heat sinks, etc. Too expensive and too heavy.

So what's a poor pilot to do? You can either shield the modules that are sensitive to the noise, or you can shield the sources of the noise (most likely the motors and the ESCs) or both.

Still waiting for my H+ to arrive and if you are still grappling with this issue I can probe around inside the drone and try to find the biggest sources of noise. Let me know if you need me to do this.

I hope this helps.

 

[Dr Delta]

The copper shielding must be grounded or it will do nothing. You need to solder a wire to it and then attach the other end to the ground terminal of the battery.

Also, you can use thin copper screen and it will work just as well but it's substantially less heavy.

If you really want to work on this, it would be better to actually find the source of the RF rather than trial and error. You need to get access to an oscilloscope. Then you make a little (1/4 inch diameter) coil of copper wire and attach it to the end of the scope probe. The coil serves as an antenna and will pick up all the noise near it. So you can scan around inside the H, moving it near to each component until you see a sudden jump in the scope signal. Whatever the probe is near when that happens, that's what you need to shield.

Last thing: usually the biggest source of noise is from power switching. The H controls its motor speed using pulse-width modulation. That means that the ESC is generating a square wave of a certain frequency and controlling the speed of the motor by varying the length of time that the square wave is positive. More positive time equals higher motor RPM.

The main source of noise occurs at the leading and falling edge of the square wave. This occurs because at the rising edge some capacitors charge up in the electronics and the motor winding acts as an inductor. At the falling edge, the magnetic field around the motor coil collapses and as it does so it generates a noise spike because the collapsing magnetic field is moving through the motor coil. So that is one source of noise from the square wave.

The second source of noise is the square wave itself. This explanation is a little bit technical but it is well known that a square wave of a given frequency is actually composed of a whole series of higher harmonic sine waves frequencies added together. I.e a sine wave at 2x the frequency, another at 4x, 8x, 16x etc. So when the power switches on or off in a square wave pattern, at every edge there will be a noise pulse containing the higher harmonics. The lower harmonics are not too much of a problem. However the higher frequencies are in the radio area of the spectrum, and they radiate out of the wire as RF noise.

So why, you may ask, do they use square wave pulse width modulation to power and control the motors? They use it because: (1) it is very easy to generate square waves, and (2) the power transistors in the ESC that are generating the power pulses are way way more efficient when they are either fully on or fully off. If we were to use sine waves or triangle waves instead, the transistors would get very hot and would fry the ESC. So to use something other than a square wave, the ESC would have to be re-designed using much bigger power transistors with heat sinks, etc. Too expensive and too heavy.

So what's a poor pilot to do? You can either shield the modules that are sensitive to the noise, or you can shield the sources of the noise (most likely the motors and the ESCs) or both.

Still waiting for my H+ to arrive and if you are still grappling with this issue I can probe around inside the drone and try to find the biggest sources of noise. Let me know if you need me to do this.

I hope this helps.

Sounds very interesting, although I do not own an H+.... but a 480 pro RS....

 

[DCH]

The copper shielding must be grounded or it will do nothing

This is not necessarily true. I would like to clarify a technical point here. A shield itself -solid or mesh- will block RF (as long as the voids in the mesh are smaller than the wavelength of the frequency that is being blocked. The higher the frequency the smaller the holes need to be. RF Signals cannot pass thru metal -period. Bonding (grounding) the shield will drain any potential charge to ground. This is definitely a good idea whenever possible, however it will not actually aid the shield in blocking RF. So to say it will do nothing unless grounded is technically incorrect.
You can prove this yourself, just lay a piece of foil over the GPS antenna and watch what happens to your signal.

 

[homedoc]

This is not necessarily true. I would like to clarify a technical point here. A shield itself -solid or mesh- will block RF (as long as the voids in the mesh are smaller than the wavelength of the frequency that is being blocked.

Yes, DCH is technically correct. The radio waves will generate circular eddy currents in the copper and they will go round and round till their energy is converted to heat. This is the same idea as induction heating. However, if I remember correctly, the eddy currents themselves can generate more radio waves while they are still circulating. That is why generally accepted EE practice is to ground all EMF shields wherever possible.

For example, in the lab if you are conducting experiments with high sensitivity electronic equipment such as radio receivers or high gain amplifiers, the accepted practise is to shield the equipment in a "Faraday Cage". A Faraday cage is a construction made of copper screening which surrounds the equipment on all sides, including the top and bottom. Faraday cages are ALWAYS grounded.

Interesting point about the aluminum foil around the GPS antenna. I wonder what happen if you tried the same experiment both grounded and ungrounded. Would the aluminum shielding effect be stronger if the foil was grounded? I don't have the answer, but maybe DCH does.

 

[DCH]

Yes, DCH is technically correct. The radio waves will generate circular eddy currents in the copper and they will go round and round till their energy is converted to heat. This is the same idea as induction heating. However, if I remember correctly, the eddy currents themselves can generate more radio waves while they are still circulating. That is why generally accepted EE practice is to ground all EMF shields.

Exactly right! Though not nearly as "energetic" and existing at lower levels these eddies can cause their own RFI. Bonding a shield is always the best practice. However it is not always done, for one reason or another -mostly out of convenience or to save a buck. Take the shield inside the top body shell of the H series for example, the one protecting the main board and separating it from the GPS board, that is not bonded to anything, yet is still rather effective.

 

[homedoc]

Take the shield inside the top body shell of the H series for example, the one protecting the main board and separating it from the GPS board, that is not bonded to anything, yet is still rather effective.

"Rather" being the operative word It might be a very good idea to ground that shield, perhaps with a wire with an alligator clip on one end for conveniently clipping it onto the shield before reassembling the unit. Might solve a few problems.

 

[DCH]

"Rather" being the operative word It might be a very good idea to ground that shield, perhaps with a wire with an alligator clip on one end for conveniently clipping it onto the shield before reassembling the unit. Might solve a few problems.

Yes, that's why I used the word and why it was in italics! I have often thought about bonding that shield. It certainly can't hurt. Who knows, theoretically it might even act like a dish reflector for the GPS signal.