So my Kingkong 90GT was getting boaring, I wanted something lighter, stonger and if possible smaller

I had a Furious PikoBLX clone lying around from something which i don't even remember why i bought it....
This was a great start for a new build!
http://www.banggood.com/CleanFlight-BetaFlight-Micro-F3-Flight-Controller-Built-in-PDB-Buzzer-20X20mm-For-FPV-Racing-p-1094615.html

This Frame popped up on banggood so I immediatly ordered it!
Its an 85mm frame with a huge prop guard (which i don't use), but the main plate is one of the smallest and lightest:
http://www.banggood.com/PC85-85mm-Carbon-Fiber-FPV-Racing-Frame-with-Camera-Mount-Propeller-Guard-p-1132526.html

I also took the Racerstar BR1103 10000kv motors combo with 4x6A escs, I didn't want a 4in1 to make the build lower profile, but i think i might get a 4in1 (if there will be) next time, there is the BR1103B with screws for the props, its a bit heavier and i'm not sure I like it more than pushing the props in, it seems enought on these power levels:
http://www.banggood.com/4X-Racerstar-Racing-Edition-1103-BR1103-10000KV-1-2S-Motor-Dark-Blue-4X-RS6A-V2-Blheli_S-ESC-Dshot-p-1117842.html

For VTX CAM combo i tool the TX02 which has a very decent camera and 200mw power:
http://www.banggood.com/Eachine-TX02-Super-Mini-AIO-5_8G-40CH-200mW-VTX-600TVL-14-Cmos-FPV-Camera-p-1088368.html

I used the Micro FRSky FD1000 which is very light, has 1000m range and FRSky telemetry (not sport),
Its has serious problems on <3m range, but after 3m it is perfect (binding is a pain, take distance while binding):
http://www.banggood.com/Mini-FrSky-8CH-LR1000-Receiver-Long-Range-Compatible-FrSky-X9D-Taranis-Taranis-Plus-XJT-p-1132192.html

I took these Racerstar 2035 quad props, they are light and very breakable, but if one prop break I break the opposite prop on it and have a 2 props prop and it works ok hahaha:
http://www.banggood.com/10-Pairs-Racerstar-2035-50mm-4-Blade-ABS-Propeller-1_5mm-Mounting-Hole-For-FPV-Racing-Frame-p-1127196.html

This micro buzzer is surprisingly stong:
http://www.banggood.com/5V-Buzzer-Alarm-Beeper-With-Cable-for-Eachine-QX70-QX90-QX95-NAZE32-F3-DIY-Micro-Brushed-FPV-Racer-p-1100338.html

For batteries i ordered 400mah 2S 35C but i think they are too small/weak... 
http://www.gearbest.com/multi-rotor-parts/pp_590449.html

Let the build commence:

The PickBLX board, notice it doesn't support Dshot on all motors... I found it the hard way, 
I have one esc on its way from banggood,
I later found someone who sold them and got one from him,
Replaced the ESC and the motors still didn't work, switch to multishot and viola   





Shortening the antenna and making it 1/4 2.4ghz wave at 31.25mm, the original is about 27mm which is 1/5 2.4g wave i think.




This is how it first looked:





This is how it looked after 3 batteries, i guess you put the camera mount on the top carbon plate and not alone....




This is how it looks today after fixing it last night:


DVR footage, soon....

27 mm is correct due to the propagation factor of the coaxial cable.

(16-Apr-2017, 02:43 PM)Mephi Wrote: 27 mm is correct due to the propagation factor of the coaxial cable.

It's more correct, at least!

Without knowing the dielectric constant of the insulator around the central wire, it's difficult to be completely correct, but when the original 'debate' about the length of the antennas on the XSR came up, I sat down and did some slightly better than 'back of a cigarette packet' calculations:

If we take the speed of light in millimeters/second as 299,792,458,000

and the centre frequency for 2.4GHz band as used by FrSky as 2.445GHz or 2,445,000,000 Hz

Then the length of one wavelength is 122.614502249mm and a 1/4 wave is 30.6536mm

Looking at the FrSky XSR antennas, I see the exposed part is 25.8mm long, implying a velocity factor of 0.84. Now, a single bit of wire sticking out of the end of a bit of coax isn't a transmission line any more, so the velocity factor of the coax itself is not the relevant factor, but the fact that the wire is insulated does have a small effect, as does the diameter of the wire. Hopefully, the FrSky engineers know what they are doing and the 0.84 factor is accurate.

Well, if it was 27 mm original i would go for that...

Probably the most uninformed video I've ever seen from the Rotor Riot guys.

"The longer the exposed wire is, the more gain"

Err...no stupid. That is NOT the way RF works.

For a dipole antenna, the simple calculation of dividing the speed of light by the frequency and taking 1/4 of that for a quarter wave assumes that the antenna itself is infinitely thin, which it isn't. The insulator surrounding the exposed wire also has an effect. Both of these things actually mean that the exposed part of the wire should be shorter than the simple calculation suggests. Also, using 2.4GHz as the frequency to calculate with is also incorrect as the antenna should be designed for the best response at the centre of the quite wide range of frequencies which the radio system uses. In the case of the FrSky system, this is 2.445GHz.

While these two gentlemen might be excellent FPV pilots, they obviously know nothing about RF and their highly unscientific test is, at best, misleading.

Forget the explanations,
Just look at the bottom line,
X4R with 31mm (or so) longer range, XSR 26mm (or so) shorter range,
Change the XSR to 31mm and you get longer range,
Would be nice to have the scientific explanation, but actual results are more important.

There are other factors in that 'test' which could easily account for the difference in range, the most important one being the positioning of the antenna with regards to the transmitter. The shadowing effect of a carbon frame (which blocks RF) would make a huge difference and just flying a bit faster, which tilts the quad forward more will change the effective range of the receiver. As will taking a slightly different path which changes the angle between the receiver and transmitter.

Either you can believe stuff because it agrees with your own personal bias or because you like the person presenting the information, or you can use facts to evaluate the information and decide if it has any value.

Testing antennas in a scientific way is hard and requires not only test equipment that is outside most people's reach but also somewhere to perform the test that completely eliminates any reflected signals and such a test chamber is way out of a hobbyist's reach.

Saying "never mind the theory, look at the results" is absurd when you can't account for all the factors that would influence the result.

This video is rubbish and you won't change my opinion about it by appealing to me to suspend rational thought.

Well I just learned a lot about receiver antennas...

Oh, also, nice micro! glad you got the camera sorted. The extra support on the antenna looks like a good idea too. If I ever get my micro back in the air I might have to try reinforcing it like that.

There's a second part to this puzzle which makes me dislike this very unscientific 'test' even more.

On FrSky's receivers that have telemetry, the antennas are not just passively listening, they are also used to transmit back to the radio to send telemetry data, so the antennas are used both to receive and to transmit (obviously, not at the same moment).

When an antenna is used to transmit, it is crucial that the impedance of the antenna is matched to the transmitter. A mismatch in impedance causes more of the signal that should have been radiated out into the air to be reflected back to the output of the transmitter.

The primary way that you adjust the impedance of the antenna is by adjusting its length. When the engineers at FrSky design a telemetry receiver, they will use detailed calculations and perform careful measurements to determine the optimal length of the antenna to ensure that everything is properly impedance matched.

Altering the length of the antenna will change the impedance match. If it changes the matching too dramatically, telemetry range will be drastically reduced and over time, the power being reflected back into the transmitter will physically damage the output.

So, all the fanboys who have been slavishly fiddling with the length of their XSR antennas based on this nonsense may actually be doing little more than reducing telemetry range and damaging their receivers!

Interesting. Apparently Ill keep repairing my chopped antennas to factory lengths

Nice build btw

(17-Apr-2017, 03:26 PM)unseen Wrote: There's a second part to this puzzle which makes me dislike this very unscientific 'test' even more.

On FrSky's receivers that have telemetry, the antennas are not just passively listening, they are also used to transmit back to the radio to send telemetry data, so the antennas are used both to receive and to transmit (obviously, not at the same moment).

When an antenna is used to transmit, it is crucial that the impedance of the antenna is matched to the transmitter. A mismatch in impedance causes more of the signal that should have been radiated out into the air to be reflected back to the output of the transmitter.

The primary way that you adjust the impedance of the antenna is by adjusting its length. When the engineers at FrSky design a telemetry receiver, they will use detailed calculations and perform careful measurements to determine the optimal length of the antenna to ensure that everything is properly impedance matched.

Altering the length of the antenna will change the impedance match. If it changes the matching too dramatically, telemetry range will be drastically reduced and over time, the power being reflected back into the transmitter will physically damage the output.

So, all the fanboys who have been slavishly fiddling with the length of their XSR antennas based on this nonsense may actually be doing little more than reducing telemetry range and damaging their receivers!

How do you explain the difference in antenna length on the X4R and the XSR and the actual degragation in preformance when comparing factory XSR to X4R, that is very noticeable without scientific tests.

Also, I wouldn't give the Frsky engineers so much credit haha, I remember reading on one thread that someone was in contact with them and in the end he got the response that that was the optimal length their engineers decided on but if he wants more prefomance he can lengthen the antena 

(18-Apr-2017, 10:08 PM)sevet Wrote: How do you explain the difference in antenna length on the X4R and the XSR and the actual degragation in preformance when comparing factory XSR to X4R, that is very noticeable without scientific tests.

Also, I wouldn't give the Frsky engineers so much credit haha, I remember reading on one thread that someone was in contact with them and in the end he got the response that that was the optimal length their engineers decided on but if he wants more prefomance he can lengthen the antena 

I can't completely explain the difference and there is a certain amount of trusting the manufacturer here. I'll admit that the difference in length is considerable compared to other receivers. Here are the antennas from four different receivers, of which three are FrSky.



What I will do, to try and shed some light on this, is to order some spare FrSky antennas and some u.fl sockets to plug them in to. If I can make a suitable adapter to SMA then I can connect one up to my spectrum analyser and plot the S11 curves for the antenna at both 25.8mm(XSR) and 33mm (D4R-II) and maybe a few other lengths in between those.

Then we would at least have some actual data to aid our speculation about why the XSR has shorter antennas.

(18-Apr-2017, 11:00 PM)unseen Wrote: I can't completely explain the difference and there is a certain amount of trusting the manufacturer here. I'll admit that the difference in length is considerable compared to other receivers. Here are the antennas from four different receivers, of which three are FrSky.



What I will do, to try and shed some light on this, is to order some spare FrSky antennas and some u.fl sockets to plug them in to. If I can make a suitable adapter to SMA then I can connect one up to my spectrum analyser and plot the S11 curves for the antenna at both 25.8mm(XSR) and 33mm (D4R-II) and maybe a few other lengths in between those.

Then we would at least have some actual data to aid our speculation about why the XSR has shorter antennas.

Didn't completely understand but it sounds like a great plan! 

(19-Apr-2017, 01:02 AM)sevet Wrote: Didn't completely understand but it sounds like a great plan! 

S-parameters are measurements that describe how well an antenna works at various frequencies. (See: http://www.antenna-theory.com/definition...meters.php)

I have the equipment to test and measure these parameters, but it uses SMA connectors for all the various connections, so I'll need to find a good way of adapating from the connector that's used on the removable FrSky antennas (often called u.fl or IPX) to an SMA connector so that I can connect the antenna under test to my test equipment.

Once I can do that, I can generate a bunch of graphs that look like this:



If I create such graphs for both the long and short antennas, we can see how they work at the different lengths. Maybe there's a surprise waiting for us in the results?