Espritmodel.com Telemetry Radio
Reply
Thread Tools
Old Mar 01, 2010, 01:55 AM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,669 Posts
Some emperical testing on the Two-way system

I did a bit of receiver stress testing today with a pal at work. We compared the Two-Way with Spektrum for range in a bad signal environment from inside our office block to outside.
FrSky looks good really. The test was a bit unfair though as the Spektrum had four satellite receivers working. The difference between the two was very little. We need to test against a comparable two receiver Spektrum unit to be a fair comparison.
PLMS is online now Find More Posts by PLMS
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 01, 2010, 11:39 PM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,669 Posts
FrSky Two-way system RF measurments

I tested the FrSky Two-way system for RF measurements on the bench at lunchtime today with a pal here at work. No big surprises really. A few observations.

I think Chase said the PRE-production (note the emphasis on pre) samples were expected to be lower power than the regular FrSky product.
Well testing showed that these are lower in RF channel power than Spektrum and others, but it showed that the difference was very little.
(Note - the overall transmission system gain is partly calculated from the spreading code and hopping technique used, not just RF channel power. So one must be careful when comparing two different SS systems on channel power alone regarding it's ultimate maximum range).

At the same time we actually noticed some new things about the newer and older firmware versions on Spektrum DSM2 RF modules (I'll explained later).

A couple of RF screen images from the tests are attached, from an Ansitsu MT8222A BTS base station analyzer.

They show the general transmission spectrum for the FrSky Two-way system.
One shows a frequency hopping transmission scheme, using a total of 36 hopping channels.
Also shown in another image is the channel power measurement on one of those 36 channels. The calculated channel power reading is on the lower RHS of the screen image. The peaks on the screen reveal very little about the real power to air BTW.
This BTS analyzer excels at these sort of measurements on spread spectrum radios, but even so it still has to sample for about 3 or 4 minutes to come up with the final power reading because of the channel hopping. We have also found that you must be very careful to match the transmission system to the instrument measurement settings, particularly the channel bandwidth, if you are to get meaningful results between brands of 2.4Ghz equipment.
One thing of note right away is that the FrSky channel bandwidth is very narrow compared to say Spektrum DSM2 DSSS. FrSky uses just under 100kHz (about 90Khz) where as DSM2 is about 1500kHz. However DSM2 does not spread the trasmitted signal over 36 channels, just two...

FrSky is unlikely to cause interference to DSM2 because of it's small bandwidth, and DSM2 is not going to interfere with 36 channels on FrSky either. That's exactly how it should be in spread spectrum radio.

We also measured the power decrease in the 'Range Test mode', and have some feedback to FrSky about that in comparison to the well know Spektrum brand system.
FrSky Range Test on the Two-way sample I was sent is spot on 15dB. That's not enough in our opinion...
In the past we found Spektrum DSM2 units to use a 20dB drop in power with the range test button pressed. To confirm it we tested the two newer Spektrum modules we have here to test with.
At that point we got a real surprise. Our Spektrum DSM2 module with firmware version V1.9 had a drop of exactly 30dB, and the second had firmware version V2.1 and it had a power drop of spot on 40dB !!
So Spektrum have been reducing the range test power in subsequent firmware revisions by the look of it. I wonder if anybody knew that (we didn't).
My friend here has a very late version module arriving soon, and we'll check that one to see if it's also 40dB. We also will recheck a very old one to see if they really are 20dB down (it was a while ago since we checked). We'll let you know.

The conclusion of that part was that we think FrSky should consider lowering the Range Test power output in the TX firmware to at least 30dB down, not 15dB. 15dB is not low enough to find installation issues in your model while doing a normal 50 paces walk-away test in our opinion. Given how Chase Wu seems to be listening to feedback I'm sure it will be looked at.

To wrap up.
The Two-way system uses Frequency Hopping SS (rather than DSSS) and has a power output close to 'normal' for a better word, probably within 2dB of Spektrum on the low side.
In Range Test mode it shows a reduction of exactly 15dB of output (in the sample I have at least).
No real surprises and nothing to grizzle about that effects flying it.

I want to do a real range test down my favorite test road soon, I'll take video of that and post it up.
Martin
PLMS is online now Find More Posts by PLMS
Last edited by PLMS; Mar 01, 2010 at 11:47 PM.
Reply With Quote
Old Mar 02, 2010, 02:30 AM
Registered User
Birmingham, UK
Joined Apr 2007
487 Posts
That 15dB drop explains why I got 150 yards + range on test with the aerial pointing pretty much at the model. It is too far for a meaningful range check, I would say you want 50 yards at most in good conditions if it's going to be worth doing, so you can see if anything changes between tests. Who in their right minds is going 150-200 yards every time?
BrumBob is online now Find More Posts by BrumBob
Reply With Quote
Old Mar 02, 2010, 02:46 AM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,669 Posts
Quote:
Originally Posted by BrumBob View Post
That 15dB drop explains why I got 150 yards + range on test with the aerial pointing pretty much at the model. It is too far for a meaningful range check, I would say you want 50 yards at most in good conditions if it's going to be worth doing, so you can see if anything changes between tests. Who in their right minds is going 150-200 yards every time?
Bob at least it's a dead easy fix in the production, just a small value change in the firmware. Hopefully FrSky are absorbing the constructive feedback we send them via this thread.

Speaking of firmware, wouldn't it be fantastic if FrSky could make their receivers 'firmware upgradable' like new Hitec gear is supposed to be. Mate, it would sure get lots of attention then, and I'd be buying lots more.
PLMS is online now Find More Posts by PLMS
Reply With Quote
Old Mar 02, 2010, 03:00 AM
Registered User
Joined Mar 2009
294 Posts
Based on our exact and practical test, that's exactly 20dB drop(by program's set), and the 1/10 practical test range is also be tested by us.

I mean the range(but not the power) of 2-way system will be a bit more short compare with V8 series products.

Meantime, all the specification(including the power) of our PRE-production samples as the same as regular products.

In order to walk shorter range under range check, maybe it should be 30dB drop

Best regards
Chase Wu
Chase Wu is offline Find More Posts by Chase Wu
Reply With Quote
Old Mar 02, 2010, 05:15 AM
Bleriot's R Us
xairflyer's Avatar
Ireland, Donegal
Joined Nov 2003
1,002 Posts
What about the transmission of data from the receiver to the transmitter, that is the bit interests me, how will that be done? will there be a display at the transmitter end.
xairflyer is offline Find More Posts by xairflyer
Reply With Quote
Old Mar 02, 2010, 05:37 AM
Registered User
Birmingham, UK
Joined Apr 2007
487 Posts
Quote:
Originally Posted by Chase Wu View Post
Based on our exact and practical test, that's exactly 20dB drop(by program's set), and the 1/10 practical test range is also be tested by us.
.....
.....
In order to walk shorter range under range check, maybe it should be 30dB drop

Best regards
Chase Wu
Chase,
30 - 50 yards (or metres) seems a reasonable target to aim for with the range check. For me, that's about the limit that I can see the contols moving if I'm on my own, and the limit for effective communication if I'm not.

The upgradeable firmware sounds like a good idea, if it's possible.

I'd also like to see the 4 channel receiver a couple of mm longer to allow some support for the aerial wire, I've had one break off. Even if it only allows a dab of glue to support the joint it would help.
BrumBob is online now Find More Posts by BrumBob
Reply With Quote
Old Mar 02, 2010, 05:59 AM
Registered User
Romania, Dolj, Craiova
Joined Sep 2007
14,272 Posts
Quote:
Originally Posted by xairflyer View Post
What about the transmission of data from the receiver to the transmitter, that is the bit interests me, how will that be done? will there be a display at the transmitter end.
I am planning to write such software for a PDA / PPC platform first.
These device are now omnipresent either as GPS navigators or GSM phones, and small enough to be comfortably fitted on any radio.
renatoa is offline Find More Posts by renatoa
Reply With Quote
Old Mar 02, 2010, 08:20 AM
Registered User
Joined Mar 2009
294 Posts
Quote:
Originally Posted by BrumBob View Post
Chase,
30 - 50 yards (or metres) seems a reasonable target to aim for with the range check. For me, that's about the limit that I can see the contols moving if I'm on my own, and the limit for effective communication if I'm not.

The upgradeable firmware sounds like a good idea, if it's possible.

I'd also like to see the 4 channel receiver a couple of mm longer to allow some support for the aerial wire, I've had one break off. Even if it only allows a dab of glue to support the joint it would help.
Thanks a lot Bob,

You are heads-up, We get it and will make some improvement in next batch of 4ch rx. As to the residual 4ch rx of this match, we will take steps to fasten the antenna.

Chase
Chase Wu is offline Find More Posts by Chase Wu
Reply With Quote
Old Mar 02, 2010, 08:26 AM
Registered User
Joined Mar 2009
294 Posts
Quote:
Originally Posted by xairflyer View Post
What about the transmission of data from the receiver to the transmitter, that is the bit interests me, how will that be done? will there be a display at the transmitter end.
Hi xairflyer,

You could find another thread about our two way system, the link as below:
http://www.rcgroups.com/forums/showthread.php?t=1200443

You also could download some docus from our website.
http://www.frsky-rc.com/download.html

Thanks
Chase Wu
Chase Wu is offline Find More Posts by Chase Wu
Reply With Quote
Old Mar 02, 2010, 04:36 PM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,669 Posts
Quote:
Originally Posted by xairflyer View Post
What about the transmission of data from the receiver to the transmitter, that is the bit interests me, how will that be done? will there be a display at the transmitter end.
FrSky have given us the system info to go make something like this. It could be a simple DIY project for us. The parts required would be a 16x2 LCD, a programmed PIC micro and a small GPS with NMEA serial output (in the model).

The LCD with PIC would plug directly into the back of the Two-Way module on the transmitter (the module has a 5V power pin for use as well as the serial output).
Here's one way the LCD screen could look. This would suit a thermal glider pilot perhaps.
Martin
PLMS is online now Find More Posts by PLMS
Reply With Quote
Old Mar 02, 2010, 05:37 PM
Bleriot's R Us
xairflyer's Avatar
Ireland, Donegal
Joined Nov 2003
1,002 Posts
Battery condition would be worth it's weight in gold, also fuel capacity for Jets would be very useful also as would the speed
xairflyer is offline Find More Posts by xairflyer
Reply With Quote
Old Mar 03, 2010, 02:02 PM
Registered User
krafty's Avatar
France, Aquitaine, Duras
Joined Jan 2006
322 Posts
Frsky in the cold

Posting deleted
krafty is offline Find More Posts by krafty
Last edited by krafty; Mar 03, 2010 at 02:18 PM. Reason: Aready covered in subsequent posts
Reply With Quote
Old Mar 03, 2010, 04:42 PM
Novice builder and flyer!
brisbane australia
Joined Mar 2007
320 Posts
Quote:
Originally Posted by krafty View Post
Posting deleted
Last edited by krafty; Today at 06:18 AM. Reason: Aready covered in subsequent posts
I'm confused - did you mean previous posts?
gmphoto is offline Find More Posts by gmphoto
Reply With Quote
Old Mar 03, 2010, 11:01 PM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,669 Posts
Idea for discussion

While FrSky are seeking ideas and feedback lets make use of it. Here is an idea for your comment or improvements.
It's an extra item to be sold with the Two-Way receivers, to make it more user friendly.
It would be a small circuit board with a handful of components to attenuate common battery voltages so the AD inputs can use them for alarms etc.
You would use the links to select the divider to suit your pack. To keep it simple there is two options for each of the two AD inputs.

There is a JPG and also an Eagle file. If you use Eagles free circuit editor you can modify the idea and repost it.

The thing I reckon is to not make it too complex or too risky (blow up the RX inputs).
CH1 could be either a RX NiMh pack (4 or 5 cell), or up to a 3S LiPo pack.
CH2 could be up to a 6S pack or up to an 8S pack.

Connection between the RX and the extra attenuator board could be easily done by either common M-F servo extension cables, or by M-M extension cables too. The board just needs the right sex headers on that side.
PLMS is online now Find More Posts by PLMS
Last edited by PLMS; Mar 03, 2010 at 11:13 PM.
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Mini-Review The video for FrSky 2.4Ghz Chase Wu Radios 0 Dec 16, 2009 01:18 AM
Discussion Realflight G5 Combat so far so good! splitpilot Simulators 0 Nov 22, 2009 06:36 AM
Discussion So far... So good rikks Mini Helis 35 Feb 17, 2006 12:42 PM
Aerofly Deluxe so far so good! Gary Morris Simulators 5 Aug 19, 2005 07:23 AM
Schulze 18.46K v7.02 - so far, so good! Jason M Electric Heli Talk 4 Aug 11, 2002 01:25 AM