HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Jul 12, 2012, 03:48 AM
Registered User
Romania, Dolj, Craiova
Joined Sep 2007
15,410 Posts
Ah, wasn't aware about differences.
In this case, the universal solution, it's outside, not internal: the trainer port.
It absolutely carries PPM !
renatoa is offline Find More Posts by renatoa
Reply With Quote
Sign up now
to remove ads between posts
Old Jul 12, 2012, 03:59 AM
Registered User
Joined Dec 2007
25 Posts
Ok. But I don't understand- I must insert trainer cable to get PPM signal. If I insert in to port simple only plug (without cable) PPM is not generated . Maybe I need to add something - like resistor or something?
kliunkis is offline Find More Posts by kliunkis
Reply With Quote
Old Jul 12, 2012, 08:42 AM
Rocket geek
Highlands Ranch, CO
Joined Dec 2007
651 Posts
Quote:
Originally Posted by kliunkis View Post
Ok. But I don't understand- I must insert trainer cable to get PPM signal. If I insert in to port simple only plug (without cable) PPM is not generated . Maybe I need to add something - like resistor or something?
The DX8 is looking for a 3-conductor plug with 2 of the conductors connected to each other before it will output PPM. Sorry I forget which ones. I've gotten as far as looking at the PPM coming out of the DX8 trainer port with a logic analyzer, but I got distracted before I got the full path working through the DHT-U.
Adrian A is offline Find More Posts by Adrian A
Reply With Quote
Old Jul 12, 2012, 10:02 AM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,692 Posts
Quote:
Originally Posted by jj604 View Post
So would wrapping the barometric sensor completely in Al foil work?
Just a thought. I have seen more fluctuation in the height readout since I shortened the wire and that now makes sense.
John
Hi John, I tried that approach today. Nope.
In fact I went further than that, I made a brass foil cover for the baro board and soldered it on to the bottom ground plane. Did nothing to help at all (and I was surprised by that).

I then tried feeding the data wires through a toroid choke (like used for ESC's). That had a little success, particularly with the biggest one I had (25mm and about ten grams !). It reduced the variation by about 80%. Still around ten meters variation though...
I also tried four tubular bead chokes, one on each wire near the sensor board. That did zero.
I then tried small ceramic caps on the data lines (yellow/brown wires) to ground at the module. Didn't help (and going past 100pf corrupted the data anyway).

So it looks like RF is getting in along the cables, but it's not easy to filter.

As it's fitted to a Quadcopter and all packed into the center (hence I also shortened the wires) it's hard to get the baro sensor away from the RX (on the same tray).

I have two choices unless somebody thinks of another. Fit longer wires to the baro and move it out on the arms (in the turbulent air !) or buy longer antennas and run them out under two of the arms as far out as possible. Or thirdly try another sensor just in case this one is odd.
PLMS is offline Find More Posts by PLMS
Reply With Quote
Old Jul 12, 2012, 10:16 AM
ancora imparo
jj604's Avatar
Melbourne, Australia
Joined Jul 2005
6,865 Posts
Thanks, mate. That saved me a lot of work!

I'm wondering if it is partly poor board design - particularly the hub. Maybe Renatoa or one of the RF gurus might have a view?

John
Quote:
Originally Posted by PLMS View Post
Hi John, I tried that approach today. Nope.
In fact I went further than that, I made a brass foil cover for the baro board and soldered it on to the bottom ground plane. Did nothing to help at all (and I was surprised by that).

I then tried feeding the data wires through a toroid choke (like used for ESC's). That had a little success, particularly with the biggest one I had (25mm and about ten grams !). It reduced the variation by about 80%. Still around ten meters variation though...
I also tried four tubular bead chokes, one on each wire near the sensor board. That did zero.
I then tried small ceramic caps on the data lines (yellow/brown wires) to ground at the module. Didn't help (and going past 100pf corrupted the data anyway).

So it looks like RF is getting in along the cables, but it's not easy to filter.

As it's fitted to a Quadcopter and all packed into the center (hence I also shortened the wires) it's hard to get the baro sensor away from the RX (on the same tray).

I have two choices unless somebody thinks of another. Fit longer wires to the baro and move it out on the arms (in the turbulent air !) or buy longer antennas and run them out under two of the arms as far out as possible. Or thirdly try another sensor just in case this one is odd.
jj604 is online now Find More Posts by jj604
Reply With Quote
Old Jul 12, 2012, 10:21 AM
Registered User
(Jo)Hannes's Avatar
Trier, Germany
Joined Aug 2007
522 Posts
Mmmmmh very Strange, the Sensor is connected to the hub via i2c -> digital, so influences this will not "change" the read values on the cable (they just might block the connection). The only thing i can immagine is that RF Noise on the power-lines influence the sensor itself while reading the preasure and converting it from analog to digital....
Maybe a Cap between VCC and GNC near the Sensor might help.
(Jo)Hannes is online now Find More Posts by (Jo)Hannes
Reply With Quote
Old Jul 12, 2012, 10:36 AM
Martin - AKA mr.sneezy
PLMS's Avatar
Adelaide, Australia
Joined May 2004
1,692 Posts
Quote:
Originally Posted by (Jo)Hannes View Post
Mmmmmh very Strange, the Sensor is connected to the hub via i2c -> digital, so influences this will not "change" the read values on the cable (they just might block the connection). The only thing i can immagine is that RF Noise on the power-lines influence the sensor itself while reading the preasure and converting it from analog to digital....
Maybe a Cap between VCC and GNC near the Sensor might help.
@John
"Maybe Renatoa or one of the RF gurus might have a view?"
LOL

@Jo.
I tried that too Jo, caps from 5V to Gnd at the sensor board. I agree the variation can only be on the analogue side of the sensor, the data lines are not being effected, but they can carry RFI back to the analogue core of the BMP085, as there is nothing in the way (direct connected to sensor).

Microware RF is a black art, and I'm still learning after 28 years of it. One thing I have learned is an open mind and patience.

PS. One more thing I could try but not hopeful is fit shield cable instead of the plain twisted wire. In the meantime I just ordered 60cm antennas from Hobbyking.
PLMS is offline Find More Posts by PLMS
Reply With Quote
Old Jul 12, 2012, 11:23 AM
Registered User
Romania, Dolj, Craiova
Joined Sep 2007
15,410 Posts
My "guru" view is... cross-eyed

If BMP085 is influenced by the telemetry RF field, I wonder why my vario implementation of PLMS project is not affected...
As a reminder, it's a PCB side-board having the baro sensor placed at some mm near antennas. Picture below

renatoa is offline Find More Posts by renatoa
Reply With Quote
Old Jul 12, 2012, 05:19 PM
Registered User
Takilara's Avatar
Norway, Rogaland, Stavanger
Joined May 2008
302 Posts
Quote:
Originally Posted by (Jo)Hannes View Post
Mmmmmh very Strange, the Sensor is connected to the hub via i2c -> digital, so influences this will not "change" the read values on the cable (they just might block the connection). The only thing i can immagine is that RF Noise on the power-lines influence the sensor itself while reading the preasure and converting it from analog to digital....
Maybe a Cap between VCC and GNC near the Sensor might help.
Could the issue be casued by problems on the downlink side? like bugs in byte stuffing/de-stuffing or overflows? Corrupt bytes? Have you logged the raw stream when this happens?

I don't often see corrupt bytes in the stream, but they do happen...

Also, could it be related to temperature?
Takilara is offline Find More Posts by Takilara
Reply With Quote
Old Jul 13, 2012, 03:38 AM
Registered User
Germany, NRW, Bonn
Joined Jul 2010
225 Posts
Quote:
Originally Posted by Takilara View Post
Could the issue be casued by problems on the downlink side? like bugs in byte stuffing/de-stuffing or overflows? Corrupt bytes? Have you logged the raw stream when this happens?

I don't often see corrupt bytes in the stream, but they do happen...

Also, could it be related to temperature?
Hi,
during development of the FRS_logger app (www.hapeweb.com) I tested some sensors extensively.
Indeed the BMP085 sensor produced extreme height variations when very close to the receiver (some centimeter). But when the sensor:

- was encapsulated by a foil against wind effects
- was positioned away from receiver and hub
I got a precision of about 0.5m, by using different filtering techniques in the app software. In my own system I get a precision of about 0.3m, so itís not the sensor itself which causes the problems.
Regards
HaPe
HaPe is offline Find More Posts by HaPe
Reply With Quote
Old Jul 13, 2012, 04:27 AM
Happy FPV flyer
Kilrah's Avatar
United Arab Emirates, Dubai
Joined Sep 2006
4,898 Posts
Quote:
Originally Posted by RENATOA View Post
If BMP085 is influenced by the telemetry RF field, I wonder why my vario implementation of PLMS project is not affected...
I see you have a PIC there... so I guess you did some filtering of the data in there?
AFAIK the biggest problem of the frsky implemtation is that the hub only takes the sensor's data and sends it out without treatment, so no filtering to smooth out the noise and spikes...

Quote:
Originally Posted by Takilara View Post
Could the issue be casued by problems on the downlink side? like bugs in byte stuffing/de-stuffing or overflows? Corrupt bytes? Have you logged the raw stream when this happens?
That's higly unlikely. The whole stream is framed and CRC'd...
Kilrah is offline Find More Posts by Kilrah
Reply With Quote
Old Jul 13, 2012, 04:36 AM
Registered User
Romania, Dolj, Craiova
Joined Sep 2007
15,410 Posts
Quote:
Originally Posted by Kilrah View Post
I see you have a PIC there... so I guess you did some filtering of the data in there?

AFAIK the biggest problem of the frsky implemtation is that the hub only takes the sensor's data and sends it out without treatment, so no filtering to smooth out the noise and spikes...
...
For the first, the software is not mine but PLMS, and posted in his vario thread. The filtering is a moving average of the last 10 readings if I remember well.

Second, I doubt about sending sensor data without treatment in hub. At least the conversion from pressure to altitude is done.
renatoa is offline Find More Posts by renatoa
Reply With Quote
Old Jul 13, 2012, 09:18 AM
Registered User
billg's Avatar
USA, DE, Newark
Joined Sep 2001
127 Posts
Jumping altitude readings

Just a quick comment about the jumping altitude readings. I have not had time to analyze the this in any detail but I did power the hub with a 4 cell battery and the jumping went away. I looks like the jumping may be caused by switching regulator noise generated in my speed control. I may have some time in the next few weeks to try some filtering in the hub.
Bill
billg is offline Find More Posts by billg
Reply With Quote
Old Jul 13, 2012, 10:56 AM
Registered User
Joined Jan 2012
56 Posts
I'm confused about fitting the DHT-U to the Turnigy 9X. From what I see, the DHT-U comes with connectors on the end so no soldering is needed? Just connect it to the prongs sticking out of the back where the module used to plug in to?

http://www.hobbyking.com/hobbyking/s...dProduct=19966

Is that correct, or do you need to cut the plugs off and solder it inside? If you can simply plug the plugs into the old module area, how would people recommend securing them as it doesnt seem like the best solution as if they fall out during use then that would be bad!
totality is offline Find More Posts by totality
Reply With Quote
Old Jul 13, 2012, 11:36 AM
Registered User
United States, WI, Fond du Lac
Joined Sep 2008
1,740 Posts
Quote:
Originally Posted by totality View Post
I'm confused about fitting the DHT-U to the Turnigy 9X. From what I see, the DHT-U comes with connectors on the end so no soldering is needed? Just connect it to the prongs sticking out of the back where the module used to plug in to?

http://www.hobbyking.com/hobbyking/s...dProduct=19966

Is that correct, or do you need to cut the plugs off and solder it inside? If you can simply plug the plugs into the old module area, how would people recommend securing them as it doesnt seem like the best solution as if they fall out during use then that would be bad!
You can just connect to the module pins. You also can use an empty module to cover the leads and keep them from being accidentally pulled out like the JRPA615.
pmullen503 is offline Find More Posts by pmullen503
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion FrSky 2.4ghz - Telemetry protocol discussions PLMS Radios 1421 Nov 26, 2014 01:39 AM
Discussion FrSky 2.4Ghz - Telemetry PDA application Mybrains Radios 18 Jun 23, 2011 10:02 AM
Sold FrSky Telemetry 2.4Ghz for JR Module, 2 RXs & extra nizee Aircraft - General - Radio Equipment (FS/W) 1 May 07, 2011 03:28 PM
Wanted FrSky 2.4Ghz DIY Kit with Telemetry purcilas Aircraft - General - Radio Equipment (FS/W) 5 May 04, 2011 05:10 PM