Hi Everyone,
I thought it might be useful to give a bit of background on what this forum is for.
3DRobotics have just released a set of telemetry radios based on the Si1000 chipset. The ArduPilot developers (particularly Mike Smith and myself) have been working hard on the firmware for these radios for months now, trying to make them ideal for APM telemetry.
These radios are unusual as they are open source and extremely configurable. That means that there is a lot of scope for diydrones members to get involved in tuning of the radios, and in adding new functionality. That is what this forum is for.
The Si1000 chipset has a huge number of configuration registers. We've spent quite a lot of time selecting the best register values for use with APM, but there is still a lot of scope for more experimentation. For example, I'm planning to do a test flight soon to see if enabling manchester encoding gives better or worse range at various data rates.
Similarly, there is a lot of scope for adding firmware features to make these radios link really well with antenna trackers and other radio add ons. We've already got them talking to APM in a much more advanced way than an Xbee can, but there is a lot of scope for taking it even further.
The basic theme is that these are DIY radios. We've put together functionality to make them work really well already, but the real potential is from the community getting involved to make them even better.
So get involved, and let's bring to telemetry radios the same sort of community development that has made ArduPilot so much fun to be involved with!
Cheers, Tridge

Hi Jake,
The radios have been produced in both DIP and SMD versions. I think the first ones are DIP, and later ones will be SMD. In my testing I haven't seen any difference in the performance between the two varients.
It also seems like the actual data rate is going to be limited by the 115.2kbps UART rate listed in the HopeRF module datasheet. Is that all the faster it can go? The radio should be able to do 256 kbps and the UART can clock to 230.4 kbps. Have you played with this at all?
Once you add in the ECC the practical serial speed is limited to 115200 anyway. The highest I've been able to get the radios to do over the air is 250kbps. In theory they can do 256, but I've been unable to find a set of register values that allow it to actually do that (the suggested values from Si don't work). That is why the top air rate in the firmware is 250. I also don't suggest running them at 250, as the range is much lower at that air rate. I made the default 64 kbps as that gives great range, while still giving good telemetry logs from APM.
I don't think these radios are really going to be much good for sending images. For that you need a quite different radio design. These are great telemetry radios, but for images I use 5.8GHz and a very different design (I use Ubiquity bullets).
Cheers, Tridge
Permalink Reply by Veikko Vierola on April 10, 2012 at 2:12pm A good stable 433MHz CE certified telemetry radio is something I'd like to have. Hopefully 3DR radio will get to that goal first. By the way, could RFDesign do the certification lab tests officially?

Hopefully 3DR radio will get to that goal first. By the way, could RFDesign do the certification lab tests officially?
I don't know. I'm hoping that other people will work out what is needed. I'd prefer to work on the code myself :-)
I'll point Seppo at this forum in case he wants to comment.
Cheers, Tridge
Permalink Reply by Jake Stew on April 10, 2012 at 5:05pm > I don't think these radios are really going to be much good for sending images.
There's been a lot of work on getting a Gumstix computer in the air. It has a lot of DSP capability and could easily perform the compression required. A 640x480 image compressed with almost no quality lost would be about 70k. At 128kbps that could be transferred in 4 seconds. With higher compression or lower resolution you could get your 1 fps even at 128 kbps.
Taking snapshots manually would be a very handy feature. If you're flying IFR and just want a picture every minute or so to make sure you're not heading into something you could get by with not that much bandwidth.

One thing I forgot to mention... I would have increased the frequency a bit.
Everyone and their brother is cranking out 433 and 915 mhz units. All of them that I've seen use these frequencies and/or just copy the reference design or use a module that copied the reference design.
Unless I've misunderstood your suggestion, that would take us out of the ISM bands, which means these radios would only be useful to people with HAM licenses. Is that what you meant?
Cheers, Tridge
Permalink Reply by Jake Stew on April 10, 2012 at 4:37pm No, I'm suggesting just not using the exact same frequencies as everyone else. i.e. use 918 instead of 915 or 434 instead of 433. It's still in the same band, just not on the "default" frequency that every cheap chip radio is using.

Hi Jake,
No, I'm suggesting just not using the exact same frequencies as everyone else. i.e. use 918 instead of 915 or 434 instead of 433. It's still in the same band, just not on the "default" frequency that every cheap chip radio is using.
The radios don't use a single frequency - they do frequency hopping over the entire allowed band. So we're actually using all of the ISM band in each case, which makes us much more resistant to interference.
In the 900 MHz band this is actually required. In the US you have to hop over 50 frequencies in order to use the full 1W EIRP limit. In Australia you have to hop over 20 frequencies.
We also do frequency offsetting within the band, based on the NETID. So if you use a different NETID from someone else then the set of 50 frequencies you use will be different from them. Both will cover the whole band, but will have different center frequencies for each channel.
Cheers, Tridge
Permalink Reply by Earl on April 23, 2012 at 6:50pm Got my 433mhz radios today 4/23/12. Looks good.
Hooked USB cable to laptop and with a terminal program talked to radio at 57kb fine.
Did an ATI5 and data returned as expected.
Hooked other radio to APM2 with supplied cable.
Both radios the green lights went solid meaning they are locked to each other.
I did an RTI5 and the APM2 radio responded as expected.
Did ATI7 and RTI7 and got expected data.
Great so far !!!!
Now at this point I expect the radios to send mavlink data from APM2 to the laptop.....NO GO ....
I even tried the mission planner. Did a control a and got both radios settings. Great !!
The problem seems to be the APM2 is not talking to the radio. Cable miswired ??? I don't think so. TX > RX and RX > TX
The APM2 telemetry data rate is set to 57kb. Verified by USB cable on APM2 to get setup data.
Sh....T. I think I know why the radio is not sending data. THE USB CABLE on the APM2 is still connected !
I will unplug and try again........
Earl
Permalink Reply by Earl on April 23, 2012 at 7:07pm Son of a gun !!!!! That was it. Units doing mavlink great at 57kb on 433mhz. Now to try range.
(it's ALWAYS the simple things !)
Earl

yep! The APM2 uses the same UART for USB and telemetry, so only one will connect at a time. It auto-switches between the two when you insert or remove the USB cable.
Cheers, Tridge

it's the opposite of having to ask somebody "did you plug it in", lol!
Permalink Reply by Gabriel on April 24, 2012 at 12:31am I got around to setting mine up today (900mhz pair on a APM1) Soldered on some gold plated right angle headers because i had them instead. The wife's hair dryer doesn't seem to be good enough to the the heat shrink but it got there. plugged it in changed the channels down to 20 (I assume that's all I have to do to make it semi-legal in Australia ??) and it works!! little quad icon in the mission planer with the gps and everything.
about firmware updates i assume i need to have the both on the same version ? and to update the air module I need a FTDI cable 3.3v or 5v ? (dyidrones store don't sell theirs separate) either http://littlebirdelectronics.com/products/ftdi-cable-5v-vcc33v-io or http://littlebirdelectronics.com/products/ftdi-cable-5v
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.249 members
183 members
693 members
24 members
51 members
© 2013 Created by Chris Anderson.
Powered by
