My Disco

[kibi]

Hello everyone.
There are plenty of disco build logs around, so I';ll spare everyone the boredom and just highlight the differences that I made with my build. The main difference to my build that I have not seen anywhere else before are centred around the core.
The TBS Core did not really blow my hair back and the price made me run even further away. I wanted to use my iOSD mini anyway, so a custom power supply was required.
My parts bin was host to a 6W, 12V - 12V isolated DC - DC converter, it actually supports input voltages of between 9V and 18V which makes it perfect for 3S or 4S use. It will provide a nice clean, isolated and stable 12V which can then power the FPV circuits.

Traco Power DC - DC converter.





Here I have mounted the DC - DC converter upside down with VHB tape and soldered the wires to it "dead-bug" style. The wires then connect to the relevant pads that the TBS core would normally use. I have done the same with the wires coming out of the iOSD mini.
I can now connect my camera and VTx with the TBS supplied cables and the iOSD will work as normal but the wiring will be much neater.
I';ll have to wait and see if I can be bothered to make a can to cover the whole lot and isolate it.





More to follow as I progress the build.

[Marty McFly]

hi where did you purchase the traco please,Marty

[kibi]

Hi Marty

I think I got mine from Mouser a while back, but RS are apparently doing them discounted at the moment.

This is the one that I used.
http://uk.rs-online.com/web/p/isolated-dc-dc-converters/7553576/?searchTerm=ten6-1212&relevancy-data=636F3D3226696E3D4931384E4B6E6F776E41734D504E266C753D656E266D6D3D6D61746368616C6C7061727469616C26706D3D5E5B5C772D5C2E2F252C5D2B2426706F3D313326736E3D592673743D4D414E5F504152545F4E554D424552267573743D74656E362D313231322677633D424F544826

There are many converters available with higher or lower power ratings or voltages to suit your needs / budget.

[Northernmonkey]

Good thread! Any progress?

[kibi]

Good thread! Any progress?

Yes, I have made some progress over the last weekend.

I have installed a Frsky D8R-II plus Rx on the top plate.
I experimented with CPPM but I did not like what I saw on the oscilloscope, so I decided to stick with traditional PWM. It';s not all that bad really. There is a firmware available to widen the CPPM signal but it is not directly aimed at my Rx, so I thought I wouldn';t mess about.



Now, I want to take advantage of the telemetry facility that is included with Frsky system. I planned to use the two analog ports to gather Voltage and Amps information to be displayed on the Transmitter.
TBS did include current measurement circuit on the bottom board of their kit. I thought this was very kind of them to do that, but then I realised that it was to interface with their Core, which I am not using.
So, no big deal I';ll get a current measurement for myself and use that for telemetry. The chip they have used is an INA139 current shunt monitor made by Texas Instruments. You can see it here.



I quickly knocked a buffer together using a rail-to-rail op-amp since the input impedance of the ADC inside the Frsky Rx seemed a bit low for my liking. It is currently glued to the side of the leg.



The multi turn pot is to adjust the voltage sent to the Frsky input accurately. The Frsky ADC has a maximum voltage input of 3.3V, so the pot is wired as an adjustable voltage divider.

I looked up the datasheet for the INA139 to figure out how it works. The datasheet revealed that in it';s current configuration it would not work very well. Experimentation proved this. The problem is that the INA139 does not output a very linear voltage in relation to current until it has about 50mV on it';s sense input.
TBS have put a 1mΩ resistor as the shunt. This is a good choice because at 50A, the power dissipation is only 2.5W, the lower the better. Unfortunately the lower the shunt resistor the lower the voltage drop for the INA139 to measure. Now this means that the INA139 only starts responding in a truly linear fashion at about 50A. This is not much use at lower currents. My experiments revealed that it just isn';t possible to get an accurate reading in the low current ranges.
I';ll have to get rid of the INA139 at some point and use that lovely rail-to-rail opamp that I mentioned earlier to build a differential current sense amplifier or look at some of Linear Technologies'; precision current shunt amplifiers. The LT parts are very expensive though, so TBS would wisely not choose one of their parts in order to keep costs down.

I got on the milling machine and quickly knocked up an antenna mount out of some scrap nylon. This mount ensures that the two antennae out of the Frsky Rx are in 90º planes to one another.





That';s about it for now. I might not be too fussed with the inaccuracies of the current measurements, I';ll have to wait and see to what extent it bothers me. It does at least, for now, give an indication. The voltage readings that I am getting back are perfectly accurate enough to set alarm levels etc. on the transmitter, which I guess is more important than current measurements.