Originally Posted by miru

Maybe we should take the FSAV input out? There are other checks in place now that can detect a 'dead' link.

That would save a couple of bytes. Since rev018 it works fine on Turnigy 9x without the cable, wonder how it works on other RXs.

Originally Posted by pawelsky

That would save a couple of bytes. Since rev018 it works fine on Turnigy 9x without the cable, wonder how it works on other RXs.

You could do a poll.....Works fine with Orange RX

Drone 2 with MiruMod using Mini Pro works great.

I decided to build one using the Nano. Acquired a couple of Nano boards (version 3.1) from Gravitech about a month ago. I first breadboarded the entire circuit including RC receiver and a cable I have used before to connect to the Drone 2 (just as I did with the Mini Pro version before wiring it for use in the Drone). Bench test using USB serial console (not connected to Drone yet) went fine. I then disconnected the USB cable and setup jumper and plugged in the cable to the Drone 2. Powered up the Drone and all went well until right after the orange blinks. The mod completely ignored any commands from the RC transmitter. I finally isolated the issue as being a problem with the serial receive path back from the Drone. The upload of the at2so.arm from the mod to the Drone did happen so the mod to Drone TX path is fine.

I checked the signals with my O-scope. Mod to Drone is fine (both sides of the level shifter). The Drone is sending to the level shifter but the Rx in on the Nano appears to be pulled high (thus no Receive path).

Swapped to my other Nano board. Identical results. Swapped to the Mini Pro (on the breadboard) -- works perfectly. (I can easily swap as the connected/active pins for the mod are in identical locations for both the Nano and the MIni Pro. A swap takes seconds. I am quite certain I do not have a short in any of my wiring that would be responsible for pulling the Rx input on the Nano high.

So, unless I did something stupid (possible but not likely in this case), there is likely something wrong with BOTH of the Nano boards I recently acquired.

I did some searching on the Web and have not heard of any issues with the Nano 3.1 (vs. the 3.0).

I have already looked at both Nano boards under magnification - did not see anything unusual. I will attempt some experiments with the Nano (have to write or look for some code to use the serial port to test with).

Anyone have similar results or any other clue as to why this may be happening? I suppose it is possible I got two boards with issues in the same order. I do not want to go buy more until I am certain.

Thanks in advance.

George

Originally Posted by gtsullivan

Drone 2 with MiruMod using Mini Pro works great.

I decided to build one using the Nano. Acquired a couple of Nano boards (version 3.1) from Gravitech about a month ago. I first breadboarded the entire circuit including RC receiver and a cable I have used before to connect to the Drone 2 (just as I did with the Mini Pro version before wiring it for use in the Drone). Bench test using USB serial console (not connected to Drone yet) went fine. I then disconnected the USB cable and setup jumper and plugged in the cable to the Drone 2. Powered up the Drone and all went well until right after the orange blinks. The mod completely ignored any commands from the RC transmitter. I finally isolated the issue as being a problem with the serial receive path back from the Drone. The upload of the at2so.arm from the mod to the Drone did happen so the mod to Drone TX path is fine.

I checked the signals with my O-scope. Mod to Drone is fine (both sides of the level shifter). The Drone is sending to the level shifter but the Rx in on the Nano appears to be pulled high (thus no Receive path).

Swapped to my other Nano board. Identical results. Swapped to the Mini Pro (on the breadboard) -- works perfectly. (I can easily swap as the connected/active pins for the mod are in identical locations for both the Nano and the MIni Pro. A swap takes seconds. I am quite certain I do not have a short in any of my wiring that would be responsible for pulling the Rx input on the Nano high.

So, unless I did something stupid (possible but not likely in this case), there is likely something wrong with BOTH of the Nano boards I recently acquired.

I did some searching on the Web and have not heard of any issues with the Nano 3.1 (vs. the 3.0).

I have already looked at both Nano boards under magnification - did not see anything unusual. I will attempt some experiments with the Nano (have to write or look for some code to use the serial port to test with).

Anyone have similar results or any other clue as to why this may be happening? I suppose it is possible I got two boards with issues in the same order. I do not want to go buy more until I am certain.

Thanks in advance.

George

If you can download the sketch the RX pin should be ok, since the IDE verifies the download. On a Nano the serial interface is shared with the FTDI circuitry, which is connected through 1K resistors to the RX and TX channel of the processor (maybe they have a lower value resistor?).
Then you can check the wiring of the level shifter, make sure the drone is ALWAYS on the LV side and the Arduino is on the HV side, no matter of the direction of the signal, the drone side needs to be connected to the source pin of the FET.

Miru,

I checked and double checked the level shifter. A Pro Mini works in the identical circuit on the breadboard - the level shifter is wired correctly. I have done a quick experiment with a sketch running loop-back serial where I connected a 5 Volt Sparkfun FT232R directly to the RX and TX on the Nano and the Nano is powered externally with 5 Volts (No USB cable connected to the Nano). Same problem. Here there is no level shifter to add doubt. The RX input line is exhibiting "strong" pull-up behavior. The Nano was ONLY connected to power and the RX/TX lines (not on my MiruMod breadboard). The same test sketch runs perfectly on the Pro Mini connected to the FT232R and the same external power supply.

When I disconnect the RX line from the Nano (mounted on the MiruMod breadboard and connected to the Drone via cable) and monitor the output from the level shifter HV side, I see serial data at the appropriate voltage levels and it decodes at 38,400 baud. Further confirmation the level shifter is fine. I remain convinced there is a problem with the Nano (two of them!).

I wrote to Gravitech with extensive testing details. I hope to hear from them if they have seen the problem before.

Thanks for the input.

George

Originally Posted by gtsullivan

Drone 2 with MiruMod using Mini Pro works great.

I decided to build one using the Nano. Acquired a couple of Nano boards (version 3.1) from Gravitech about a month ago. I first breadboarded the entire circuit including RC receiver and a cable I have used before to connect to the Drone 2 (just as I did with the Mini Pro version before wiring it for use in the Drone). Bench test using USB serial console (not connected to Drone yet) went fine. I then disconnected the USB cable and setup jumper and plugged in the cable to the Drone 2. Powered up the Drone and all went well until right after the orange blinks. The mod completely ignored any commands from the RC transmitter. I finally isolated the issue as being a problem with the serial receive path back from the Drone. The upload of the at2so.arm from the mod to the Drone did happen so the mod to Drone TX path is fine.

I checked the signals with my O-scope. Mod to Drone is fine (both sides of the level shifter). The Drone is sending to the level shifter but the Rx in on the Nano appears to be pulled high (thus no Receive path).

Swapped to my other Nano board. Identical results. Swapped to the Mini Pro (on the breadboard) -- works perfectly. (I can easily swap as the connected/active pins for the mod are in identical locations for both the Nano and the MIni Pro. A swap takes seconds. I am quite certain I do not have a short in any of my wiring that would be responsible for pulling the Rx input on the Nano high.

So, unless I did something stupid (possible but not likely in this case), there is likely something wrong with BOTH of the Nano boards I recently acquired.

I did some searching on the Web and have not heard of any issues with the Nano 3.1 (vs. the 3.0).

I have already looked at both Nano boards under magnification - did not see anything unusual. I will attempt some experiments with the Nano (have to write or look for some code to use the serial port to test with).

Anyone have similar results or any other clue as to why this may be happening? I suppose it is possible I got two boards with issues in the same order. I do not want to go buy more until I am certain.

Thanks in advance.

George

You may want to try disconnecting shifter-arduino line only (keeping 3 other connections in place) and repeating your experiment with two channel scope watching both sides of shifter. Does the shifter output (HV side, disconnected one) now follows the input(LV side) or does it stay always high even if the LV is at zero? If it does follow, you got a problem with nano, and if it does not, the problem is with shifter.

Edit: I can see that you've already done the test and nano is to blame...

Should have said I checked both sides of the level shifter. My scope is quad input with 16 channels of logic decoding. The drone side RX and TX (LV) track fine. No matter what state the Drone to Nano LV side is in, when connected to the Nano RX on the HV side, RX is always high (with next to zero noise on the line) on the Nano. When the RX line is disconnected from the Nano, the HV signal tracks the LV.

On the LV side, the swing is about 1.7 volts and the HV side is just under 5 if I remember correctly (forgot to write it down). Looking at the Nano TX line, there is a very, very small amount of noise (insignificant to the standards). On the RX pin on the Nano, there is almost none (thus why I said "strong" pull-up).

Thanks again,
George

Originally Posted by gtsullivan

Should have said I checked both sides of the level shifter. My scope is quad input with 16 channels of logic decoding. The drone side RX and TX (LV) track fine. No matter what state the Drone to Nano LV side is in, when connected to the Nano RX on the HV side, RX is always high (with next to zero noise on the line) on the Nano. When the RX line is disconnected from the Nano, the HV signal tracks the LV.

On the LV side, the swing is about 1.7 volts and the HV side is just under 5 if I remember correctly (forgot to write it down). Looking at the Nano TX line, there is a very, very small amount of noise (insignificant to the standards). On the RX pin on the Nano, there is almost none (thus why I said "strong" pull-up).

Thanks again,
George

Check the resistor on the Nano that connects the RX line to the FTDI portion, maybe they put a 100 Ohm in there instead of the 1K as called out in the schematics.

Resistor pack RP1 has 102 marked on it. 102 == 1K as it should be. Right part number but not necessarily the right part.

Thanks,
George

has anyone experienced this problem?

using the AR6115e Spektrum receiver, about 50% of the time when you switch the drone on, at the point it loads the code to the drone and normally the motor lights flash red green to indicate its ready, the arduino red led will stay solid and wont complete. I have to power the drone on/off a few times then it works.

Very strange, using an arduino nano, tried two nanos and two receivers of the same type, any ideas?

Originally Posted by SJW2511

has anyone experienced this problem?

using the AR6115e Spektrum receiver, about 50% of the time when you switch the drone on, at the point it loads the code to the drone and normally the motor lights flash red green to indicate its ready, the arduino red led will stay solid and wont complete. I have to power the drone on/off a few times then it works.

Very strange, using an arduino nano, tried two nanos and two receivers of the same type, any ideas?

Sounds like it is waiting for the start character, check the Drone-TX -> Arduino-RX wires.

cheked them all, and remade the whole thing twice, same thing on both, weird thing it works, but about 50% of the time

on my AR600 based unit its fine all the time, i didhowever use thinner grade wires in these ones as they were more flexible, cant see it being that though

works 100% fine when checking in hyperterminal

Originally Posted by SJW2511

cheked them all, and remade the whole thing twice, same thing on both, weird thing it works, but about 50% of the time

on my AR600 based unit its fine all the time, i didhowever use thinner grade wires in these ones as they were more flexible, cant see it being that though

works 100% fine when checking in hyperterminal

Had this problem, and it was the RX line (problem with soldering, hard to see, had to dismantle half of my setup). The other problem could be the plug (had that once as well)

The RX line from the drone plug to LLC?

what was the rpbolem with the plug? all the solders seem ok, and well it just plugs in and out fine lol

Originally Posted by SJW2511

The RX line from the drone plug to LLC?

Anywhere on the line from Arduino's TX to Drone's RX. In my cast this was between levelshifter and the plug (which are on one board, that I've custom made), but could be anywhere on the line.