I have made a "Tilt and Roll Trainer Slave" by an Arduino NANO board and an accelerometer. It is plugged into the trainer port of my DX8 and is supposed to manipulate elevator and aileron by tilting this device a few degrees in tilt and roll directions respectively. The plan is to be able to control my RC planes in the winter by simply tilting this device in my right hand while the hand is located in a pocket. The signal out on my 3.5mm jack looks fine on an oscilloscope, running 4V pulses as from a Trainer radio. It worked fine in the lab. for a long time, but one day my DX8 failed, and the trainer function was "confused", the plane trottle was all the sudden influenced by the aileron signal for example. I have tested a DX6 as Slave, and my DX8 is still "confused". I wonder if I have destroyed a digital input on a microcontroller in my DX8, maybe 4V (through a 1k resistor in series) was too much? Would it be possible to replace any component, or do I have to purchase a new DX8? Any advice?
Spektrum Possible To Repair A Dx8 Trainer Input?
- Thread starter Aage
- Start date
Tony
Staff member
This is actually deeper than I have ever looked into a DX8, but I do have some ideas. First thing I would do is back up your models. Just put them on an SD card and save them to your computer. Then, I would get new firmware downloaded from Spektrum and reflash the Tx. I'm hoping this is just a corrupted file. If that doesn't work though, maybe find someone that has a used DX8 that has been dropped or whatever and use it for parts. Start with replacing that board that has the port on it and go from there.
Other than that, I'm not sure how to go about this. I'm not sure what voltage they run stock, but I would think it would be in the mv category. I'm just guessing here though.
Sounds like a cool little project though thats for sure.
Other than that, I'm not sure how to go about this. I'm not sure what voltage they run stock, but I would think it would be in the mv category. I'm just guessing here though.
Sounds like a cool little project though thats for sure.
murankar
Staff member
my suggestion is to find out if the trainer port is setup for 3 volts or 5 volts. Then work it from there. Try what tony suggested and then scratch our heads after that. I do work arduino but not like this. I am working a basic monitoring system with it.
Also do you have your pin outs correct?
Also do you have your pin outs correct?
Thank you for taking your time to comment!
I connected a DX6i as SLAVE to my DX8 as MASTER with a 3.5mm jack cable, and "sniffed" the tip of the jack with an oscilloscope while in trainer mode. The signal was normally high (around +1.6V) and the pulses in the PWM-sequence were -2V, i.e. a peak-to-peak level of 3.6V. However, I was a little bit surprised that the high level changed from time to time, indicating that there may be a capacitive coupling. The peak-to-peak level was quite constant.
My Ardiuno board outputs +4V as high level and 0V during the pulses, via a 1k resistor to protect the input pin of the radio. I hoped that was good enough, but should probably have taken the voltage lower with a voltage divider, and possibly inserted a capacitor in series.
I will try to find a SD card and test Tony's proposal.
I connected a DX6i as SLAVE to my DX8 as MASTER with a 3.5mm jack cable, and "sniffed" the tip of the jack with an oscilloscope while in trainer mode. The signal was normally high (around +1.6V) and the pulses in the PWM-sequence were -2V, i.e. a peak-to-peak level of 3.6V. However, I was a little bit surprised that the high level changed from time to time, indicating that there may be a capacitive coupling. The peak-to-peak level was quite constant.
My Ardiuno board outputs +4V as high level and 0V during the pulses, via a 1k resistor to protect the input pin of the radio. I hoped that was good enough, but should probably have taken the voltage lower with a voltage divider, and possibly inserted a capacitor in series.
I will try to find a SD card and test Tony's proposal.
murankar
Staff member
As far as sending a negative voltage not sure, but here is what I know. If you use a 12 bit analog to digital converter you will have a range of 4096 points of reference. This range will cover -2048 - 2047 because zero counts as a point. From what I know the stock on board ADC does not have a negative range and will not allow you to show negative numbers. Logically (AS I SEE IT) if you had this ADC then you wont need a negative signal. What this would like is mid stick should be reading zero. Something else to keep in mind, the 5 reference is affected by the load on the arduino. So you may want an external 5v reference voltage.
Today I tried to upgrade to the latest firmware version. Unfortunately, that did not help, so I think the digital input representing the trainer port has been damaged. I probably have to replace the board or the entire radio....
Polarity of pulses: The Arduino Nano is operated from a 4V Li-Ion battery (Voltage regulator removed), and just use a digital output pin to drive the Trainer pin (through a 1k resistor). The pin is set normally high, and then drive the signal low (zero volt) for each of the trottle, elevator, aileron and rudder pulses. They are 1ms long for stick-neutral position, and drop to 0.5ms for minimum and 1.5ms for maximum stick position. There is a 0.4ms high level between the low pulses. This pulse train is repeated with 23ms period. I use the accelerometer to measure pitch and roll and modulate the aileron and elevator pulse durations.
Polarity of pulses: The Arduino Nano is operated from a 4V Li-Ion battery (Voltage regulator removed), and just use a digital output pin to drive the Trainer pin (through a 1k resistor). The pin is set normally high, and then drive the signal low (zero volt) for each of the trottle, elevator, aileron and rudder pulses. They are 1ms long for stick-neutral position, and drop to 0.5ms for minimum and 1.5ms for maximum stick position. There is a 0.4ms high level between the low pulses. This pulse train is repeated with 23ms period. I use the accelerometer to measure pitch and roll and modulate the aileron and elevator pulse durations.
murankar
Staff member
I only know of the uno and up. All of those have both a 5v and 3v supply to the board. I have no knowledge of how the smaller boards work.
I am also going to preface this with that I am not an electronics engineer. I only know what I need to know for my current project.
I good luck with the problem solving.
I am also going to preface this with that I am not an electronics engineer. I only know what I need to know for my current project.
I good luck with the problem solving.