As I understand this so far my Rx gives a PWM signal thats 2.90V peak and varies it in width from 1.0 to 1.8 mS (without trims etc) and my APM2 does about the same at 5.23Volts.
None of my servos respond to this signal.
Min Neutral Max Peak V
APM 1.2ms 1.8ms 2.0ms 5.23
RX 1.0ms 1.4ms 1.8ms 2.90
All the first 4 outputs vary in width properly but they are the wrong peak voltage!
Any ideas anybody?
Thanks John Fletcher
I seriously doubt that is the issue. The servos should accept a normal 5volt PWM signal. Further, common sense says the outputs of the APM cannot be higher than the input voltage so what are you powering the APM with? Obviously it's higher than 5 volts. Most servos will take 6 volts, but it's not normally recommended, and meanwhile in a normal RC plane or heli, the receiver would also be at 6 volts putting out 6 volt PWM, so either your servos are damaged or there is something else going on. Again, the APM is going to put out what you put into it and it had better not be over 5 volts. Maybe you have a 6 volt BEC?
By chance are you powering the APM via USB? IF so the outputs are NOT powered so while the signal is there, 5 volts to power the servo isn't. IF you measured the signal and it's there on the output pins, then the servo should work period.
It's also not clear how you "measured" both pulse width and voltage of the outputs. Even based on your own measurements and assumptions, the signal scenario is that the servo expects a 3.3 volt signal and your APM is sending 5, so the problem is the low part of the 5 volt signal isn't read as a 0 or low thus the servo just sees the APM signal at high all the time. I seriously doubt this on so many levels. My big concern is that we KNOW there is a voltage drop through the mega 2560 outputs, so if the voltage is that high, your input into the APM is even higher.
First, Thanks very much for the reply.
The 5.23 Volts is about 0.15 below the output voltage from the bec and this is normal bec output (about 5.25 to 5.5V seems typical). I wasn't aware that servo's could accept this much higher voltage than the receiver gives out, now I know that they do!
I've found some other things out along the way.
When I measured the signals, I also measured the rails both by DVM and found:-
5V input at 5.38V.
5V output at 5.36V.
Yet my scope shows the signal is as described at 5.23V and the output from several receivers is 2.9V (futaba), this is true for APM1 too.
The APM was not connected to a computer at all.
What it seems the problem is that the 5V output rail drops to almost nothing when a servo is connected, which explains the lack of response. From what you have said it seems there is some way of switching the 5V output on and off and it might be that this has failed. It's my guess that to protect the computer usb 5V the normal 5V supply to the APM is disconnected and the usb 5V used instead.
Lots of this is difficult without a diagram to help see what happens and where.
I hope this helps to explain what I've found and how and I'd be pleased if you could help some more!
Thanks John Fletcher
Well, the solution goes like this :-
First, installed the power is normally applied to the 5V rail on the OUTPUT side of the APM. APM1 does not care which side it's powered from and works even if the receiver is powered and the servos still work properly and everything is OK. APM2 does not work like this, powering the INPUT side or the receiver causes the APM to function normally and 5V to appear on the output 5V bus, but there is no power only volts. As soon as it's loaded the voltage disappears.
This is a difference in design between the 2 boards, it's taken quite a few hours to discover this!
The signal levels really are the way they are described in the table, and there really is a voltage gain of around 2.3V.
It does not matter what state the link at JP1 is in for this issue.
I really hope this helps someone and that it's of interest, APM2 is quite different to APM1.
John, I could be wrong but I thought most of the power situation is discussed in the instructions for the new APM 2.0 hardware already. None of that was a surprise other than your signal level measurements, but even then, it follows the 10% rule, especially in consideration you are out of spec on your input power to the APM.
It's discussed in 'Alternative ways to power your board', but I wasn't looking for an alternative I wanted to know why what I was doing didn't work. It worked for APM1!
Lots of the time people are solving problems that don't exist because they missed the information or because the infomation they need is buried somewhere. It would be great if I had time to read through lots of this. What I'd like to see is a manual for the board and Mission Planner.
As for the input level, I've seen several questions about this and how critical it really is and everywhere there are references to being powered by BEC, but BEC output varies from about 5.2V through to about 5.8V and is quoted as 5V most people are going to use that as 5V without checking, but a manual would tell us all just exactly what was needed.
Do I need to provide 5V exactly? Is 5.1V in spec. In short what is the spec?
Normally I look these things up either in a pdf or on paper, here the information seems diffused.
Many Thanks anyway.