Damien,
Sounds like you have a lot more experience with the electrical side of this than I. I just thought that since it sounded like you were experiencing issues initially that you may have overlooked this thinking it didn’t work. But if you were using the RC calibration page to see the inputs then you were doing it right. The status page is another way to this.
View the videos, I think they should be a big help in leading you through the set up.
While it seems you’ve done some due diligence, I find it nearly impossible to believe that an F427 based board does not support inverted logic for the SBus protocol (as nearly every board in recent and not so recent history has). It’s an extremely common protocol.
Hi Damien,
Did I understand you correct you don’t have the special Tradi-Heli parameter in your parameterlist on mission planner if you connected to your FC? So which ardupilot firmware version did you upload to the FC. You know that tradi-heli has a own special firmware.
I am with @bnsgeyer that best is to use a serial protocol like SBUS and I am with @Yuri_Rage that most modern FC will use the SBUS signal correct. For this I don’t see anyreason at the moment to change your good old FUTABA (I like and use also this old stuff). IF you have still a problem with the SBUS on your FC a simple small signal transistor will do the job to invert the receiver signal externally. If this don’t will work and if your receiver is correctly configured to us the port as SBUS out port think about change to are more modern FC.
I use on my FUTABA ch1-4 to control the servos, ch5 as mode switch (only 2 modes possible due to the switch on the FUTABA t6ex) ch6 as motor interlock.
Ok, just checked you log file.
This shows the Tradi-Heli special parameters (Starting with “H-”) so firmware seems to be ok
Hi Bill,
I have a degree in Electronics, worked professionally in the telecomms and aerospace industries for 34 years and built my own proportional radio control set, (including the pcbs) when I was a youngish teenager… that’s why the bother I’m having with my PixRacer is so irritating!
I think there’s something I am missing in all this but I’ll have another go at it later this evening… and of course, thank you for your continued assistance!
Ok, I’ve been through sufficient of the setup to nominally get it working but now have a number of other problems! The main two are: it’s VERY jittery and my right front swashplate servo is no longer functioning. Currently I have no idea why that should be, but it is the remaining original servo bought with the T450…I have a spare so will swap it out to see what effect that has. The tail rotor servo is also very sluggish…is this a feature of the FC, I wonder? I note there is a pass through option (in the swashplate setup video) so it will be interesting to see whether the sluggishness improves if I implement the pass through. Motor ESC is strangely intermittent. I can still run the motor without arming (when it actually bothers to run ).
Clearly I have a lot of work still to do…but that’s for tomorrow.
Best regards,
D
Hi Juergen,
As far as I can see, I have the traditional helicopter firmware…which I think you confirmed in a later email.
I agree, a simple transistor would probably work but it really should have a couple of resistors added to bias it correctly. That’s all a bit of a faff to set up and not very practical except for test purposes. I haven’t got a transistor to hand (sure I have some somewhere but can’t quickly find them) and so I opted for a proper logic inverter which I ordered over a week ago. Unfortunately that parcel appears lost in the post at the moment but I can continue with my efforts to get the PixRacer working using the 8 to 1 encoder with associated clumsy wiring…I unfortunately have plenty of problems to resolve at the moment!
I have spent far too much time on this already, struggling to cope with the various problems and so I am already considering scrapping at least the PixRacer but also the Futaba radio. I don’t really want to do either but this is getting ridiculous. I have a nagging feeling that there may be a degree of incompatibility between the latest firmware and the V1.0 PixRacer which may be the cause of some of my problems…but I have no immediate way of proving this.
Do you have any recommendations for a different FC please? It has to be tiny to fit on the T450… I fancy the Radio Master setup with the Nexus FC…I shouldn’t have to consider wasting what I have already purchased but how long should I continue struggling?
Thank you for your support, it is very much appreciated.
Best regards,
D
‘Some due diligence’ Thanks Yuri!
I’m going with Juergen’s ‘try an invertor’ suggestion (which is certainly very feasible) as soon the parts I ordered arrive.
If you would like to demonstrate a PixRacer V1.0 working with S.BUS, I would be very happy to learn how you achieved it. As far as I am aware, the PixRacer I am currently using is brand new and unused, so I’d be surprised if it was damaged.
Thanks for your thoughts and interest.
Best regards,
D
You may find that you can get it working on a UART (I’d try SERIAL4 based on hwdef - it’s inverted by default). Connect the signal wire to the TX pin and set:
SERIAL4_PROTOCOL,23
SERIAL4_OPTIONS,4
@andyp1per may be able to help further. So far as I can see in the source, SBus should work without issue or significant configuration woes on the RC_INPUT pin.
What is jittery? the servo(s)? Do you have any telemetry radios on this? Be sure that the swashplate servos are all the same make and model. Also be sure that they are the correct servos for a swashplate (torque & speed).
Yes this is typical for heli default setups. The yaw axis is set to a very low max acceleration which makes the tail rotor servo appear to be slow and sluggish. Change the ATC_ACCEL_Y_MAX to 80000. I went back and looked at the accel max for all 3 axes and it looks like you changed them. Why? I would suggest putting the pitch and roll back to their defaults of 110000.
Interesting? you should never be able to run the motor with out arming and then enabling the motor interlock switch on your controller. describe the physical connections of servos and ESC to the flight controller and post a parameter file. To save the parameter file, power and connect to the FC, go to the Config tab and click on the Full Parameter List on the left. You will see a save to file button on the right. Use that button to save the parameter file and then post here.
Hi Bill,
Yes, the servos are very jittery. I haven’t replaced the ‘dead’ servo yet but it is entirely possible that it was the cause of some of the jittering if it was actually dying. I’ll have another go later with a fully charged battery etc. All servos will be identical when I swap that one out (Align HDS45002T).
I don’t believe I have changed any of the acceleration perameters… that’s way too deep for my current capability…all I have been trying to do is get it functional…such refinements would surely come later, once I’d had a few inflight scares!
Current situation without me changing anything, is that the motor will not start and I suspect there is some confustion between rc channel 3 (throttle) and Motor 3 (swashplate rear)…ie I suspect I still have some of my previous feedthrough in the setup.
I think you are right about the t/r sluggishness.
As ever, life gets in the way and I can only look at this short bursts unfortunately but I will persevere. Thanks to your input, I am clearly heading in the right direction.
Keep you posted.
D
No telemetry or anything fancy, just the radio, the fc and the esc.
Hi Yuri,
I think RCIn is just a UART isn’t it but your idea is worth a try…next on the list after I’ve tried the inverter idea.
My percepception is that there is some confustion out in the world because S.BUS uses inverted logic levels…ie inverted relative to a standard UART. I think Flysky came up with inverted S.BUS to address this but that of course means some people perceive the Flysky protocol to be inverted when in fact it isn’t really. My current guess is that my V1.0 PixRacer doesn’t know that Futaba S.BUS is inverted…hence it is well worth my trying Juergen’s suggestion.
I’ll update the thread when I have tried both options or if I get it to function. 
The only reason I asked about the telemetry radio is that I have had cases with some cheaper servo brands where the RF from the telemetry radio has caused jitteriness with servos.
That is strange then. Because they weren’t at their default values.
I am all too familiar with that.
Yes, but typically less user configurable.
According to the hwdef and also any documentation I can find from mRo, Futaba SBus (inverted) should be supported.
Can you confirm which firmware you’ve uploaded (version and board type)? EDIT: Disregard - I can see from your log that you’re using the Pixracer target v4.5.7, which is exactly the target I’ve been referencing to try and sort this out.
HI Bill,
Here’s today’s update on my on-going struggle to get airworthy:
My T450 has had a full life up to this point but why a servo should choose to die just now, I do not know. I suppose I should be grateful that it didn’t cling on until I had reached some altitude. The situation now (with a new servo fitted) is that broadly speaking, the system is working (Whahhaayyyy, thank you!)…and is no longer jittery, it is ACTIVE.
The reason for the system not working was as you said…I had clearly misunderstood the Servo_Function set up - I read the webpage but didn’t watch the video.
So that’s great news. There’s always a ‘but’ though isn’t there? If I roll the airframe left, the swashplate rolls left too and the pitch has a similar directional error. This presumably implies my AHRS_Orientation is set wrongly. The PixRacer is slung underneath the T450 chassis and is upside down but pointing forward, so I reckon that means I need Roll180 in my orientation settings…which is what I have set.
I appear to have three compasses…I suspect no1 is in the GPS receiver (orientated correctly with the arrow) and 2 and 3 are in the PixRacer (set to Roll 180). These orientations are displayed on the compass calibration page. I have seen a ‘Compasses Inconsistent’ error message, so perhaps my compasses need recalibration (again)?..but I am currently unable to do that due to an intermittent USB cable - the moment I start throwing the airframe around to calibrate the compasses, it disconnects MavLink. I really need to use the wireless connection but have so far been unable to. I can connect the pc to the PixRacer Wifi with no trouble, but Mission Planner refuses to connect, giving this error: Only one usage of each socket address (protocol/network address/port) is normally permitted. I have read a number of articles about this online, but no-one appears to have published a solution. I am also now unable to arm, presumably because of the above problems…(so that’s actually another improvement!)
Tomorrow is another day.
Best regards
Hi Yuri,
Copter 4.5.7 Official
Naturally I can’t quickly find the PixRacer page I have previously seen, but my understanding is that the basic PixRacer is now R15. My two units are V1.0 and mRO’s current product appears to the PixRacer Pro…which I assume is different again.
I have no idea what the differences are.
I’m currently concentrating on the FC aspects of this project but will get back to the S.BUS problem as soon as I have the mental capacity!
Best regards,
D
That is not true. As long as the HUD of Mission Planner is showing the orientation correctly as you move the aircraft in different orientations then it is not AHRS orientation. It is always very helpful that if you run into a problem and think it is a configuration issue then post a parameter file.
Verify that your RC transmitter stick inputs move according to how I describe them in the video. Set your swashplate servos to get the proper movement of the swashplate. Your swashplate movement to tilting of the aircraft should be opposite to the aircraft tilt. There usually isn’t much more you have to do.
Ok, I can confirm:
The HUD/Artificial horizon moves correctly: Nose down ->horizon rises, nose up → horizon falls, roll right → horizon rolls left etc
Tx set up is as in your video - green bars move in the same direction as the sticks.
FC works as negative feedback ie nose down, swashplate tilts backwards to compensate. Lateral FC control is correct too. ie swashplate moves against any pitch or roll.
Unfortunately, the Tx Pitch stick effect is reversed: Stick forward, Swashplate tilts backwards but roll stick and swashplate are correct.
My Tx channel directions are set as follows:
Ch1 - Reversed
Ch2 - Reversed
Ch3 - Reversed
Ch4 - Normal
Parameters attached…It was 1.00am when I sent my last message - too late to be sending parameters - apologies!
If I attempt to correct the Cyclic pitch (fore/aft) reversal by reversing the relevant Tx and FC Servo channels, the above situation becomes:
The swashplate follows the sticks in both pitch and roll.
HUD/Artificial horizon remains correct
FC tilts the swashplate in the wrong direction for both pitch and roll (ie nose down, FC tilts the swashplate forwards. Roll right → swashplate rolls right)
Radio Calibration - green bars follow the sticks except for Pitch which is reversed.
I think I might take up knitting.
TREX450 - Aero Format Tx Wrong.param (17.5 KB)
Here is what I suggest. This will have you making changes to the parameters and the transmitter setup
First if you have the transmitter set for reversed on channel 1 (Roll) and you also have the flight controller RC1_REVERSED set to 1. So instead of the double negative. I would set the transmitter to normal for channel 1 and set RC1_REVERSED to 0.
For Channel 2 (pitch) on your transmitter, set the channel to normal. Do not change anything with channel 2 on the flight controller. That should fix the pitch axis and make it behave properly.
I think that should be it and it should work as we want it to.
