Servos Offline

[ZK-SVH]

Hi everyone

I have a Skyview HDX with autopilot, operational in my aircraft for about 190hrs and 18 months. I’ve really been enjoying it.

Recently though I’ve found that my servos are going offline. I get a aural “Message” warning, with the word “message” being momentarily displayed, but when going to the message there is no message there, except a couple of times I’ve been fast enough on the button to open the message “Pitch servo offline” together with “roll servo offline.”

At the same time flashing very briefly in the autopilot task bar I get the two white crosses indicating that the servos have gone offline. However it is always only momentary, literally the blink of an eye and I don’t always see it. The white crosses never remain as they do when I power the servos down by pulling the breaker.

If the AP is engaged it disconnects.

I’ve read every relevant thread I can find and looking at the user guide this message is supposed to indicate loss of power to the relevant servo. Power to my servos comes from a 5 amp breaker. From this breaker each servo is ”fed” with a separate wire. Likewise each servo has a separate earth. So I’m reasonably confident it’s not a continuity problem there. I’d initially suspected that it might be transponder interference I’d read about, but the data and power wires are down one side of the airframe and the antenna wires the other side.

The antenna is about 30 inches from the servos, and in any case it all worked flawlessly for quite some time, and flight testing with the Tx off I still intermittently have the scenario described above.

The two servos plug into a hub adjacent, with a network cable going forward.

Is there a possible failure mode in the breaker (Klixon) that results in a fleeting interruption of the power supply? (Will swap it out as soon as a replacement arrives.)

Something wrong with the network cable exiting the hub maybe but this hub also has remote magnetometer plugged in and I’ve never had an indication of a problem there.

Is it possibly a software thing?

Any thoughts greatly appreciated.

Thanks

Peter

 

[jnmeade]

Sounds similar to my experience, which is reported here and you probably read about. In my case, both servo power wires terminated in one ring connector and that termination in the ring connector failed, although physically it looked fine. It drove me nuts finding the problem. Eventually, it was accidentally wiggling the wire when doing some other task. My poor diagnosis and detection technique but there is no doubt it was hard to find.
Is it possible that a connection is OK when cold and becomes unreliable when warmed up or when it is jiggled?
Good luck. I guess my lesson was not to assume that because something looked OK it was OK, but to test and check.
Is there any way you can induce the failure?

 

[airguy]

You have separate grounds for each servo, so that's not likely the problem - but both servos are fed from a common power supply, and both servos are going offline/online together - that's your smoking gun. You most likely have a power supply problem to them.

 

[ZK-SVH]

Thanks for the input guys. What is really perplexing to me is the completely momentary nature of the event, if it’s a power interruption as it seems it is never more than the briefest instant, the white crosses flash for a fraction of a second and disappear.

 

[swatson999]

Do you have a schematic?

 

[swatson999]

BTW, a) you should be able to make those messages latch, instead of self-clear, and b) did you download the data from the system to something like Savvy Analysis, as well as take a look at the Alarms file data dump?

 

[ZK-SVH]

Aaah you’re right 999 I will change the settings so the messages latch and educate myself on the diagnostic file procedures etc.

I’ve now installed a new circuit breaker but with the weather and life won’t be able to fly for a few days.

 

[swatson999]

Well, if you can't fly, at least download the data and analyze it (before it gets overwritten).

 

[Marc_J._Zeitlin]

Thanks for the input guys. What is really perplexing to me is the completely momentary nature of the event, if it’s a power interruption as it seems it is never more than the briefest instant, the white crosses flash for a fraction of a second and disappear.

Sure sounds like an intermittent momentary power loss, like would occur when strands of wire are just barely touching the inside of a connector and there's some vibration.

 

[ZK-SVH]

Hi Swatson. The power and ground for the servos are wired like this, I’ve omitted the stuff going to the network.

Much to my enduring embarrassment when I wired everything it was only when I got to the end of the install I realised that I hadn’t drawn up a schematic as I did it. I still feel like such a fool, as I did each little bit I just lost sight of the big picture…... Everything has been fine until now…

When I get back to the plane (next week) I’ll take the panel out and recheck all the relevant connections. (I did that before originally posting and they all seemed secure maybe I missed something) I’ve focussed on the commonalities in the circuits for the servos, power grid the bus bar, the circuit breaker connections and the breaker itself in case there’s dune weird failure mode and where the two grounds terminate together.

 

[maartenversteeg]

A question: what's the reason/purpose of the double wiring in between the main bus and the circuit breaker?

 

[airguy]

Thanks for the input guys. What is really perplexing to me is the completely momentary nature of the event, if it’s a power interruption as it seems it is never more than the briefest instant, the white crosses flash for a fraction of a second and disappear.

Yup, welcome to the frustrating world of intermittent electrical faults. Hardest thing to diagnose/find.

 

[jakej]

Hi Swatson. The power and ground for the servos are wired like this, I’ve omitted the stuff going to the network.

Much to my enduring embarrassment when I wired everything it was only when I got to the end of the install I realised that I hadn’t drawn up a schematic as I did it. I still feel like such a fool, as I did each little bit I just lost sight of the big picture…... Everything has been fine until now…

When I get back to the plane (next week) I’ll take the panel out and recheck all the relevant connections. (I did that before originally posting and they all seemed secure maybe I missed something) I’ve focussed on the commonalities in the circuits for the servos, power grid the bus bar, the circuit breaker connections and the breaker itself in case there’s dune weird failure mode and where the two grounds terminate together.

Gotta ask - why did you parallel the power to 1 x C/B ? That's not the way to do it, you need independent power to each servo so that IF one goes awol it will not impact the other one - makes troubleshooting a lot easier too. Wiring the way you have could, in some circumstances, result in both servos needing repair - hopefully the realization of potential double repair cost scares you enough - just saying

 

[ZK-SVH]

Hi Jake

I just doubled up my 22g power wires to the servo cos I wasn’t sure that a single was big enough, wired both servos through one CB following the diagram. Am I better off in the course of working this out putting each one on a separate CB? This is all new ti me as a complete beginner it was a huge job.

Thanks Peter

P.S. The reason I chose Dynon was for the support I’d seen come through both this Forum and Dynon compared to the other vendors. It’s really appreciated.

 

[jakej]

Peter, all important ,essential, critically equipment should be on independent circuit breakers or fuses. Doubling up those items can create an unwanted stressful situation If one ‘blows’. I see what’s on the diagram however you can do better. The. ‘accepted norm’ is one ’breaker’ for the above reasons.
As you’ve probably realised, troubleshooting a circuit would be a lot easier just chasing a power wire ( & ground) fault, or a single device fault, than having to do double the work.
IF the fault is power/ground then that is easier however if it’s 1 x servo, which one? Independent circuits make that easier too.
let us know how it turns out
ps pm if you want.

 

[airguy]

Peter, all important ,essential, critically equipment should be on independent circuit breakers or fuses. Doubling up those items can create an unwanted stressful situation If one ‘blows’. I see what’s on the diagram however you can do better. The. ‘accepted norm’ is one ’breaker’ for the above reasons.
As you’ve probably realised, troubleshooting a circuit would be a lot easier just chasing a power wire ( & ground) fault, or a single device fault, than having to do double the work.
IF the fault is power/ground then that is easier however if it’s 1 x servo, which one? Independent circuits make that easier too.
let us know how it turns out
ps pm if you want.

For troubleshooting or fault-finding, I agree that more is always better - but reality is that we cannot instrument everything to the maximum degree possible in our aircraft. From an operations standpoint, I don't care if one or two (or three) of the autopilot servos have a problem in flight - if the AP system is compromised, then it's INOP for the flight and troubleshooting isn't happening until later. I have one breaker for both my servos, and if I have a servo die then I reset the breaker ONCE - if it pops again or the servo goes offline again, now I'm hand-flying until I get to a time/place convenient for troubleshooting. It's not realistic to say that in all cases every circuit needs a separate breaker, we would need a flight engineers seat in the cockpit, not to mention that every added connection or device is another point of failure.