Xact servo over FBUS - "sensor lost"

[zeo0nie1]

I had the very bright idea of putting together an FBUS-based system.

I'm using two Xact 5800 servos over FBUS with a TW Mini receiver. I keep getting "sensor lost" messages, and the servo sensors flash red for a bit before recovering. The servos *seem* to work correctly otherwise.

I have updated the firmware of the transmitter, receiver, and servos to the latest versions:

* Servos: 3.3.0
* Receiver: 3.0.1
* Transmitter: whatever Ethos upgraded me to.

I changed the Application ID and Physical ID of both servos to distinct values.
I deleted all telemetry and re-discovered them.
I deleted some nil-value telemetry items from the servos (position, mode, velocity, perhaps?)

Any idea what could be going on? Bad splice job? Thanks!

 

[Wayne]

Question, does it have this error if the radio is moved away from the receiver?

 

[zeo0nie1]

Question, does it have this error if the radio is moved away from the receiver?

Yes, unfortunately.

 

[Wayne]

How is the solder joint on the receiver? Servos are moving correctly at all times?

You have me stumped on this one. Does the TW alloy you to alter the telemetry signal strength?

The sensor lost is usually a result of the telemetry feed not getting back to the transmitter. I have not run into this myself, but one of the guys here runs a submarine and he occasionally gets the lost sensor message when under water. (2.4 does not like water, he uses TD for the this reason, but the telemetry feed stays on 2.4 on the TD units.)

 

[zeo0nie1]

Thanks for thinking this through.

I just tried "Reduced telemetry power (25mW)" in the receiver options, it didn't do anything that I can tell.

OK, a little bit more info after some tinkering:

Looking at the telemetry screen, I am guessing the dot on the left of a sensor name indicates that it is being updated?

With the two servos plugged in, the normal Rx readings are firing constantly, the SRVx values are firing sporadically. When the SRVx delay is too long the row goes red and the transmitter complains about losing a sensor.

With ONE servo plugged in, the dot on the left of the SRVx sensors fire consistently at what looks like once per second.

I am also noticing that the servos randomly "jump" - make noise on my table across the room while I have the transmitter here with me while I type this out - has happened a couple times in the past ~10 minutes.

It looks like *something* is happening with the wire protocol - messages are clobbering each other too often or something, I don't know. It's pretty frustrating that the protocol isn't documented, I don't think I'd have tooled up with FrSky if I'd have known that was the case.

It could be my soldering. I'll investigate. In the meantime, I'll just use a PWM receiver. There is still value in these in that I can center them without having to buy a separate programmer like you do with the KSTs (also frustrated that their programming protocol is not documented, let alone having open source firmware to do it.)

Anyways, thanks for your thoughts on it. Will try again when new info or firmware comes to light.

Edit: yes, the servos seem to move correctly the whole time.

Edit 2: I set "Data Rate" to 1900ms and 2300ms on the two servos, hoping to avoid collisions. It didn't help.

 

[Wayne]

Hmm, I did have one plane with 6 of these servos in F-bus mode on an ARCHER receiver and all was good. I was not monitoring their telemetry though as I didn't see a need for that plane to do so. Had no issues once I had firmwares updated and servos programmed. Was rather pleased with how well it worked.

When you say the servo is "jumping" what do you mean? Do you actually see the servo arm move, or is it more the typical noises of a servo with a very tight deadband?

I like my solder joints to be shiny silver, it looks like yours may be dull silver? (hard to tell in the photo) If they are dull it is a sign that things are not good there. Would not think that would cause much issue in this application though.

 

[zeo0nie1]

Good to know that it can work in this configuration

I only heard the jumping, it sounded like servo arm made the thing jump off the table from a spasm. At some point I'll try to keep a video on it, but for now I've moved on - the servos are in a wing already.

I dipped the splices in rosin before soldering, I think that's where the dullness comes from. I think that if there's a bad connection, it's probably the one to the board, which in my infinite wisdom I had already secured with hot glue.

I notice that the GR6 comes with a connector with FBUS access, I think I'll try splicing that into the loom at some point to take my hand-soldered connection out of the equation.

Edit: also, it's a bummer that you can't use the channel output pins on a GR6 for FBUS. I think I saw a video somewhere demonstrating that functionality for a different series of receivers.

 

[Wayne]

Yeah, for some reason FrSky only does that on the bigger receivers.

 

[zeo0nie1]

One of the servos continued to be problematic in PWM. I swapped the bad one out and now two servos are working fine with telemetry over FBUS to the TW Mini on my desk.