Replies to This Discussion

3D RoboticsPermalink Reply by Chris Anderson on October 13, 2012 at 9:58pm

"Now does this mean i can plug all my outputs from my 6ch RX into the inputs on the board, with the BEC from the ESC still going to my RX like normal?"

 

Yep! Standard ESCs have a power regulator onboard, which can power both APM and the RC receiver. 

• ▶ Reply

Permalink Reply by Finnius on October 14, 2012 at 12:06am

Thanks!

So the power output from a standard BEC is fine, its ~5V i think from memory?

Just out of interest, which channel (out of 6) would the the power go from the RX to the APM?

It couldn't just power the APM each time a channel is used, because the APM's processor is running all the time...

Quite a simple way to power it, i must say!

Thanks for the reply!

• ▶ Reply

Permalink Reply by Drone Savant on October 14, 2012 at 4:31am

Be careful with which ESC's you pick. There have been a few conversations lately about IF or IF not it is right of us as a group to be using (and suggesting that others use) the ESC's for power due to their unpredictable nature. So be sure that you have the "right ESC". Aka... purchase 3DR branded ESC's for best compatibility as they have been "tested" to work with the APM boards. 

http://diydrones.com/forum/topics/apm-2-5-only-boot-via-usb?comment...

"Like I keep saying, if you have the right ESC the APM will work fine, but grab an ESC off the el-cheapo discount rack and they don't power up correctly and they don't put out a voltage that works.  Simple as that."

http://diydrones.com/forum/topics/apm-2-5-only-boot-via-usb?comment...

"To be honest, I think the days of most people powering off the ESCs are numbered."

Bad ESC power can lead to things like this so beware... its not fun in flight: 

Good luck (and remember... ESC power may not be as simple as you think)!

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on October 14, 2012 at 5:00am

That's wild.  How do you do that?!

It would appear that when the motors RPM spikes, the ESC regulator must dip?  That's insane.  I've never seen that demonstrated like that.

Is the Zigbee taking a ton of power or what?

• ▶ Reply

Permalink Reply by Drone Savant on October 14, 2012 at 5:04am

I wish I knew sir. =] This was during my side by side .cn clone vs OEM APM tests... the clone had an identical setup but did not do this. I have never gotten to the bottom of the cause of it. 

I switched the method of power over to a dedicated Castle BEC which seemed to solve the situation. I haven't done much testing on that particular pair of APM boards for a month or so. I'll let ya know if I ever figure out the exact cause. 

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on October 14, 2012 at 5:43am

Wait, so the genuine board did a brown-out, but the clone did not?  When you say "identical setup" were you swapping boards into the exact same frame/ESC's, or you had seperate machines?

Check for solder on the power jumper pins.  If they are soldered, but you can still recreate the condition, resolder those pins, add a bit of solder, and try again.

I have flown these systems A LOT, and have never had any problem like this.  I solder all my own pins, and always use a dedicated power supply for the APM.

• ▶ Reply

Permalink Reply by Drone Savant on October 14, 2012 at 7:52am

I had purchased two identical DJI F330's specifically for a clone vs. OEM shoot out. They did indeed have 1 for 1 the exact same parts on each. They were two separate craft. 

At one point I did swap the board between frames and the issue was still present. If you look closely in the video you can see the sister craft on the floor behind the one being filmed. I was calling them DJI Red & DJI Blue on my Twitter feed a few months back if you follow along there. 

Both of the boards are in my APM pile for now but I'll certainly recheck them and follow up. I'm currently on a 3dr brand flight hiatus for the most part. Our group has for the most part been focusing on clones of *other* autopilot brands and general flight comparison testing. 

I will keep you in the loop though. 

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on October 14, 2012 at 8:04am

I don't follow anything on twitter.  I do not use twitter, it's not a good interface for doing anything useful, IMO.

• ▶ Reply

Permalink Reply by Boston UAV on October 15, 2012 at 8:42pm

Brownouts are not fun:

This was 4.9V to the output rail via ESC (worked fine for ~30 flights). After rebuilding it failed again during the test flight. Going to try a dedicated UBEC on the RX.

*At this point I'm guessing its a brownout, but there could be another cause.

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on October 14, 2012 at 4:58am

The ground trace on the input of the APM is common to all channels.  Same for the power trace, they are all common.  Same deal on the Rx.  So, you really only need a single positive and negative wire between Rx and APM.  However, probably good idea to run at least two for some redundancy.

Now, you have to be careful powering the APM2.5 in particular.  Check this thread:

http://diydrones.com/forum/topics/apm-2-5-only-boot-via-usb

It would appear that not just any ESC can power the APM2.5 via the output rail.  I do not really understand why, I don't think anybody has gotten to the bottom of this.  I have not heard of this particular problem with any board before the APM2.5, nor does this problem exist with any other Flight Controller.

I recommend always powering the APM with a seperate supply in any case.

• ▶ Reply

Permalink Reply by Finnius on October 14, 2012 at 6:05am

So, overall with the APM2.5, should i power it from the RX into the APM input or from a separate ESC into the APM output?

I am doing a Bixler build for my first drone - and was hoping to just power it using the standard ESC through the RX into the APM2.5.

But now i might use a separate battery and ESC - so would this seperate ESC plug into the APM output?

I read this is the other discussion:

"If J1 is open the APM/RX is powered by the input rail side. The output rail side will be powered by the ESC(BEC). J1 links the output side power rail to the input side power rail."

That sounds like the APM2.5 needs 2 sources of power?

Cheers

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on October 14, 2012 at 7:12am

It only needs two sources of power if you have JP1 removed.  Unless your ESC is Opto-isolated, then it will have it's own internal power source.  This power would backflow to the output rail.  JP1 links the output rail to the input rail, but it goes through a diode which drops about 0.2-0.3V going from the output to the input.

This is the problem. Because the BEC in the ESC outputs 5.0V, it gets dropped to ~4.7V when it reaches the input side of the APM.  This is too close to the brownout voltage of the APM, IMO.

I suggest leaving off the JP1, and powering the input side seperately.  Use something like a 7805 linear regulator, or a Jdrones power supply.  They make a nice little 3A switching regulator that they claim is very "quiet" and only weighs a few grams.

• ▶ Reply

• ‹ Previous
• 1
• 2
• 3
• 4
• Next ›
• Page