R&D Hex for Survey/Imaging/Conservation

crayfellow

crayfellow

Member
Mactadpole said:
1. I use an adjustable BEC (built into PDB board) to power the gimbal without a filter.
Click to expand...

OK, I will check the PDB in your build thread.

Mactadpole said:
2. 1)It depends on how you feed the battery power to the PDB. Solder both battery wires together and then to Attopilot and you get it all through there. I don't like this setup because there is potential for the power module to fail and then it all falls out of the sky. Ask me how I know. I feed the battery wires separately to the PDB. One set goes through the power module and I just scale it in Mission Planner to show what is theoretically going through both. The other set I have going through the Frsky current sensor so I also get the voltage and current on the Taranis. I find the Frsky to be much more accurate. Yes this is more complicated, it provides redundancy while also creating a few additional failure points. Pick your poison.
Click to expand...

Yeah, I was thinking of keeping the Attopilot on just one of the two, then putting a FrSky sensor on the other. It just so happens that I have a couple of the per-cell voltage sensor (this guy as opposed to the current sensor) so I will just plug that into the balance port on the battery that is not routed to the Attopilot.

This reminds me, why do we see people modifying lower rated Attopilots as opposed to just using the 180A?

Mactadpole said:
2) I do have a 5V feed to the Pixhawk rail as redundant power from PDB BEC.
Click to expand...
OK - just an extra cable from the BEC to a free port on the Pixhawk?

Mactadpole said:
3) Your choice. I feed it from BEC because I don't like dealing with multiple batteries. I'm sure there are arguments against this.
Click to expand...
that is my line of thinking too. Seems like if noise is an issue, filtering is still simpler than a whole separate battery (unless I'm missing something).

Mactadpole said:
I have a 5V and 12V built into the PDB I use (see build thread). You can of course use separate ones. Castle creations are the best IMO. In theory you should have good failsafe options with the Pixhawk and it will come home at the flip of a switch.
Click to expand...
That is exactly the approach I plan to take. I will check them out.

Thanks!
 

Motopreserve

Motopreserve

Drone Enthusiast
Not sure if you guys use (or have seen) the Hall sensor type current sensor. If you use it on one side, at least that's one side that would have no additional failure points. And it's rated much higher.
 

crayfellow

crayfellow

Member
Motopreserve said:
Not sure if you guys use (or have seen) the Hall sensor type current sensor. If you use it on one side, at least that's one side that would have no additional failure points. And it's rated much higher.
Click to expand...
Yeah - I see now from @Mactadpole 's build thread that he uses that one (I found it here), along with the FrSky voltage monitor. Cheap insurance.
 

Mactadpole

Mactadpole

Member
Yep. I certainly feel as safe as possible with this setup. I have compared both against a Whatsup meter and the frsky inductive sensor is much more accurate.
 


crayfellow

crayfellow

Member
makes sense. I'm thinking it will be near impossible to have everything on hand before starting the build, so at some point I'll just have to set about doing it.

Sidenote: I am using 45A Anderson Powerpoles and have a Tricrimp tool on the way, again as cheap insurance for batt connections. Just as an experiment I played around with soldering the 30A ones and I can see that technique will not work on the 45A ones :)
 



Motopreserve

Motopreserve

Drone Enthusiast
Perfect. I went through the same thing :)

Let me know when you get to it. There is a trick even with the overpriced correct crimper!!!
 

crayfellow

crayfellow

Member
Motopreserve said:
Perfect. I went through the same thing :)

Let me know when you get to it. There is a trick even with the overpriced correct crimper!!!
Click to expand...
I'm sure. Coming from the microelectronics world I'm wondering more about soldering 10ga wire on a PDB and counting on that solder joint alone to keep $xxxx in the air. Seems like a PDB with some kind of hard-mounted compression connectors bolted on (heck, maybe something PowerPole friendly) would be a good product. A giant wire laying awash in solder, even when the joint is good, is nerve-wracking to me even with dual batts.

Then most people are using, what, 16ga or 14ga to power ESC's, right?

Does this win for longest build thread before any actual work has occurred?
 

Mactadpole

Mactadpole

Member
I just realized we've been doing calculations based on a single 16000mah 6s battery from Tattu. They don't offer an 8000mah, all I see is 10000 from them.

Might want to go back and see what that changes in eCalc.

Think I used 16 or 18 on my wires. I'll check later for you.
 

Motopreserve

Motopreserve

Drone Enthusiast
If you're using the stock wiring on the ESCs - I think they're typically 16 gauge. KDE is 15??? Some weird size.

I haven't had any issues with the main power wires - but I'm always sure to offer some stress relief. And checking the connections regularly should be part of your routine maintenance. And of course, use a real soldering iron.

And I do think you win for longest thread before thread locker has been used. :)
 


Mactadpole

Mactadpole

Member
Motopreserve said:
If you're using the stock wiring on the ESCs - I think they're typically 16 gauge. KDE is 15??? Some weird size.

I haven't had any issues with the main power wires - but I'm always sure to offer some stress relief. And checking the connections regularly should be part of your routine maintenance. And of course, use a real soldering iron.

And I do think you win for longest thread before thread locker has been used. :)
Click to expand...
Agree with Moto on the power wires with plenty of stress relief. I have also been coating all the connections on the PDB with a thick layer of liquid tape but this does prevent you from checking the actual solder joints. I mostly look for any damage to the smooth liquid tape surface. Maintenance is key.

I have one of those old school style big soldering guns (it was my dad's, gotta be 40 years old) that I use for the PDB's and battery connector changes. It gets HOT!!
 

Motopreserve

Motopreserve

Drone Enthusiast
You know, I was just thinking about my joking about the length of time you've been posting, researching, posting some more, more researching....

It is obviously good to be smart and thorough. No question. But what I also realized a long time ago is that at a certain point - you need to be ready to lose the investment. I know this may not be a popular opinion. And it's certainly not what people (or myself!!) want to hear. But honestly, I had to admit that no matter how smart and diligent I tried to be - I would not be able to think myself out of bad product quality control, or glitchy code.

At some point, we need to trust the technology, the advice we've been given, and the research we have done to ensure a safe and proper build. But without "letting go" I found that my ability to fly safe, think on my feet, and avoid pilot error was being crippled by my fear of losing the craft.

So I now do everything in my power to design and build the best possible craft. I don't cut corners in any critical areas, and I feel I do a pretty damn good job. But one bad day at a factory - even those we feel are a cut above the rest, means the power has been taken out of my hands.

So a very long winded way of saying... get the soldering iron out. It's time to have some fun and build this thing right! :)
 

crayfellow

crayfellow

Member
yep for sure! Just happens that the most important components aren't here yet since I had to space out the expenses (PDB, ESC's, Motors). I am flying a test rig (250mm quad) in the meantime. Also have a Pixhawk rover to get used to the Ardupilot code, MAVLink, available software tools, etc.

So while it may seem like I'm slow, this is R&D so it has to happen alongside client projects that fund it :)

I will probably just ignore the gimbal mount, rails, etc. for now since I won't even have it for at least a week. Just means I'll need to disassemble the frame once I figure that out. Most likely will be rails of some sort.
 

Motopreserve

Motopreserve

Drone Enthusiast
You're on it! And please understand, I did not mean my post to come off as a critique. I just realized that there is an obvious incentive not to crash - but also the potential for a negative side effect.

Happy balance - just like the craft :)
 

crayfellow

crayfellow

Member
Motopreserve said:
You're on it! And please understand, I did not mean my post to come off as a critique. I just realized that there is an obvious incentive not to crash - but also the potential for a negative side effect.

Happy balance - just like the craft :)
Click to expand...
Did I mention we have 4 little kids we're homeschooling too? :D

BTW I saw the dude using the "big copper washer" PDB method - that's right up my alley...
 

Motopreserve

Motopreserve

Drone Enthusiast
There's no rush. You'll get there.

And certainly kids should come first, although I don't have any - so I have too many Multirotors :)
 

crayfellow

crayfellow

Member
Motors (KDE3520XF-400) and ESC’s (KDEXF-UAS55) are on the way.

One thing to keep in mind for others looking at a similar build is the KDE compatibility note regarding Pixhawk:
http://www.kdedirect.com/blogs/news/18220995-xf-uas-esc-and-flight-controller-compatibility-list

But it looks to me as if I will not have to do anything different given the plan to power the servo rail. This gives a (-) and (+) loop to eliminate errant noise caused by the lack of separate loops for each output port.
 

You must log in or register to reply here.
Top