Good battery source?

[clolson]

This may not be the best place to ask this question, but it's for a UAV I will be building, and there haven't been very many new posts here lately.

I need to find a 3.7v li-poly battery in the 1200-1500 mAh range to power my Xbow unav/stargate. Tower hobbies does not sell any single cell li-polys. Neither does hobby lobby (they just discontinued their last one weeks ago.)

I'm having a tough time finding anything online. Does anyone know of any good sources for buying a single cell li-poly battery in the mAh range I need?

I'm adverse to soldering, but could probably manage if I absolutely had to ...

Thanks,

Curt.

[kd7ost]

Curt,

How about one of these?

Dan

http://www.allelectronics.com/cgi-bin/category.cgi?category=160&item=LBAT-43&type=store
http://www.allelectronics.com/cgi-bin/category.cgi?category=160&item=LBAT-43T&type=store

[RCS1000]

Cheapbatterypacks.com

[mckaneorg]

Curt,

Let me know what you think about the XBOW. I have a BTA and a PICO. The price is much different but I just wanted to see what you thought about it and what you like about the XBOW over the others out there.

-j

[djklein21]

call steve belknap at diversity model aircraft
858-693-8188
www.flydma.com

[Vindication]

Just be careful about using a LIPO w/ your hardware. You need a proper voltage cutoff our you will kill your lipo. NiMH's and such don't need this.

[dbeamon]

try balsa products they have them all...

[clolson]

Curt,
Let me know what you think about the XBOW. I have a BTA and a PICO. The price is much different but I just wanted to see what you thought about it and what you like about the XBOW over the others out there.
-j

So far I really like the xbow unav in terms of it suite of sensors and its openness and programability. On the flip side, it's not quite a full drop in autopilot. There is a lot of do-it-yourself pieces that have to be put in place.

You need to buy (and hack) a receiver to get the ppm signal off it to the unav.
You need to scrounge batteries. They recommend a single cell 3.7v lipo to power the unav, which was harder to find than I expected and is what this thread is about.
The software side of the unit is very primitive. I plan to eventually throw out most of their code and write my own.

So I see a *lot* of potential with this unit, but there is also a fair amount of effort to get it up and flying.

Curt.

[mckaneorg]

So far I really like the xbow unav in terms of it suite of sensors and its openness and programability. On the flip side, it's not quite a full drop in autopilot. There is a lot of do-it-yourself pieces that have to be put in place.

Sounds like its a decent platform. I have several units myself as mentioned above and I wanted to know what you like the best. Have you tried any of these with a heli? I have been thinking of diving into a custom project based on some of the AP systems out there. They all seem good but none quite perfect. If I have to hack it as you say, I would just prefer to use a gumstix platform to design a runtime specific for flight data. There has also been talk about collision avoidance and weather avoidance. I feel I could incorporate this into a Linux based operating system or even a HP PDA running .NET systems. I am just not willing to totally re-invent the wheel on this.

Thanks for your input though, very helpful!

[Unterhausen]

I just decided to buy a second unav based on the software they released. At least it works and so smacking it into shape is not an impossible problem. It sure beats spending $10-$60k for source code like we were going to do on other platforms we considered.

On the other hand, their coding standards are not very good even by the standards of this mechanical engineer. I spent 5 minutes the other day trying to figure out a while loop and finally realized that they had hidden a curly bracket in a totally non-standard place. Then I realized that they switch between different methods of formatting the code even in the same file. Ouch.

[clolson]

Sounds like its a decent platform. I have several units myself as mentioned above and I wanted to know what you like the best. Have you tried any of these with a heli? I have been thinking of diving into a custom project based on some of the AP systems out there. They all seem good but none quite perfect. If I have to hack it as you say, I would just prefer to use a gumstix platform to design a runtime specific for flight data. There has also been talk about collision avoidance and weather avoidance. I feel I could incorporate this into a Linux based operating system or even a HP PDA running .NET systems. I am just not willing to totally re-invent the wheel on this.

Thanks for your input though, very helpful!

I don't have any personal experience with a heli, but when I was out at xbow last month, they gave us a demo of a helicopter-like vehicle. It had two ducted fans mounted side by side that could be pitched back and forth. Pitch them both in the same direction and you have forward/back motion (and pitch) of the vehicle. Pitch them in opposite directions and you get yaw. Vary their relative rpms and you get roll. This vehicle is *impossible* to hand fly. But they added a xbow unav simply to add damping on all the axes and it allowed the vehicle to be flown very much like an R/C helicopter and was such that any reasonably accomplished R/C helicopter pilot could fly the vehicle (with the addition of the xbow unav for stability augmentation.) I'm not a helicopter guy, but I'm sure there's more issues for you than just rotational stability, so I can't fully answer your question, but my sense is that the xbow is probably as good as any other mems based sensor out there.

One other thing to note, the unav has a simple serial interface. So it's very possible to connect it up to any computer that has a serial port and make it work. It don't recall seeing a serial port on th gumstix though so that may be a problem?

Curt.

[mckaneorg]

One other thing to note, the unav has a simple serial interface. So it's very possible to connect it up to any computer that has a serial port and make it work. It don't recall seeing a serial port on th gumstix though so that may be a problem?

Curt.


Interesting... I will have to check that out. The gumstix can pretty much do whatever you want. I would not use the serial port to update it, rather I would use something else like wifi or bluetooth. It does have USB and I could use it to connect to a long range modem system. But then again, I would have to rebuild the software to do exactly what I want it todo. The other option I have been playing with is the NSLU2 Linksys Storage device. That is also a very real possibility.

I will buy an xbow to add to my AP collection since none of them have been really amazing technology yet.

-j

[Unterhausen]

To be clear, the unav is really just a sensor. So you could hook it up to the gumstix, which does have plenty of serial ports -- you need two for the unav, one for the sensor data and one for gps. I'm almost positive it would require an adapter to connect them, although they both use the same style of Hirose connector. I think the xbow software that runs on the ap computer should work fine on the gumstix, although one person on this forum had trouble compiling it.

[Vindication]

Why do you say it needs two for the unav? And what other sensor data do you NEED?

Ie I thought you would need 2-- one for unav one for gps. I've never used unav.. just trying to find out.

-Vind

Oh I was the one that had trouble compiling it. I'm wondering if it has something to do with using the Gumstix version of buildroot which isn't up-to-date with the generic one...

[clolson]

The unav includes:

- 4hz gps
- gyro and accelerometer sensors
- 3 axis magnetometer
- static and dynamic pressure sensors
- ability to read an R/C receiver output (via a little hackery)
- ability to drive several servos (I forget how many off the top of my head.)

You only need one serial port to talk to the unav.

The unav doesn't do any kalman filtering or sensor integration on board, so you have to hook it up to an external computer to do that. They are setup to plug into a stargate board (Xscale 400mhz cpu) but you ought to be able to run their code on gumstix, a standard pc, or anything else with a decent c compiler and standard development environment.

[Unterhausen]

I stand corrected, it has 2 serial ports connected to the stargate and on the discrete pin connectors, but it doesn't have to use the second one. That's what I get for looking at pinouts and not the programming info. You can hook up to the serial port at rs232 voltages with a 3 pin connector, which might be the easiest way to hook it up to anything other than a Stargate.

It will drive 10 servos including what they call a "high speed" servo.

Vindy, you never did tell me what went wrong on your compile. Crossbow tends to ship older tools too, which is fine because they work. The rapid development of linux can be a real pain if you are trying to do anything other than desktop development.

[Vindication]

Unter,

I'll have to look at it again. But it probably won't be for a couple weeks at least. It's finals time in my Masters program...

-Vind