RC Groups

RC Groups
    Power Systems
        Thinking about UBEC's and this seems WAY too easy

#1 THUREN Jul 18, 2005 04:00 PM

Thinking about UBEC's and this seems WAY too easy
 
Just a thought as I am fairly "noobish" when it comes to small electronics. Anyway, would something like this work as a BEC? It's a plain voltage regulator but from what I can see, it seems pretty self contained, where added circuitry would not be needed. Note it says it can handle a 3amp output... Just something I was thinking about... :)

Don...
PDF sheet
more...
http://www.st.com/stonline/prodpres/...cons/to220.jpg

#2 Martyn McKinney Jul 18, 2005 04:14 PM

Typically, devices this size can dissipate approximately 5 Watts without a heatsink.

If a 5V device were run from a 10V supply and was supplying 1A, this would be 5 Watts dissipated by the regulator.

If supplying 3A from a 10V supply, this would be 15 Watts and the device would shortly go into thermal protection.

#3 THUREN Jul 18, 2005 04:31 PM

Interesting.... can something like this be run in parallel with no other components? :)

#4 Martyn McKinney Jul 18, 2005 04:33 PM

In parallel ?

With what ?

#5 spastic Jul 18, 2005 04:37 PM

Quote:

Originally Posted by Martyn McKinney
In parallel ?

With what ?

more than one of the fets in parallel is what i think he ment

these are what most 2-3cell ESC's use, thats why they are limited to 3 cells when the difference in voltage becomes larger they become less usefull and a switching regulator is needed

#6 mmormota Jul 18, 2005 04:59 PM

Quote:

Originally Posted by THUREN
Interesting.... can something like this be run in parallel with no other components? :)

Just use a heatsink instead.

#7 THUREN Jul 18, 2005 05:39 PM

Thanks so far everyone.. :)

So if I were to get 3 of these regulators, attach them to a small aluminum heat-sink, solder the equal legs together, and wire the assembly to my 3 cell lipo pack for testing, do you think it would be worth it? I could mock it up with some light bulbs to simulate 3-5amps and see how hot it gets before smoking my Rx/servos.


My main question I guess is, if wiring the legs of 3 or 4 of these together, is the correct way to do this, or if other things are needed... :)

#8 Martyn McKinney Jul 18, 2005 05:51 PM

Should work, although if they were to do this on the Space Shuttle, because these are voltage sources, not necessarily exactly equal, they would use small balancing resistors on each of the outputs.

Usually bypass capacitors are added from the input and output to ground in order suppress any spurious oscillation.

Putting 3 regulators in parallel would not smoke your Rx and servos.

The output would be 5V at a max of 3A (when used with a 10V supply).

If the load current exceeded 3A (approximately 15W dissipated in the regulators if a 10V supply were used), the regulators would go into thermal protection and shut down, supplying zero volts to your receiver.

#9 vintage1 Jul 18, 2005 06:21 PM

Quote:

Originally Posted by THUREN
Thanks so far everyone.. :)

So if I were to get 3 of these regulators, attach them to a small aluminum heat-sink, solder the equal legs together, and wire the assembly to my 3 cell lipo pack for testing, do you think it would be worth it? I could mock it up with some light bulbs to simulate 3-5amps and see how hot it gets before smoking my Rx/servos.


My main question I guess is, if wiring the legs of 3 or 4 of these together, is the correct way to do this, or if other things are needed... :)

The problem is that they are not going to be exactly the same voltage: In practice one will hog all the current and get hot while the others twiddle their electronic thuibs waiting for the voltage to drop below THEIR regulated output...

Far better to use a bit of heatsink and a single regulator.

#10 hul Jul 18, 2005 09:27 PM

I use one of these to supply the retract servo. The rest of the system gets its power from elsewhere (SBEC in my case, could be UBEC, conventional BEC in the speed controller or RX pack).
This way I don't loose power to the flight controls if the retract servo stalls due to bent linkage or whatever. The voltage regulator shuts down on overcurrent or overtemperature and protects the servo, wiring etc. Heat dissipation is not an issue in this application, occasional short time use only.

Hans

#11 Martyn McKinney Jul 18, 2005 09:47 PM

Clever !

#12 THUREN Jul 19, 2005 01:14 PM

Well, I ordered up 5 of these for testing.... :D


After going over the spec PDF for a bit it looks like this regulator has some pretty good power handling capacity. At 12.5 volts it can still handle 5.5 amps before shutting down. After that it drops fast to about 3amps at 15 volts but as long as I can keep it cool, it seems to do pretty good... I'll do some testing and report back... :)

#13 Kulamata Jul 19, 2005 01:53 PM

Yes, cooling, with a light weight aluminum heat sink, and some thermal compound to ensure good thermal contact will help a lot. An easier solution might be to use mutlples, but to individual, isolated loads; an extension of the idea above. Wiring them in parallel is NOT a good idea, for the reasons given above.

I'd suggest giving a good thermal margin as well. Try to stay at least 25 degrees C below the data sheet max at the most extreme condition.

I'd also suggest using a good grade of electrolytic capacitor for the input and output bypass capacitors; 85C and rated for use in switching supplies, as there will be some fairly fast edges in the system.

Digi-Key is a good source.

#14 THUREN Jul 26, 2005 02:24 PM

OK.... Some very interesting, very good, info... :D

Test mule is my 3D plane, 4 cheap 9g servos, Nippy Black 2510/114, 2200mah 3 cell lipo with a charge to 12.6volts, brushless ESC with the red wire pulled.... What I did was high-temp epoxy one regulator to an approximately 1"x1"x.050" piece of aluminum, wire it into the battery side of the esc, and power the Rx through and amp meter.... I put a little tape on the hinges to simulate drag on the servos, and with no airflow, went crazy with the controls. I could never get the amp meter to read over 1.5 amps unless I REALLY yanked on the controls and these were only short bursts. The regulator/heat sink never got too hot to touch and keep in mind that the cut off temp is 250deg. F.... :D

I'm pretty happy so far....

#15 matttay Jul 26, 2005 11:55 PM

Just curious, but what was the average current just sitting there and with the servos being lightly exercised?


All times are GMT -5. The time now is 12:52 AM.