Switching BEC Test Results - RC-Monster Forums

BrianG · 03:14 AM

I picked up a couple of the cheap (~$10) UBEC units from eBay. This is what I use in my Hyper 8 and it works well so far, but I decided to run some tests to see how well they really work.

First, some pics:

This unit is based on the LM2596 switching regulator IC and use a very minimum amount of parts. The design of these units are assembled just about verbatim from the example circuit in the datasheet (in the link above).

Now, some numbers. I hooked them up to a linear adjustable 12v power supply I made and loaded them with various resistors. Pay attention to the output voltage, output current, and efficiency ratings.

Typical of switchers, they tend to be most efficient when run closer to their rated output current. I did overload it (4.4A) to see how it would react and I was quite happy. The voltage sagged a little, but the AC ripple was pretty constant for about 5 seconds when the switching frequency dropped, AC ripple increased, and the unit started to get quite warm (~125*F). So, it can take short durations of current draw beyond its rating. Under normal conditions, the temperature stayed under 105*F. And AC voltage ripple was quite low at a nominal of 2.2mV.

I was a little disappointed to see the output voltage was not quite what it claimed. The 5v setting fell a little short, and the 6v setting fell at least 0.5v short.

Dropout voltage is a little high at about 1v. This means that your input voltage has to be at least 1v higher than the output voltage you want. Still, it will still work for 6 cells/2s lipo should you choose to do so.

Edited for spelling :)

GorillaMaxx360 · 04.05.2007, 09:43 PM

Nice Write Up Brian

BrianG · 04.05.2007, 09:51 PM

Thanks. :)

Just goes to show that just because something is inexpensive doesn't always mean it's cheap.

MetalMan · 04.05.2007, 10:35 PM

Nice test :). I've got one of those, but haven't been using it because it has caused radio interference (75mhz FM). It's possible that a part on it is dead, or my receiver just really sucks, though.

neweuser · 04.05.2007, 10:39 PM

I had run two of those and had the same interference problems with JR Radio I have, so I went to the ultimate ubec from Mike and have no issues now. But I would agree for lower voltage systems it would be ideal.

Serum · 04.06.2007, 03:08 AM

Nice test Brian, i've seen some other becs that are even cheaper, which look way better build. perhaps they hold the 6V better than these.

BrianG · 04.06.2007, 03:13 AM

Update to the test results: The overloaded test result is skewed. I forgot my linear supply is only good for ~2-2.5A before the pre-regulated voltage sags to a point that is under the regulated voltage, which will pass a lot of 120Hz AC ripple (120Hz because I'm using a bridge rectifier) riding on the DC voltage. So, that noise on the input can't be good for the switching BEC. I'll probably re-run that test with batteries instead.

The interference surprises me. I don't know about you newe, but MM's radio is 75MHz, which is far above the BEC's switching rate of ~150kHz - a factor of 500 to be exact. Must be something to do with harmonics or something. Even then, I only got ~2.2 millivolts of AC ripple.

If it were me with the interference, I'd shield it with tin foil or some type of metal braid and tie the shield to the ground wire. I guess another feather in the cap for Spektrum! :)

BrianG · 04.06.2007, 03:19 AM

What BEC are you talking about Serum? The ones I got were actually $7.50 each, but $10.50 after shipping from China. I think they are built well, it's just that the parts are crammed on there for room. There's only the IC, inductor, diode, resistor, and input/output caps. Pretty simple, which is why I like it.

If I had an extra Koolflight one, I'd test that as well. Although, I should pick up an EMF meter to measure radiated noise as well.

04.06.2007, 08:21 AM

Very nice data. Thanks for testing and posting this info. If you decide to test more BECs, I know there's demand for seeing this kind of test data. :)

What equipment are you using btw?

BrianG · 04.06.2007, 08:49 AM

Your welcome! :) I can test more styles/types if someone wants to send them to test.

As far as equipment, I'm just using several meters going at once to take the various measurements. For loads, I'm just using whatever resistors I have on hand that will draw the current I want to test. If I was going to do more of this type of thing, I'd use a more stable power source and probably make an active loading circuit so I could dial in any load I want (basically just a constant current source).

04.06.2007, 12:27 PM

I just got one of those, it came in a very slick hard plastic case, anyway it has no instructions. How to do you put in in 6V mode...with or without the little jumper?

Thanks,

DM

squeeforever · 04.06.2007, 12:40 PM

I believe with is 6V.

BrianG · 04.06.2007, 03:19 PM

Yup, Squee is right (as usual); jumper out=5v, jumper in=6v. However, a different model may work differently. Just measure with a voltmeter to be sure.

04.06.2007, 03:28 PM

Thank you guys.

BrianG · 04.06.2007, 04:17 PM

I was just looking at the datasheet for the regulator a little more. While the regulator can be used for a higher voltage than the IC's default output voltage, it'll be more stable at 5v. Really, if you want a different output, the adjustable version of the IC should be used. So, unless you absolutely need 6v, I'd just select the 5v setting. I think I'm going to mess with the feedback resistor to see if I can get a more solid 5v with a 3A load.