1

I need a clean install of an iPod charger in my car. That means not a wire running from a 12 volt socket. I've looked at the Belkin flush chargers and they are beautiful, but require a socket.

I'm considering wiring one to a panel mount USB adaptor using a DE-SW050 voltage regulator. It'll take up to 30 volts (so 12 is trivial) and outputs 5 volts at 1amp max (iPad compatible). The wiring seems trivial: An in, and out, and a common ground. With a breakout board and the Cables to Go panel mount, (and a fuse in line with the 12 volts) this seems like a trivial solution.

Is it viable and device-safe? It seems like such a clean design that there must be a catch.

The other question I had was if it was possible to then add an LED to indicate that the 5 volts was being used. (Putting it in-line perhaps? But what about voltage drop?)

Help would be much appreciated.

Chris Hanson
  • 131
  • 4
Scott Anguish
  • 11
  • 1
  • That looks a costly device - surely something like an LM317T and a couple of resistors and capacitors would do the same job at a fraction of the price? – Majenko May 28 '11 at 22:18

2 Answers2

2

The automotive environment is electrically very noisy, with potentially large spikes on the 12 V power bus (especially during engine start). This can be mitigated through the use of large filtering capacitors, maybe an inductor inline on the supply, and perhaps a transorb or other protection device. I don't know enough about power supply filtering to tell you exactly what to do, but you might be able to Google "automotive power supply filter" or similar. Note that the regulator may require input and output bypass capacitors, too.

Then there's the issue of iPad fast charging. I'm not sure how it decides it can draw more current for charging than what is normally allowed by the USB spec. I think there are resistors of a specific value between certain pins on the supply side. If you don't care about fast charging (which I've heard the iPhone 4 also uses), then you can ignore this.

You don't want to put an LED inline in order to see that power is being drawn. An LED has a maximum forward current above which its lifespan is dramatically shortened. This forward voltage is typically 5 - 25 mA, far less than the 1000 mA your regulator is capable of delivering (and far less than the 500 mA the iPod will likely draw). An alternative would be to set up a high-side current sense resistor (inline on the +5V side) and high-side current sense amplifier, use the output of that to drive a MOSFET or other transistor to turn on an LED. There are probably other ways to do that, but that's what comes to mind.

It really is easier to buy an off-the-shelf USB charger with a suitable input voltage, and cut off the supply connector to wire it in (with a fuse!) to your car supply. As another commenter mentions later, it may still lack appropriate filtering. You should be able to buy an automotive power filter to put inline with the USB charger. Then plug the Cables to Go panel mount into the charger.

Rick M.
  • 21
  • 2
  • Ugh, that seems like way over my head. The device takes up to 30volts. Do the spikes likely get that high? The spec sheet seems pretty much in/out voltages. 1amp is the max that MacBookPro USB puts out. But this seems to indicate that this isn't going to be as simple as it would seam. Thanks sir! – Scott Anguish May 28 '11 at 06:52
  • Well, it's not that the spikes would damage the regulator, it's that it wouldn't react fast enough to clamp them on the output. They might "get through" to the connected device. Moreover, the power is noisy, which might introduce noise on the analog outputs or erratic behavior in the device. Erm...see my updated answer. – Rick M. May 28 '11 at 07:10
  • 2
    @Scott Automotive regulators for 12V vehicles have to be rated to handle up to 60V for t < 100ms and 28V continuously (i.e. double battery jump start.) Often, negative spikes are tolerated too. And yes, 60V spikes are not uncommon in even small vehicles. They are called load dumps and they can occur when you switch lights on or off, but also any number of other power hungry systems can cause spikes due to the high inductance of the alternator. – Thomas O May 28 '11 at 10:20
1

You're going to need voltage divider on the USB data lines as explained in this answer. This controls the charging modes for iDevics and is needed for them to start charging.

Next, the datasheet says that there are already integrated capacitors in the regulator, but I'd still recommend that you add some extra ones at the output.

Another thing worth pointing out is the derating figure in the datasheet. Temperatures inside of the vehicle can get high, especially if you're going to place the regulator in an area with low ventilation. If you want to run it at 1 A, do take into account the graph. To me it seems that the maximum safe outside temperature would be around 35 degrees. You need to be sure that you have lower temperature when you're using the charger. Another problem is the unusual package. To me it looks like there's no easy way to add a heatsink to the device.

Community
  • 1
AndrejaKo
  • 23,261
  • 25
  • 110
  • 186
  • Thanks. It seems more and more attractive to just solder directly to a third party completed unit and tap into the wires on the 12 volt socket to power it. :-( Now I wish I hadn't just ordered the damn thing. – Scott Anguish May 28 '11 at 06:56
  • @Scott Anguish Well, it it will make you feel better, most of the cheaper car chargers I've seen don't have much (or even any) protections Rick M. recommended. Still, if you can, do add them. Most of the iDevices are very expensive and you don't want to risk them on a cheap car charger. Capacitors should be cheap and easy to obtain, but make sure you use ones rated at 35 V or higher, just to be safe. The datasheet for regulator recommends 470 $\mu F$ on sensitive output. – AndrejaKo May 28 '11 at 07:03