Arduino voltage regulator question

I'm looking at powering my arduino from a 12v battery. The data sheet for the onboard MC33269D-5.0 voltage regulator shows a max input voltage of 20v. When I power the board from the 12v battery instead of the USB power, the voltage measured at the +5v pin on the board goes from ~4.4v to ~7v. This is somewhat concerning in that I assume the board is then operating at the ~7v. The board doesn't seem to be harmed yet, but the voltage being different dosn't support my origional ideas. Anybody have ideas on what is going on?

Below are the board specs. The 12v is into the barrel connector. The arduino has an ethernet shield attached. Removed the shield and not much changed as far as using the barrel connector. The ethernet shield is power hungry. Using USB powerwith no shield, the 5v point measures 4.88v, the same as what is measured at the hub where the arduino is connected. Attach the ethernet shield and the voltage at the 5v point drops to 4.25v. Guess I'll use a 7805 regulator chip to make a 5v supply and feed it into the USB port when battery use is required. The voltage regulation thru the barrel connector seems to be totally unpredictable.

DFROBOT DFRduino Duemilanove www.dfrobot.com

Summary

Microcontroller ATmega328 Operating Voltage 5V Input Voltage (recommended) 7-12V Input Voltage (limits) 6-20V Digital I/O Pins 14 (of which 6 provide PWM output) Analog Input Pins 6 DC Current per I/O Pin 40 mA DC Current for 3.3V Pin 50 mA Flash Memory 32 KB of which 2KB used by bootloader SRAM 2 KB EEPROM 1 KB Clock Speed 16 MHz

If you supply the ATmega with 7V you could damage it. Check your multimeter that its properly calibrated because it could very well be 5V there and your multimeter or its leads are causing the wrong measurement. Check it with something else.

The regulator should output exactly 5V, no more no less (unless there is a short circuit which can pull it down a bit).

That said the regulator being used can't handle that much current, so maybe using a 7805 through the USB connection would be better.

Looking at the parts on the board, there is a L53B 5v voltage regulator (datasheet below) on the board. I'm checking voltage on the lower pin header (looking down, barrel connector to the left). The pins are from left to right reset-3v3-5v-G-G-Vin (Vin labled 9v on ethernet shield) I'm checking the voltage between the 5v pin and adjacent ground pin. I checked the the multimeter against two others and the readings are consistant. The board and ethernet shield are working ok when connected to the USB port. I'll just have to keep tinkering. When connecting the 12v battery, 5.1v is momentarly indicated at the 5v pin, but this changes to ~7v. A 9v battery settles out at ~6.5v. Tried some other stuff, but the results so far are unpredictable.

http://www.datasheetarchive.com/pdf/Datasheet-028/DSA00500725.pdf

Are you saying that all three boards behave this way?

Sorry for shortcuting my statement by using "two others" instead of "two other multimeters". I checked the voltage on the single arduino board using two other mutimeters, as it was suggested that the origional multimeter might have an issue. And yes, I do have a reasonable understanding of how voltage regulators are supposed to work, hense my concerns about what I am seeing. I've read the other post concerning the "auto swap" of the power source when a measured voltage reaches ~6.6v. I'll have to do some more trouble shooting, and develop some type of workaround for the issue short term.

there is a L53B 5v voltage regulator (datasheet below) on the board.

Hmm. My impression was that the DFRduino was a "faithful" reproduction of the Arduino duemilanove PCB, rather than one that had had any re-engineering applied. If that's the case, then the lm2940 regulator you reference is NOT a correct replacement for the mc33269D regulator used on the original duemilanove!!!! They have different pinouts!

(Check whether the tab of the regulator has continuity with the metal casing of the USB port. If the board is properly laid out for lm2940, these should both be GND. If the PCB is laid out for MC33269D, then the regulator tab will be +5V (vout of the regulator) while the casing would still be GND.)

I'll have to check the board with my meter. As I don't have a good picture or layout of the origional board, the below is what I suspect to be the voltage regulator on the board (located to the right of the barrel connector). To the lower left appears to be an SS14 diode connected to the barrel connector center pin.

I can't see where the traces go to tell whether the board has been modified to use the 2940 pinout. The other components and the relative placement can all stay the same even if you change the regulator and traces to match. For example, the "freeduino" design switched from a 33269-based design to 7805-based at quite a late time, without needing to change much else.

A quick check seems to indicate that the chip is probably properly connected/wired on the board. When I connect a 9v battery to the barrel connector I get 6.5v at the 5v pin. I wonder if others who have purchased the DFRduino boards have actually checked the voltage supplied to the chip when connecting to an external power source? With the apparent 7v being supplied to the 5v header, my board did not immediately die, but I was also planning to use the 5v to power my ssc-32 servo controller.

did you mistype that first sentence?

did you mistype that first sentence?

The sentance is typed as intended. I consider a board trace equivalent to a "wire" if that caused an interpertation problem. The large pad to which the L53B chip is soldered appears to be part of the board ground system.

Ok... Rats! it was such a good theory!

Sometimes I just use terms that I’m use to, such as a “sketch” is just code and a “shield” is just an adapter. Too many new age tootie fruitie terms floating about to keep up with everybody’s “widget”. :slight_smile: