Go Down

Topic: Problem with external power supply to arduino (Read 5747 times) previous topic - next topic

harsha271094

My arduino uno works perfectly when powered through the USB cable.Recently 2-3 weeks back the arduino is not responding when powered through 3.5mm power cord!!!!! I thought that something is wrong with my cable ,and instead started powering through Vin pin .It worked fine till yesterday!!!! Suddenly arduino stopped responding to Vin pin also.i observed that the ON (Green colour led indicator) is glowing very very dim.Then i checked the voltages at the regulator and Vin pin.When i supply 12 V(or even 9 V) ,input voltage at the regulaotr(7805)is showing around 3-4V and voltage at output is showing  0.3(under 1v).It still works fine with USB.
     I saw the same question in another forum!! but that solution didnt help me!!!!!!
so PLZ..PLZ....PLZ.....suggest me what to do!! i am depending on the arduino for my project!!!
Thanx in advance!!  :)

retrolefty

What is the voltage and current ratings of the power source your using for external power?


harsha271094

12v(1.2A) and 9v(500-700mA) battery.
i worked on a project with these two power sources  and they worked fine!!!
Now my arduino is not allowing this power !!!

virtual1

On the power end of things, get out a multimeter and put it on DC volt check and make sure it's actually delivering voltage to the plug.  Maybe a loose battery, loose/corroded/rusted cell interconnect, broken wire, bad end, or wrong polarity. (center positive)  I've seen all of those, more than once ;)

If that's good, check the solder points on the jack on the arduino on the solder side.  If the soldering was poor, or the jack jarred physically, it can crack away the pin from the solder and make an intermittent.

If that looks good, plug in your power and then use the meter to check the pins on the bottom of the plug on the arduino to make sure it's making it all the way to the board.  Pushing the plug into the jack on the arduino will cause it to reroute power on the board, but you should be able to find +9 between two of the pins on the jack.

if you have power there but the arduino still isn't powering up, but works from the computer, there's probably an electrical problem with the board, such as a failed voltage regulator.

harsha271094

Checked at the bottom soldered part!!!
Battery normal voltage-8.5V.After connected to Vin pin or the power cord.voltage of battery falls to 6.5 v.(at bottom soldered part).and volage at the regulator is falling to 3 V and output being 0.7 V!!

harsha271094

Ohk!!!! I am sure some part of my arduino has malfunctioned at the power lines....But dont know which part exactly!!!
The voltage drops abruptly at this Vin pin.Can anyone tell me which part might have gone wrong?? and can we replace that part even if we found that malfunctioned part!!! :)

cartoonist

#6
Feb 03, 2014, 10:10 pm Last Edit: Feb 06, 2014, 11:51 pm by cartoonist Reason: 1
The voltage regulator on the Arduino is defenitly NOT a 7805. The difference is that the centerpin is ground for a 7805 and is output on the regulators used for Arduino. Furthermore the 7805 has about 3 Volt dropout from in to out.
Arduino uses low-drop 5V regulator,means less than 1.5 volt dropout between in and out.

On the Arduino V2 is a MC33269-5G , a 800mA 5V regulator, often marked as a 269-5G.
datasheet: http://www.onsemi.com/pub_link/Collateral/MC33269-D.PDF

On the Arduino V3 is a NPC1117 ST50T3G , this is a 1000 mA 5V regulator, often marked as AMS1117.
datasheet http://www.advanced-monolithic.com/pdf/ds1117.pdf

Most likely your regulator has died
because you used a 12 volt powersupply for extended periods of time and probably feeding secundary circuits and/or components from the 5V Arduino rail. The 5 V regulator gets very hot first and finally dies out because of temperature stress.
Always better to use a 7.5 to 9 volt 1 Ampere powersupply to feed your Arduino. Offcourse your Arduino CAN be fed with up to 18 Volt or so without harm, but not for extended periods of time because the regulator will overheat soon.

(I striked this part because the conclusion is rather speculative.)

Replacement of the voltage regulator can be done if you have some soldering skills.
You do not need a new P.C., you need a new O.S.  Linux is free, safe, easy, fast and reliable.

retrolefty

Quote

Most likely your regulator has died because you used a 12 volt powersupply for extended periods of time and probably feeding secundary circuits and/or components from the 5V Arduino rail. The 5 V regulator gets very hot first and finally dies out because of temperature stress.


That would be unlike any linear regulator datasheet I've ever reviewed. All to my knowledge state that they have automatic shutdown protection for over current and over temperature conditions. I don't know why this myth prevails to this day?
However over voltage input can be a fatal failure as well as reverse current conditions.


cartoonist

#8
Feb 04, 2014, 12:15 am Last Edit: Feb 04, 2014, 01:14 am by cartoonist Reason: 1
I agree with you that all regulators have protection mechanisms.

The AMS1117 datasheet says under 'application hints' :
Quote
Thermal protection circuitry will shut-down the regulator should the junction temperature exceed 165°C at the sense point.


I myself had very recently a AMS1117 stop working and certainly not due to overvoltage, it was fed by a switching 12 volt supply, the current was too high for a long period and the regulator died with a small poof and some smoke.

So I don't rely my precious electronics circuits only on the protection mechanisms of these little things.  I will Just be safe and avoid overheating.
Thermal stress makes these things collapse for mechanical/technical reasons, not for electric reasons.
You do not need a new P.C., you need a new O.S.  Linux is free, safe, easy, fast and reliable.

cartoonist

If I look at the schematics of the Arduino, specially the part between the V-in pin and the USB-vcc.
I see only 2 parts inbetween these points.
A voltage regulator and a 47uF 25V capacitor.
Considering the description of the problem,  I concluded that the regulator is most likely defective.
You do not need a new P.C., you need a new O.S.  Linux is free, safe, easy, fast and reliable.

virtual1

Quote
Thermal protection circuitry will shut-down the regulator should the junction temperature exceed 165°C at the sense point.


I myself had very recently a AMS1117 stop working and certainly not due to overvoltage, it was fed by a switching 12 volt supply, the current was too high for a long period and the regulator died with a small poof and some smoke.


I suspect the protection in those regulators is meant more for instantaneous over-current than for overheat.  So if it's rated 500mA continuous, 2A surge, the safety trips at maybe 2.3A, but there's probably *nothing* in the safety to watch for 1A continuous causing an overheating package.  And after awhile, out comes the Magic Smoke.

I've seen my arduinos get quite warm after awhile, at the regulator.  I try not to draw too much from them.  The same thing can happen to your ATI chip, it may say "rated 50mA continuous, self-resetting", but it probably won't end well if you leave ten pins cooking at 40mA all morning.

retrolefty

Quote

I suspect the protection in those regulators is meant more for instantaneous over-current than for overheat.  So if it's rated 500mA continuous, 2A surge, the safety trips at maybe 2.3A, but there's probably *nothing* in the safety to watch for 1A continuous causing an overheating package.  And after awhile, out comes the Magic Smoke.


No. If you read the datasheets and review the block diagram of the linear regulator you should find that the two protection modes are independent of each other as far as being able to shutdown the device. For sure there is interaction as to what causes either protection mode to be activated, there is nothing to prevent either condition from doing it's designed function. It's a OR condition.

Lefty

cartoonist

@ Lefty

It is true that an overcurrent or shortcircuit will not damage a voltage regulator like the AMS1117 very easy.
A user that will accidentally apply an overvoltage or uses a low quality PSU that will pass on an input voltage spike to the output and so to the regulator is probably ofter a cause of failure than thermal damage.

But often datasheets do not tell all the specs, f.i. it does not mention how many thermal cycles a device can have before it goes out of specs, after which a technical failure of the device can happen.  A shortcircuit or overcurrent can cause a repetive shutdown/reset  that cause repetive mechanical strain inside the device. Other mechanisms can also make the device go out of specs. So when a fault condition lasts long enough ................... defective regulators result.

But my supposition that harsha271094 had a defective regulator due to overcurrent/overheating was premature. He did not mention anything about the current just that he had used a 12 volt PSU.

Therefor I edited my earlier contribution. It was not my intention to fear-mongering Arduino users who have a regulator that gets hot sometimes.
You do not need a new P.C., you need a new O.S.  Linux is free, safe, easy, fast and reliable.

retrolefty


Go Up