Mega2560 board, IC7 gets very hot?

Hello everybody!
I don't know what could have happened, board has been working alright for some time, but now i noticed that the IC7 get really hot? What can cause that? There is nothing connected to the board. I measured the current board draws, it is about 400mA. Does anyone have a suggestion what could cause the problem?

IC7 controls two things.

  1. The pin 13 LED
  2. The FET T1

T1 controls the connection between the usb power bus and the 5v bus on the main board. If Vin is less than 6.6v, the usb and the main power bus are connected to power the Arduino from usb power.

The Mega2560 has problems when powered by the Vin (barrel jack) and the usb simultaneously. If you had your Mega connected to usb and external power applied to Vin, it could have damaged the board. Read this topic about that:
http://arduino.cc/forum/index.php/topic,82046.0.html

SurferTim:
IC7 controls two things.

  1. The pin 13 LED
  2. The FET T1

T1 controls the connection between the usb power bus and the 5v bus on the main board. If Vin is less than 6.6v, the usb and the main power bus are connected to power the Arduino from usb power.

The Mega2560 has problems when powered by the Vin (barrel jack) and the usb simultaneously. If you had your Mega connected to usb and external power applied to Vin, it could have damaged the board. Read this topic about that:
http://arduino.cc/forum/index.php/topic,82046.0.html

ST;

You have totally misunderstood the problem stated in that linked topic and have now posted a untruth about how one can damage their Mega board that is sure to at best confuse beginners about what and how they can power their mega boards.

There is absolutely no problem with using external DC power in the correct specified voltage range (7-12) to the arduino external power connector (barrel jack or via the Vin pin) and also have the USB plugged into a PC at the same time. That is perfectly safe and in that case the board will be powered via it's on-board +5vdc regulator and the USB +5vdc power will be cut off via the mosfet switch controlled by the auto-voltage selector circuit.

The danger the linked posting was referring to is when a user attempts to power his arduino board using an external regulated +5vdc power supply wired to the Arduino's 5V pin and ground pin. In that case when you also plug in the USB to a PC then you are effectively 'hardwiring' the USB's +5vdc and the external regulated +5vdc together and that in some cases that can cause problems, although many do it without having an issue.

So please tell you you now understand the issue more clearly and that your posted warning above about using Vin power and USB power at the same time is in error and it is in fact perfectly ok and is how the board was designed to be used?

Lefty

My bad. I understood that problem to include a voltage less than 6.6v applied to Vin also. If that is not the case, you have my apology.

edit: Since you put your response to me so eloquently, I thought you might have forgotten to mention to the OP why IC7 would get hot.

SurferTim:
My bad. I understood that problem to include a voltage less than 6.6v applied to Vin also. If that is not the case, you have my apology.

Well supplying Vin at less then 6.6v while also plugged into USB is operating outside of the boards stated specifications and requirements for the board, and indeed could cause problems. But your 'warning, danger, danger' did not couch it to this specific case but rather stated a global " has problems when powered by the Vin (barrel jack) and the usb simultaneously." which is not the case when supplied with proper Vin voltage clearly stated.

edit: Since you put your response to me so eloquently, I thought you might have forgotten to mention to the OP why IC7 would get hot.

I don't have a clue. IC7 is an op-amp and should not ever run hot unless it's defective or something on it's output pins is shorted to ground or Vcc. As the normal board current runs at less then 100ma, something inside or wired to IC7 is drawing 200+ ma of current that it should not be. Maybe the mosfet switch has a shorted gate and is applying +5vdc back to the op-amp's output pin. It could be troubleshoot with the proper methods and test equipment, but via a internet forum not so easy.
Lefty

SurferTim:
The Mega2560 has problems when powered by the Vin (barrel jack) and the usb simultaneously. If you had your Mega connected to usb and external power applied to Vin, it could have damaged the board.

I usually don't quote myself, but I do not see this as "WARNING DANGER DANGER". I always like your input tho. Sometimes in my haste, I forget a detail that could be important.

SurferTim:

SurferTim:
The Mega2560 has problems when powered by the Vin (barrel jack) and the usb simultaneously. If you had your Mega connected to usb and external power applied to Vin, it could have damaged the board.

I usually don't quote myself, but I do not see this as "WARNING DANGER DANGER". I always like your input tho. Sometimes in my haste, I forget a detail that could be important.

Yes I do go over the top sometimes, I'm sorry, but it's not my fault really, I just sometimes frizzle at the warnings and cautions given out to beginners here sometimes that are not fact based. They have enough on their plates as it is learning to use their arduino board properly and learning the electronics they need to so as to not cause self inflicted damage.

I have no negative issues with you, just with the statement you made this one time. :wink:

Lefty

@retrolefty: Now that I seem to have your attention on this, do you see a problem with IC7 if Vin > 10v? That would effectively overdrive the Vin input to IC7, would it not? Could that cause a latchup condition on the output?

Just trying to figure out why it gets hot.

SurferTim:
@retrolefty: Now that I seem to have your attention on this, do you see a problem with IC7 if Vin > 10v? That would effectively overdrive the Vin input to IC7, would it not? Could that cause a latchup condition on the output?

Just trying to figure out why it gets hot.

The op-amp input pin is wired to a voltage divider the cuts Vin in half to the pin. I don't know what the input pin voltage limit is for that specific op-amp, anyone have a datasheet link to it? But as the Vin limits are rated as:

Input Voltage (limits) 6-20V

That means the input pin of the op-amp is designed to work at up to 10v which is well above the Vcc the op-amp is powered by, so obviously the op-amp is designed to handle input pin voltage higher then it's running Vcc is. Again a datasheet is needed to see what the true voltage limit for the op-amp is. I think the op-amp specification we need to know is called it's input common mode rejection voltage rating?

Lefty

Datasheet.

Max voltage any input = 5.5v
That means 11v is the upper limit to Vin without overdriving the op-amp. :frowning:

edit: Seems like you could fix that with one well-placed zener diode. :wink:

SurferTim:
Datasheet.
http://www.ti.com/lit/ds/symlink/lmv358.pdf
Max voltage any input = 5.5v
That means 11v is the upper limit to Vin without overdriving the op-amp. :frowning:

Well the op-amp has internal clamping diodes applied to the input pins to prevent the pin voltage from exceeding Vcc + diode voltage drop, so hence the 5.5vdc rating. However as the current flow is limited by the resistance of the voltage divider the 'clamping diode fault current' (what .4 ma at vin=20v?) thus generated may be fine working in this matter. That is not unlike how much higher voltages then Vcc can be applied to a AVR input pin if it's in wired in series with a high enough resistance to limit the diode clamping current to a safe value. However, I can't find a specification in that datasheet that states the maximum safe input clamping diode forward current rating. However I feet it's pretty safe that running Vin up to the 20v max rating does not subject the op-amp to any damage because of a combination of the internal clamping diode and the series resistance of the voltage divider. I think clamping current at under 1ma is most likely 'safe' for the op-amp.

Anyone know more about this?

Edit: A on-semiconductor datasheet for this device type gives a specification called

Maximum Input Current 10 mA

Which would mean that .4ma of clamping diode current when Vin is at 20vdc or less would be fine.

Lefty

I like your datasheet better. This is what I was concerned about, and it is mentioned as one of the main features:
No Output Phase?Reversal from Overdriven Input

Sounds like it is on its way to "transistor heaven"?

SurferTim:
I like your datasheet better. This is what I was concerned about, and it is mentioned as one of the main features:
No Output Phase?Reversal from Overdriven Input

Sounds like it is on its way to "transistor heaven"?

How so? I thought I made the case of inputting voltage higher then Vcc to the chip is fine as long as you limit the input current externally to a safe value.

Lefty

My bad again. What I meant was the OP is just unlucky. The IC7 failed because the transistor gods are angry with him. IC7 was probably a faulty part to begin with. Do you have a problem with that?

Did anyone ask OP how he was powering the board and measuring the current? 6V wallwart?
12V wallwart? Did the 5V v.reg get hot too? My Mega board draws about 90 mA with no loading
on the headers.

Hello all!
I haven't been able to participate on this for a while.
First of all, board has been powered mostly from USB, sometimes with a lab power supply, which has a variable voltage, but I've been using it with 12V.
I found the schematic for the board(on the product page, what a surprise :roll_eyes: ), and have been doing some trouble shooting with it, and it looks like the IC7 just has gone bad for some reason. The FET it drives is ok and working fine, and the resistor and LED are fine ok, led lights up when driven externally. The 3.3V regulator is also working as expected, which really shouldn't be an issue, because it's behind the FET but checked it anyways.
For now, I just removed the IC7 from the board, and have to live without that 3.3V when on USB power. Maybe I'll get a replacement IC for it later.

Hi,

Apparently i have done the same mistake with UNO Rev3. Powered through USB port and gave 5v input in Vin simultaneously.
Now the op amp VCC and Ground pins are shorted.
I could not upload any code and the led connected to pin13 is always on.

Is there anyway i can make my UNO work? Would it be fine if i just replace the component.

Any comments would be helpful.