ATMega128 Overheating .. Rectifier at fault?

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello everyone, I would like to ask a couple of questions bout rectifiers (which i hate) and how it heats up my Atmega128 (which i love)

My story starts with #6, if you dont know what #6 is I do suggest you check out their site over [http://blunt.media.mit.edu:8080/6/], but it is basically an MIT group that did a re programmable and easily usable ATMega32 broad. They used a rectifier, and two capacitors a 47uf capacitor and a 4.7uf capacitor (assume the u stands for micro.)

Now after I designed the first board which contained an ATMega128 with a different purpose then the #6, to light green and red based on certain inputs all worked fine but I found the ATMega128 to be heating up, I found out that the voltage coming out from my 5V regulator (tried several regulators) was more than 5V, tracing the problem i ended up removing the Rectifier since I already just input a DC Voltage to my board, so why use the Rectifier right?

I ended up making four board designs all the same but without the Rectifier and came up with those results:
Board#1 Worked onspot, a bit fishy sometimes bas stable in the end.
Board#2 Didn't work so as I was debugging it I placed an AmpMeter between the Regulator and the 5V line of my board and it suddenly worked! I removed the AmpMeter and reconnected the lines to themselves and the board wouldn't work again. I was baffled, I know the AmpMeter has some resistance but it should be negligible resistance no? and this resistance shouldn't be what decides if the board functions or not.

Needless to say I got the same results on Boards2,3,4. So the rectifier came back into the equation, all though am not sure why, but when I place it back into the circuit it works just fine again (expect it overheats) I don't understand why the rectifier is the determining factor even though I am not using any AC input. I tired to lower the capacitor values as I read online the bigger the capacitor the smaller the ripples but the higher the current, still to no avail.

My Circuit takes around 250mA max while this rectifier can hold up to 1A, the only problem as I see it is that it is letting out more than 5V on the Regulator pin which most probably the reason, as I tired with a 6V input which produced a 4.32V output at the regulator, and then it was fine, but I need a 12V input.

I attached the graph of how the Regulator/Rectifier are connected, the #6Board (Eagle Cad) if anyone needs it for any reason and the Rectifier datasheet mine is 107[/url]

If you could take out a minute or two to help me I'd appreciate it, I am still pondering why the board doesn't work without a rectifier even though I am not using any AC components (Capacitors, MAX232, RS232, ATMega128) and in addition I built the #6 board myself and it doesn't heat up nor anything which I find to be very weird. Is there a way to work it without having a rectifier? or is there a way to keep my regulator output at bay with 5V? Thanks in advance

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The schematic looks to be OK, so the implementation may be the problem.

A picture (photo) of your board would be a good idea, so please post one (or more)

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tricia, and Ulyana. You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

But also be aware that your regulator has a maximum *power* dissipation as well as a maximum current capability. If you're feeding it 12v and pulling 250mA at 5v, you're asking it to dissipate (12-5)*.25 = 1.75W.

Depending on the package and mounting, this may or may not make it unhappy. Most will shut down in the event of overheating, but not all.

However, that said... the AVR will generally take up to 5.5 volts without problems but should of itself only consume a few milliamps. Is it possible that you're pulling excessive current through the ports, for example? Perhaps driving the LEDs directly and exceeding the total current capability of the device even if indidual pins are not overloaded?

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Neil, if the 7805 runs hot, it will shut down. OOps, just re-read:

Quote:
Most will shut down in the event of overheating, but not all.
Leaves us with over-voltage. Hmmmm ...

OP mentiones warm M128, 6V Vcc and troubles with the rectifier. My guestimate: wrong wiring / short on the board, groundloop via supply, PC and/or scope (or anything else)

I misplaced a 9V wire once on a pin of the M32 (radio-thingie), and everything worked .... except Vcc was surprisingly high ... M32 survived. Strong chippies, these AVR's :)

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tricia, and Ulyana. You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It looks to me that there is a problem with the traces on the PCB. one thing to check is a bridge between one of te AC lines and the gnd output of the bridge.

One question came to mind is the bridge rectifier made up of a group of diodes or a one piece device?

The one piece units are marked with ac/dc, but it is possible in a discrete bridge to reverse something.

All the best on your quest
Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello everyone, well thanks to your feedback I took out the rectifier but kept the capacitors (noobish mistake to remove them in the first place), so the circuit works fine, whats left is that the regulator is heating up. I will attach a heat sink to it, but so far it seems to me to be too hot for comfort, but since you guys stated it should switch off if it runs too high that is a sign of relief, but I do have some questions about what was posted above, mainly to Neil;

Quote:
Depending on the package and mounting, this may or may not make it unhappy.

Did you mean in enclosed surfaces? or something techwise?

in addition to

Quote:

Perhaps driving the LEDs directly and exceeding the total current capability of the device even if individual pins are not overloaded?

The circuit as a whole, tested at the first 5V pin that powers the circuit consumes 310mA much less then the 1A which the regulator can handle.

Finally, I don't believe there to be a short in the circuit as I have tested it many times, and the main program on the AVR runs correctly, all components respond correctly, All what is left now is to attach a heat sink and leave it over night in hopes it wont be destroyed in the morning!

Thankfully Atmels AVR were made with people like me in mind, they can handle it! :D

Edit: am also searching online for power dissipation resisters and techniques for safety's sake, thanks

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

In order: any regulator chip has a maximum input voltage, a maximum output current, and a maximum dissipation. Generally, exceeding or trying to exceed any of these parameters will cause it to overheat. Most regulators are designed to shut down if the overheating gets too much but that's not always guaranteed. The data sheet will show the permitted dissipation for the regulator you have, and how it changes depending on the surrounding temperature, whether you have a heat sink, or whatever.

The second point: each output pin on the AVR has a current which it, on its own, can supply. Again, check the data sheet for details. However - another datasheet fact - the package can only supply a certain amount of current before its internal connections melt. It's possible that this figure is somewhat less than the total per pin multiplied by the number of pins. For example, on the ATMega32, an output pin can source 40mA; but the package as a whole can supply only 200mA. That means that if you try and drive more than 200mA from the output pins (source or sink) then the package limit will be exceeded *even though* you have not exceeded any individual pin. This is something to watch.

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Toufic86.

Above you have stated an input supply of 12 volts and a current of 310 mA out of the 5V regulator. That means that in the absence of the bridge rectifier that you are expecting the 7805T (plastic case) regulator to handle (12-5)*0.31 = 2.17 watts of heat. It will probably not do it in the absence of a suitable heatsink...and maybe not even then (depending upon the 7805 package, the heatsink and adequate exposure to air). Refer to the datasheet for maximum power dissipation performance graph.

Just because the 7805 is rated at 1 amp does not mean that it can deliver 1 amp under all circumstances. The thermal overload protection will cut in if you exceed its heat rating. There is a minimum input voltage required to allow the device to correctly regulate (usually around 2.5 volts), so using an input of say 8 volts and your 0.31 amp load would produce a dissipated power across the 7805T of a smidgeon under 1 watt. Definitely "doable".

The long and short of it is that your power supply design "sucks" in its present form. Sorry. If you must operate from 12 volts to produce a 5 volt output, I would suggest that you consider a switching power supply that will withstand the high input voltage with a much smaller heat loss.

And...by the way, the rectifier was probably there to prevent the design from being destroyed by someone accidently connecting the input leads "arse about". Still a good idea IMNSHO.

Cheers,

Ross McKenzie ValuSoft Melbourne Australia