Go Down

Topic: Arduino circuit/breadboard not working at higher voltage? (Read 8072 times) previous topic - next topic

xl97

Feb 18, 2011, 05:45 pm Last Edit: Feb 18, 2011, 05:54 pm by xl97 Reason: 1
I am very new to embedded electronics for the most part..

I have an Arduino Duemilanove (2009?) dev board..

but followed some links and tuts to make an Arduino compatible circuit...

here is parts  tut I followed:

http://www.instructables.com/id/Perfboard-Hackduino-Arduino-compatible-circuit/

• 28-pin DIP IC Socket - $0.30 - buy mouser
• 16MHz crystal - $0.55 - buy mouser
• momentary push-button switch - $0.15 - buy mouser
• 1k ohm resistor - $0.05 - buy mouser
• LM7805 5v voltage regulator - $0.35 - buy mouser
• 2 x 22pF capacitors - $0.12 - buy mouser
• 10nF capacitor (ceramic disc code '103') - $0.10 - buy mouser
• 22uF capacitor - $0.02 - buy mouser
• ATMega168 or 328

Mouser parts/BOM list:

https://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=20ea42b1c5


I did make my own "FTDI cable" using an older phone data cable.. and the tuts on the net as well.. (tried to make a parallel and a MAX232 version...none worked)..
flashed my chips using my real Arduino.. using the hacked data cable to upload my sketches from computer/IDE to breadboard circuit.


this all seemed to work fine..

until I decided to re-create the breadboard circuit on a smaller breadboard from Radioshack..

copied circuit.. ..

I hooked up my PSU.. gave it around 7.2 volts.. as what I would normally use in my projects..

and it wouldnt fire/run the blink sketch?

I go back to my original breadboard..same results!..

after some tinkering..

I found that if I dialed down my PSU to about 4-5v.. it started working?

I am completely stumped and need help figuring out WHY?  Im not much of a hardware guy.. just following along.. (really just like to play with the code sided of things)

here is my circuit/breadboard.. I do 'not' think its my circuit..as once the PSU is dialed down... it starts working?






I grabbed my (cheapie) meter and tested the output after the vRegulator.. on the rails of the project.. its was around 5v or so..

I have no direction..sorry if this is newbie territory, but I have never encountered it..nor have any idea WHY this would be a problem?

I can only think its the vRegulator somehow?  but thought I could throw like 12-25v at it or something?

Im hoping/thinking.... the circuit and everything is really' fine.. its somehow tied to how I have the LED on the breadboard..that is giving false indication?

after a few more tests.. I can see now that when I adjust the voltage on my PSU as the circuit is running..it affect the leds powering on/off..
not sure how I go about fixing this? resistor the LED? (would this have to be a fixed resistor then to always prepare for a certain/static voltage?)


I crank down to 4.3 volts.. and the led blinks normally...red led).. Im surprised the Arduino runs at that? after the vRegulator?


Thanks for any feedback..  (please dumb it down.. so I can understand it!)  haha..







johnwasser

Sounds like strange voltage regulator behavior.  The rails should be at 5V almost regardless of the input voltage.  The regulator probably has some minimum voltage it will work at but 7.2 volts should be more then sufficient.
Send Bitcoin tips to: 1L3CTDoTgrXNA5WyF77uWqt4gUdye9mezN
Send Litecoin tips to : LVtpaq6JgJAZwvnVq3ftVeHafWkcpmuR1e

xl97

Yeah I thought faulty/odd vReg problems as well..but switched it out with another one.. same results?!!
Lacking much electronics/hardware experience.. I feel a bit 'dead in the water'..

Hoping I didnt damage anything...


I popped out both chips I have and put them in my Arduino board.. fired up..worked fine.. steady..

throw either of them in my two breadboard layouts.. same problem..

maybe the vRegulator I got in the list is wrong? (I didnt look at spec/data sheet)



if I turn up the PSU voltage.. while running.. I can see a visual difference in the led brightness... eventually stopping the blinking..

hmm.. anybody?




retrolefty

Quote
here is my circuit/breadboard.. I do 'not' think its my circuit..as once the PSU is dialed down... it starts working?


It's your circuit design. Look at a datasheet for your specific voltage regulator for typical applications, it will show two capacitors wired from ground to both the input and output terminals of the regulator. Your regulator is most likely oscillating, but your meter won't show that as a oscilloscope is required to see it. Anyway find and wire in the recommended regulator capacitors and I bet all will be fine.

Lefty


CrossRoads

Why didn't you install the 2 other caps?
And put some caps on the input & output of the regulator.
You should really have a current limiting resistor in series with the LED - you're going to damage the LED or damage the output pin wired up as it is.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

xl97

#6
Feb 18, 2011, 10:08 pm Last Edit: Feb 18, 2011, 10:13 pm by xl97 Reason: 1
thanks for the reply guys..

as I said Im new.  so the 'why' questions should be self-evdident.  =)

just dont have the knowledge/experience to know better/different.

ok.. so I will go back and add in the

* 10nF cap (103) in front of the vRegulator (across the GND and OUTPUT)  BEFORE the  power +/- lines reach it.

* 22uF cap across the 'rails' on my breadboard.. (after the GND/output from the vRegulator)


@CrossRoads-

I guess Im more confused by your comments now?

I see/added in the two missing caps..

but there are two more?  seems like I have 4 missing components from your comments

also this:
"You should really have a current limiting resistor in series with the LED - you're going to damage the LED or damage the output pin wired up as it is."

not really that far along yet.. =(

(knowing how to correctly wire up the leds/components for testing/development..  I really want to get to the code parts of playing more) LOL

I thought the chip was capable of turning on smaller accent leds types without much fV and low mA?..
(but what do I know).. I never know if I have things wired up right/wrong..


to be clear I put the + lead to the chip/pin always,.. and the - to the GND rail..
(not other way around)?

where does the resistor go?  between I/O pin and + lead from led?





Thanks Lefty, FCrossroads..everyone for replying/helping out.


appreciate it. =)

CrossRoads

#7
Feb 18, 2011, 10:59 pm Last Edit: Feb 18, 2011, 11:01 pm by CrossRoads Reason: 1
If you look at any arduino schematic, you will see "large" caps (47uF, 100uF) on the regulator input & output, that was the caps I was referring to. The 100nF goes across the atmega328 VCC & Gnd pins close to the chip itself.
http://arduino.cc/en/uploads/Main/arduino-uno-schematic.pdf
If you look at pin PB5 (19, which is D13), you will see a series resistor with the LED. You want that.
Unless you know for sure your LED has one built in, operating without the resistor will lead to either a damaged LED (cheap) or a damaged output pin(replacement ATMegas ar $5.50 - not so cheap). If your atmega feels hot to the touch when running, that is a sure sign that something has been damaged.
There is also a 100nF cap across Vref to help that keep noise down if you are making analog input measurements.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

xl97

wow..

I barely understood much of that (sorry) I know your trying to help/educate me int he best practices.

That schematic looks foreign to me so far.. =(  although I do understand the D13/P19 reference..

(Im trying to stick with it.. once I do things a few times/repetition.. I understand and feel comfortable..just not there yet) =(

here is the 'fix' I did on both breadboards to stabilize my vRegulators on my projects:






looking at the schematic (again)..

Im thinking you mean the RIGHT side/box/object for 19 (PB5).. not the one on the left? since it goes nowhere?

although I have no clue how to tell or see there is a 'series resistor'...

matter of fact the more I look at the diagram/schematic.. the more Im getting lost.

so to swing back on topic..

I do 'not' see or understand these 'large caps' why or even placement.. vs the caps I just put in place per Leftys suggestion and cobbling it together from the links.

is what I did wrong?


thanks

retrolefty

Almost correct. That larger capacitor you have wired between ground and +5vdc should be changed so that it's negitive lead is still wired to ground as you presently have it, but it's positive lead is connected to your regulator's input pin, which is any of the pins in row two on your picture.

Lefty

CrossRoads

Not wrong, just not very effective.

The parts I am talking about on the schematic are:
PC1, PC2, these are like the 22uF cap you have by the regulator now. That is good. Get another one of those.
See figure 1 here:
http://focus.ti.com/lit/ds/symlink/ua7805.pdf
100uF might be overkill, but it won't hurt. I've been using 1uF on the input and 0.47uF on the output, but only because I had a bunch on hand.

Move the little cap across  pins VCC & GND over by the crystal the way that C6 is connected up.
Wire up a current limit resistor the way that RN2A is connected.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

CrossRoads

So if you measure the voltage on the reset pin, is it changing with your switch?
I ask because switches like this usually have pins 1/2 and 3/4 connected internally - so you may have pin 1 wired to ground and pin 2 to the reset pin, so the switch is not doing anything.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

xl97


Almost correct. That larger capacitor you have wired between ground and +5vdc should be changed so that it's negitive lead is still wired to ground as you presently have it, but it's positive lead is connected to your regulator's input pin, which is any of the pins in row two on your picture.

Lefty



hey Lefty-

thanks for the reply..

I have changed the error per your suggestion.. (I think)




however,...I have a question.. (sorry if this is 'noob' stuff)

would it be any different or problematic if wired up like then?



thanks




@CrossRoads-

thanks for the suggestions/links... not very good are reading/following schematics yet.. symbols/signs..etc (all new still)..
so I'll have to go over it more in depth.

Easy to say wire something up like this or that or like the C6..etc  for someone with a bit more .. 'comfort' I guess?

I really appreciate the help though.. (just wish I felt like I was doing your time/help justice by understanding it and being able to implement it easily)


but if I 'do' in fact have the vRegulator portion all wrapped up.... I can now move on to focus on your comments solely.
(is the link you supplied supposed to be to the regulator spec sheet?... confused as you talk about the VCC & GND of the chip...etc directly after)?

(was/am looking for figure one in .pdf you linked to)


Thanks

=)

retrolefty

Quote
however,...I have a question.. (sorry if this is 'noob' stuff)


That's fine, it's electrical the same.

Lefty

CrossRoads

Yes, the last PDF I linked to was a 7805 regulator, with figure 1 showing 0.33uF on the input side of the regulator and some smaller value on the output side.

The C6, etc parts I referenced are from the arduino schematic I linked to also. Print that out, find the parts, and get them setup up on your protoboard the same way.  If you're gonna play with electronics, you're gonna have to dive & start getting used to looking at schematics.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

Go Up