NerdKits - electronics education for a digital generation

You are not logged in. [log in]

NEW: Learning electronics? Ask your questions on the new Electronics Questions & Answers site hosted by CircuitLab.

Support Forum » Two Black Bars in RUN mode compilation:

March 20, 2011
by Ralphxyz
Ralphxyz's Avatar

Once again I am having the Two Black Bar in Run Mode situation.

On two of my breadboards (third party not Nerdkit's this time) I get the two black bars in run mode I would like to do a compilation of all of the solutions to this problem. This seems to pop up often for me even on boards that worked one day but stop working the next.

I know others have dealt with this. There never has been a definitive answer it seems there are a number of "things" that cause this problem so I would like to have all of the solutions consolidated here in one spot.

So here we go:

I do not use batteries. ( I know a low battery can lead to the"butterfly" problem not sure about the black bars).

I get the black bars with a power supply and/or USB power.

My chip has a loaded program that runs fine on other breadboards. So it is not the chip.

Excluding power there seems only a few "locations" that would cause this problem.

The chip Power wiring. (The two black bars appear so this "seems" to be negated).

The chip Ground wiring. ( ditto )

The Program/Run switch wiring. In either direction I get two black bars

The Program/Run switch itself. With no switch in place (pin 14 floating) stil two bars

The USB connections "should" not be a factor but even with USB removed two bars.

Oh speaking of USB I am able to program the mcu on the boards that are giving me the two black bars but cannot run the program.

One "solution" I had in the past was to use 20ga. copper wiring not the 22ga. wire.

The 20ga. certainly gets a better purchase in the breadboard.

So where else might the problem lie. I would like to consolidate the solutions.

It seems really simple there should be a "spot" where I can do "this" and my problem "should" be solved.

Ralph

March 20, 2011
by Noter
Noter's Avatar

A missing crystal or other connection problem with the crystal can cause the black bar problem.

March 20, 2011
by Noter
Noter's Avatar

Hmm ... if you can program the mcu in question then doesn't that indicate the wiring, power, and crystal are all good? Which pretty much says that the program you are loading is the problem. For some reason it doesn't run.

March 20, 2011
by hevans
(NerdKits Staff)

hevans's Avatar

Hi Ralph,

Just throwing a possibility out there. Given that you say you can program tha chips but can't seem to get back to run mode. I think the issue could be that the usb-ser cable is keeping the chip on with power over the signal lines. Have you tried completely disconnecting the usb-ser cable and then resetting the power.

Humberto

March 20, 2011
by Ralphxyz
Ralphxyz's Avatar

I started out with a brand new Nerdkits ATmega328p mcu and the Greetings program ran on my other breadboard.

I'll pull the USB again to makesure. I know the USB keeping the power on the chip (zoombie mode) has been the problem in the past.

I had been using a switch on the yellow wire but then I tried powering from USB with the red wire to the same DPDT switch and then this started happening. I put LED's on the yellow and green wires, the green is always on then when I load a program the yellow comes on and never turns off so I know the chip has zombie power. In fact the indicator led on my power supply will light even when the switch is off and that is going back through the regulator.

The USB has definitely caused this problem in the past.

Thanks,

Ralph

June 29, 2011
by ForScience
ForScience's Avatar

I was having the same issue, and now I have stumbled on a stranger issue. For the longest time, none of the suggestion would work. So i removed power, and tada, the temp sensor program was running. OFF THE USB! I have 4v on the rails. When I disconnect usb and hook up power I notice I have 4v going to LCD on pins 2,4,11,12,13,14 (at the LCD), and under usb power I only have it on pins 2 and 4. What would be causing this? I am able to successfully load programs to chip, however they will only run off the usb...I am stumped.

Wrap your minds around this one guys.

June 29, 2011
by Ralphxyz
Ralphxyz's Avatar

When do you have 4 volts on the rails?

You might be seeing voltage from the yellow USB wire.

Noter has a good solution for that you might try using the MOSFETs to kill the yellow wire when you power off.

The yellow wire energizing the mcu causes lots of problems.

Ralph

June 29, 2011
by ForScience
ForScience's Avatar

I can completely remove the usb and have the same issue with the power source. I am using a wall adapter if that helps.

June 29, 2011
by ForScience
ForScience's Avatar

what i am wandering is why are so many of the pins on the LCD being fed power when running of the battery or wall adapter?

June 30, 2011
by Rick_S
Rick_S's Avatar

Maybe take an overhead photo of your setup and we could help better.

  • What kind of wall adapter are you using and what is it's rated output (voltage and amperage)?

  • Did you confirm your wall adapter is putting out DC voltage?

  • Are you running the wall adapter through the voltage regulator?

  • Did you add any filter capacitors across the rail at the power source to 'clean up' the power from your wall adapter?

  • When running your kit off "USB", are you using the red/black wires for power?

  • When using your wall adapter, did you disconnect the red USB wire?

  • Seeing voltage on the pins you described is not necessarily an enigma. Pins 11 thru 14 are data lines and if data is being sent, you'll see voltage on them.

There are many things that can cause a circuit to not function properly. This list is far from complete but may hit some of the more common issues early users face. A word of caution though, while a wall adapter is very convenient, used improperly, it can also be damaging... So make sure you have a proper adapter and it is being used correctly. If you answer the questions above, we can help assure you that your are.

Rick

June 30, 2011
by ForScience
ForScience's Avatar

I am using a Enercell AC adapter. It allows the selection of the output voltage from 3-12v. It is set to 9v currently. Its rated amperage is 1a.
Yes it is putting out DC voltage at 9v.

Yes it is being run through the voltage regulator which is putting out 4v. It is ran through a switch to make disconnecting power more convenient.

no I do not have a capacitor across the power source. The only capacitor being used is the one connecting pins 7 and 8 of the MCU.

Yes the USB and everything else is hooked up correctly. I have dismantled and rebuilt from scratch to guarantee this. Sorry, I do not have access to a camera at this time.

The last point I could understand, however when running off the USB cable, there is only voltage present on pins 2 and 4 and the programs run fine.

June 30, 2011
by 6ofhalfdozen
6ofhalfdozen's Avatar

Ralph,

dunno if this is it, but it got me recently. A "loose" wire on from the mcu to LCD. The wires seemed ok, but I disconnected and reconnected the "communication" wires from the breadboard (not the data lines but the bit ready, etc lines) and voila it worked. just one more thought...

June 30, 2011
by Rick_S
Rick_S's Avatar

Something is either not right in your circuit or your meter. The output from your voltage regulator should not be that low if it is provided proper voltage. It should typically vary only by about +/- .2 volts. If you are only seeing 4 volts you need to determine why that is.

Assuming you are checking your voltage with a volt meter, the 1st thing I would do would be to check the voltage output from your wall adapter. Once you have determined that it's output is > 7v, I would disconnect everything on your breadboard from the power rails and then connect the power to the regulator. Then test the voltage into the regulator (should be what you just read when checking the adapter) and the output voltage. It should be between roughly 4.8 and 5.2V. If not, you may have a bad regulator.

If the regulator output is good, I would leave the LCD off the circuit and build the LED blink project. With the kit built this way, I would agian check the output voltage of the regulator.

If it's still good, try loading the program using the regulated voltage making sure the red usb adapter wire is NOT connected.

If this works continue with the LCD. If you had an issue anywhere along the way to this point, please post back what you discovered. There is something going on and often, it's difficult to see. That is why a photo helps a lot because extra eyes can often see something you may not. (By the way, that is not to mean anything against you, I've fried my share of my toys by a simple wireing error that I have missed even after a double check)

Rick

June 30, 2011
by Ralphxyz
Ralphxyz's Avatar

I have had problems with 22ga. wire.

It will be fully inserted in the breadboard but not making full contact.

The breadboards actually wear out, you can test with a continuity meter to make sure you have good contact.

20ga wire makes much better contact but you wear a breadboard out real quickly.

Ralph

June 30, 2011
by Rick_S
Rick_S's Avatar

Dang Ralph, you must really give a breadboard a workout. I've never worn one out

Rick

June 30, 2011
by ForScience
ForScience's Avatar

At work, so this will be a quick update. Yes I am using a multi-meter for my testing. Unfortunetly this meter is built for commercial electrical installers, so it's built to read higher voltages. I will be grabbing my snap-on multi-meter the next time I make it out to my house. I love that thing but it's to pricey to bring around with me everywhere.

So before work I tried your suggestion Rick. The program tries to connect to board and just hangs there, does nothing.

The loose wire might be it. I know when I was hooking up with the 9v battery, the LCD would flicker at times bringing up the program instead of the bars. But an ohm reading from pin to pin tests out ok. I'll give it another shot tonight after work.

June 30, 2011
by Ralphxyz
Ralphxyz's Avatar

Ok, see the round highlighted holes.

They are very iffy maybe they will work but often with 22ga. wire it just flops around in them there is not a good purchase when you plug in 22ga. wire.

You can see that a lot of the holes are starting to get rounded they are no longer rectangular that is wear.

Not only is the plastic wearing but if you look at the sixth hole from the bottom left you can see that the metal internal clip is malformed.

Notice the hole next to it (7 bottom left) is nicely defined.

Now this breadboard is probable a year old so it has seen a lot of service.

I caused some of the breakdown using 20ga wire but that was limited so most of it is wear and tear.

Ralph

July 02, 2011
by ForScience
ForScience's Avatar

Question about the capacitor, it says to use the largest one. However, the smallest one matches the "104" in the picture. I have 2 capacitors with "22J" which are larger than the "104". Now I went with the 104 in my circuit. Next item, I read:

However, you'll notice that pin 20 is actually named “AVCC”. This refers to the power for the Analog to Digital Converter (ADC) inside the microcontroller. Since the ADC deals with analog voltages, it is particularly sensitive to noise. Therefore, in some applications, it will have a specially regulated power supply, so a seperate AVCC pin is present.

That is directly from the manual, could my power supply be causing an issue with this pin?

Any thoughts appreciated

Happy birthday weekend America!

July 02, 2011
by ForScience
ForScience's Avatar

So I take it the loose wire had to be the culprit. After going through and rebuilding, and ohming each connection after being made, the circuit is working fine. It would still be nice to know about my earlier capacitor question. It would also be nice to know what or where the issue was to make it simpler to diagnose if and when it happens again.

Thanks for all the support.

Chris

Post a Reply

Please log in to post a reply.

Did you know that an analog comparator can tell when one voltage input crosses another? Learn more...