With one breadboard tied up in testing a project mock up I shifted to a new breadboard that had not been used before.
I was using it to test some new RTC modules I picked up. So had setup up a simple NANO/ I2C LCD/ I2C RTC circuit. Also decided to add in a IR detector that I wanted to test for a project.
Basically using the IR to turn on and off the LCD back light.
I started noticing my nice display would work fine after power was applied but then randomly the display would go into a series of 165's and not reset until power was reapplied.
My first thought was the new clock, so swapped with a second one but same results occurred.
I had also added some sketch changes to add leading '0's to the hour/mins/secs when below 10, so double checked that by reverting to old sketch.
Again problem still occurred.
I connected my I2C analyzer and monitored the buss but data and timing appeared to flow properly until I bumped the breadboard slightly.
After a bit of wire wiggling and such I found the 5 and ground lines from the NANO to the breadboards + and - buss would result in the display errors.
After wire swapping and pin moving I finally had to double up the wires to clear all but major vibration issues.
SO:
CASE IN POINT if you have some flaky issues with your test circuit and it is on a breadboard check you breadboard.
I will trash that one and dig back in my junk closet for that twenty year old one that I had. It was always reliable.
For the cost of the breadboards nowadays, probably a good idea to replace it as soon as you notice any interrmitant problems.
Also avoid large diameter components, especially near the center trough.
On the bad on you can remove the individual spring pieces and keep the ones that haven't been deformed.
Use as replacement in other bread boards as needed.
I this is a new board just opened, and now in the trash.
My old board has provided hundreds of uses with no failures, just do not make them like they used to.
Amen: I wire up fairly involved circuits on breadboards (30 - 40 components), and the connection reliability, even on good breadboards, is poor. If somebody could come up with a better design, they would do the world a big favor and get rich at the same time.
The only technique I have to ameliorate connection difficulties is, use fairly large wire. If you have to plug something in which has questionable leads, solder bigger wire to the leads. The time saved by not having to chase connection ghosts will be much greater than the time spent soldering. Also, I think boards eventually just wear out. Two years of steady use is about tops.
I read somewhere recently that breadboards suffer from huge capacitance problems too. I think that might explain the crap output I get from a circuit I made with an ICL8038 function generator, where according to my 'scope, a large proportion of the squares in the square wave are missing.
I'm going to find a dil socket and dead-bug it, see if it improves.
One point, if it dosen't oscillate on the bread board then it more than likely not oscillate on your PCB.
I have used a bread board at 20MHZ with no problems but obviously it depends on the circuit.
Also, if it works on the bread board, it may not work on a PCB as you may have to add capacitance to your PCB circuit to get an oscillator design to work.
I guess the point is be aware the capacitances are there and they will effect your circuit (pro or con).
Also the long hook up wires used in a breadboard design will act as caps or inductors/antennas.
2pF good or bad... depends. On lower frequencies, no difference.
Xc = 1/(2pifC)
Square waves contain the fundamental plus odd harmonics, requires up to about 9th harmonic to look relatively square. So calculate the capacitive reactance at that frequency. If it is in the range of the resistances around it and not at least 2 to 3 times higher, you'll see ill effects.
It can also cause unwanted coupling from one part of the circuit to another, especially troublesome if you are using analogRead().
I think I used to complain on using Fahnestock clips, then moved to wire wrap which was more effort but much more stable, then to breadboards which was in between the two for quality and ease.
But I do think older breadboards had more metal in the spring clips and they were plated with better metals. You could also take the back off and bend and re-align the clips if needed.
I think the trashed one that does not work I will tear apart and compare to one of my dead thirty year old ones and maybe one of the sack full of twenty year old ones just to see the quality change. If I can just find the closet, box or room full of electronics that they are hiding in.
I also think I may just pull out my wire wrap gun and test board for wire wrapping and move back to that for test efforts.
I hesitate to say how old my oldest protoboards are... I bought several when I was in high school, working at Radio Shack around 1979. They look dingy and dirty, but still work. I have hosed them out with cleaner a couple of times, held upside down and then followed with dry compressed air.
Not because I was having problems, but just because.
I've measured between 2pF and 15pF between adjacent rows on a variety of protoboards. I'm sure it depends on the geometry and the dielectric constant of the plastic.
