One of those moments when you looked down, and knew that something HAD to be wrong? Then you look over everything, try to figure out what it could be, and finally come to an acceptance of the situation?
I measured the output of one of my LM7805s today, and had one of those moments, what I saw just couldn't be right. I spent about five minutes measuring other known voltages, cycling the power supply, and everything else I could think of before I could accept it.
5 V? FIVE VOLTS? How dare they? Which manufacturer made that 7805?
Looks like a cheap multimeter . . . just the luck of the draw?
The meter is a B&K, and it's always read within a few millivolts of any known voltage I've thrown at it, it was just luck of the draw... lucky chip, lucky current draw, lucky temperature.
If I loaded it up enough to warm the chip up, the output voltage would rise, but after lowering the load, as it cooled down again, it would go right back to that. So being under a ceiling fan near an air conditioning vent must have played a part of it. That, and planetary alignment or something.
I don't think I've ever had one within even .1 volts of being exactly 5, so I had to take a picture of it... since I'll surely never see it again!
(To emuler, it was made by TI.)
since I'll surely never see it again!
Even with the 7805, I'm pretty sure you'll see it again. Statistical distribution of the output voltage of the parts manufactured will see to that. The usual Gaussian curve and all that.
My mind keeps telling me that there should be a gaussian curve, but I haven't (so far) seen it in practice. For example, I've never had a 7805 that was *below* 5V. In fact, I don't think I've previously had one below about 5.07 or so. Ah, well.
Weird... I usually get around 4.9... I sense a disturbance in the circuit...
" ... and a feeling of joy came over me when the meter showed SOO "
Yes, it happened.
To Dave: even it was a el cheapo multimeter, it would still be unique.
Nard
She is called Sylvia (2018), lives at Mint18.3 https://www.linuxmint.com/
Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips
Quite often you will see what we called a "bi-modal distribution". You pay more money for the more accurate parts and the rest of us are left with a distribution either side of nominal.
That doesn't explain why Gluteal-Celft only ever seems to get parts on the high side and daqq on the low side!
Quite often you will see what we called a "bi-modal distribution".
And for Gluteal-Cleft and Daqq's high- and low side ... global distribution ?? :lol:
(Seriously, I have no explanation for that. And personally I hardly pay attention to the precise output-value. If it's between 4.9 and 5.1V it gets status=OK)
Nard
She is called Sylvia (2018), lives at Mint18.3 https://www.linuxmint.com/
Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips
Yes, it was funny. Like the guy who was so used to eating bad food that he got a tummy upset when a neighbour gave him good food for a change. :lol:
Recently we got a batch of TO92 78L05's loaded in china, a 5V zener +/-10% would have been a better option. Thankfully we replaced this dogey part with a TO92 LP2950-CZ (LP2950-ACZ seemed slightly worse - go figure) and all reg problems are gone. (75C-80C ambient.)
The difference between a 10% reg with a BIG -ve temp coifficint and the 1% reg with minimal temp coifficient should be enough to kill the 78L05 usage, but people still use them.
Ron.
rberger,
How did you use the parts then? Did they have enough load at the output and enough bypass capacitance at the input and output?
I've witnessed some of these cheap 7805:s output very interesting voltages at no-load conditions and even turn themselves to 2 MHz oscillators without proper caps. Don't remember if they were 78L05s or just 7805s.
- Jani
Perhaps the voltage that Europeans lack is shipped to America. Somwhere between (Asia), the voltage is just about right...
The 7805's I got from Fairchild NEVER hit 5.00 They are always between 4.79 and 4.9vdc NEVER higher. Won't use them ever again. Digikey refused to return them to boot the bum's
Jim
Hmm 4.79 is within 5% - the Fairchild 78xx datasheet says they should be within 4% or 2% depending on the type. Are you sure you didn't overload or underload them?
- Jani
It can always be a fake Fairchild product... I think it has been discussed, that the Chinese don't care if your product will burn...
The batch I purchased were from Digikey. Where they got them from is of little concern to me. The load was 100ma, the input voltage was +12vdc. The package was to-220. Nothing out of the ordinary.
What has me pissed is I purchased $100.00 worth of rubbish and the supplier(Digikey) refused to take the parts back and send me National Semi's even though I was willing to pay the extra cost. Bum's.
Jim
Jepael
How did you use the parts then? Did they have enough load at the output and enough bypass capacitance at the input and output?
- Jani
Good point about fake chips! I've ran into a case where counterfeit brand A chips (x nanoseconds) were in reality brand B (x+5 nanoseconds). Revealed by X-ray.
The chips had been scrubbed from the markings and new ones were lasered. The chips were clearly thinner, plus on top of the chip there was "made in somewhere" and behind the chip "made in elsewhere" :)
Don't remember the brands, speeds or countries.
- Jani
