-
Do I need to change the same thick cables ?
-
The 7805 IC is very hot. Input voltage is 12V.
-
Does it has thermal shrink tape ? I couldn't insert shrink tube.
natong:
2. The 7805 IC is very hot. Input voltage is 12V.
Its dissapating close to 2W without a heatsink, expect it to get very hot and not last very long.
-
I replaced new cable and connector. Now 5.02V equal both cables. Maybe bad soldering or bad contact.
-
I added a small heatsink.
-
Could you please teach me about power dissipated ? How much min. power dissipated that doesn't require a heatsink ?
For your 5V supply, get a buck converter instead of a 7805 (or other linear regulator). Much more efficient. Indeed you're dissipating 0.26 * (12-5) = 1.82W in that regulator.
The voltage difference... maybe it's simply the error of your multimeter (did you measure it a few times, alternating sides in between each measurement?), or a poor contact indeed.
A TO-220 package can do up to about 1W without heat sink - depending on amount of ventilation, ambient temperature, and internal construction. Anything over 0.5W though would benefit from a heat sink.
Yours looks pretty small. Make sure your enclosure is very well ventilated, as if not your part will of course still overheat.
wvmarle:
For your 5V supply, get a buck converter instead of a 7805 (or other linear regulator). Much more efficient. Indeed you're dissipating 0.26 * (12-5) = 1.82W in that regulator.The voltage difference... maybe it's simply the error of your multimeter (did you measure it a few times, alternating sides in between each measurement?), or a poor contact indeed.
A TO-220 package can do up to about 1W without heat sink - depending on amount of ventilation, ambient temperature, and internal construction. Anything over 0.5W though would benefit from a heat sink.
Yours looks pretty small. Make sure your enclosure is very well ventilated, as if not your part will of course still overheat.
Thank you very much. Well, I will replace with a switching regulator and reduce the input voltage to 9V.
Never parallel voltage regulators - this is a good way to destroy stuff.
srnet:
Its dissapating close to 2W without a heatsink, expect it to get very hot and not last very long.
Why do you think it will not last very long? The Datasheet of 7805 says "The internal current-limiting and thermal-shutdown features of these regulators essentially make them immune to overload." I would expect if the device dissipates too much power occasional power failures due to temperature protection kicking in is more likely than failure of the chip itself (but the Datasheet does not say what the temperature limit for triggering the protection is).
Also Datasheet says thermal resistance from junction to ambient is only 19°C/W. Dissipating 2W means ~40°C higher junction temperature than ambient - as long as the ambient temperature is below 85°C the junction temperature should stay under recommended 125°C. This should be fulfilled with well (passively) ventilated case at room temperature - or not?
I don't think a bare TO-220 can dissipate that much heat to the air, even in the best ventilated case. They don't have much surface area!
The 19 deg/W figure is "high K JESD 51-5"
That's not air, that's mounted on a multilayer high thermal conductivity PCB according to test regime
JESD 51-5, or some such.
1W is plenty for a TO220 in free air!
Despite the datasheet claims, I have overheated a few 78xx regulators and damaged them permanently.
Who was it that said "Nothing is foolproof for a sufficiently talented fool"?
Well I know little about those things, I am more happy with my low power projects where 1mA is HUGE current 
I am just collecting knowledge about those "high power" stuff because one day I may need it.
In other datasheets for other linear regulators I have found the thermal limit is quoted around 160-180°C - enough to damage the die if sustained for longer time. For example one says:
"The device contains a thermal shutdown protection circuit to turn off the output current when excessive heat is
dissipated in the LDO. The thermal time-constant of the semiconductor die is fairly short, and thus the output
cycles on and off at a high rate when thermal shutdown is reached until the power dissipation is reduced. The
internal protection circuitry of the device is designed to protect against thermal overload conditions. The circuitry
is not intended to replace proper heat sinking. Continuously running the device into thermal shutdown degrades
its reliability."
MarkT:
The 19 deg/W figure is "high K JESD 51-5"That's not air, that's mounted on a multilayer high thermal conductivity PCB according to test regime
JESD 51-5, or some such.
Interesting, looks like you cannot be cautious enough when reading datasheets. Hidden catches everywhere. Do you have any clue how much this "high thermal conductivity PCB" is more effective than "normal PCB" or breadboard? I would guess it is less important for THT packages than SMD but what may be the real value when datasheet says 19°C/W? 40°C/W? Or even worse?
MorganS:
Despite the datasheet claims, I have overheated a few 78xx regulators and damaged them permanently.
I have seen that happen too, in the 40 years or so that I have been using the things.
Running most stuff at elevated temperatures, especially if there is repeated thermal cycling, can be expected to cause premature failure.
natong:
Well, I will replace with a switching regulator and reduce the input voltage to 9V.
The switching regulator is a good idea and will work just as well on 12 V as 9 V.
Question is - how come it is suddenly perfectly easy to choose what voltage you have to supply it? Why can you not just start with 5 V then?
But if its not battery powered why 7805 is a bad choice? Easy to construct circuit and will easily hold 2 mqs. I used 7805 in my mq sensor projects with supply voltage of 7v 1.5A. Small heat sink on 7805 with thermopaste. Voltage drop of 7805 is 2v as per datasheet. Supply voltage of in my case mq2 is 4.8-5.2v as per datasheet.
7V or 12V supply voltage makes a HUGE difference in amount of heat dissipated.
Paul__B:
Question is - how come it is suddenly perfectly easy to choose what voltage you have to supply it? Why can you not just start with 5 V then?
Thank you all,
My tiny cables are very long ~30m. Voltage drop too much. I start test with 12v and it's ok. Now waiting the shipments of 5v switching regulator MP1584EN module and 9v switching wall adaptor.
As recommended, In the future I will split one regulator per gas module.