Surface Mount Resistor Popped

Sounds like a Size 1210 which would be 3.2mm X 2.5mm. The dissipation rating can vary by manufacturer. For Vishay's CRCW resistors, the 1210's are rated at .33W at 70C regardless of value. For KOA/Speer's RK73B resistors, the 1210's are rated at .50W for 1K and under and .33W for values over 1K, again at 70C. Depending on the resistance value it sounds like your resistors are nearly at or just above their rating although the failure mode you described sounded more like some sort of surge event rather than just simple overheating. Hope this helped.

-- James T. White snipped-for-privacy@SPAMhal-pcGUARD.org

at

or

Thanks for the info. The value was 6.8k, so it's probably .33W.

What's funny is that it was working ok for awhile, then it went bad. I don't think there was a surge, I think what happened was what often happens when regular carbon composition resistors overheat, they start to go down in value, and as they start to go down, the dissipation increases, which gets into a runaway condition and then the resistor just goes pop. Or fizzle, or whatever. The thing self destructs.

I think what I'm gonna do is get a 1W or 2W flameproof resistor from Fry's NTE resistor mess, and try to squeeze it in there. I don't think it's gonna be difficult to make it fit. The higher wattage should hold up against having the 120VAC line across it(!)

at

or

120V, 6.8K will dissipate 2,11W If this is continuous duty you might want to try something bigger. Are you sure that it was 6.8K?

--

    Boris Mohar

Dark

dissipation rating

are rated at

the 1210's

again at 70C.

nearly at or

sounded more

Hope this

I

start

think

hold

might want

It has to handle 56VDC, and possibly ringing current, which can be up to

120VAC, 2S on 4S off. So I don't think it'll ever have to dissipate 2W. But then never say never. ;-)

I'm fairly sure, it's the same as the other circuit on the board, and the resistor says 682 on it. I went by Fry's last night, and they were out of 6.8k 1W and 2W resistors, so I'll have to check my junque box. Or else put two 3.3ks in there, if they'll fit.

That sounds like a phone line (though 56V sounds higher than the USA standard 48VDC onhook voltage), and the 120VAC ringing voltage is superimposed on the 56V. That means it's an offset sine wave with peaks of 56+120*1.414=225.68V and 56-120*1.414=-113.68V. If you do the calculus, you can calculate the RMS equivalent. The power is MORE than just 120VAC across the resistor, and if I envision the RMS equivalent voltage correctly, the average power is MORE THAN the sum of the powers of the separate AC (120VAC^2/6800) and DC (56^2/6800) components, or MORE THAN 2.12+0.46=2.58 watts.

Woops...

"Ben Bradley" wrote : "Watson A.Name wrote : >"Boris Mohar" wrote : >> "Watson A.Name wrote : : >> >Thanks for the info. The value was 6.8k, so it's probably .33W. : >> >

: >> >What's funny is that it was working ok for awhile, then it went bad. : >I : >> >don't think there was a surge, I think what happened was what often : >> >happens when regular carbon composition resistors overheat, they : >start : >> >to go down in value, and as they start to go down, the dissipation : >> >increases, which gets into a runaway condition and then the resistor : >> >just goes pop. Or fizzle, or whatever. The thing self destructs. : >> >

: >> >I think what I'm gonna do is get a 1W or 2W flameproof resistor from : >> >Fry's NTE resistor mess, and try to squeeze it in there. I don't : >think : >> >it's gonna be difficult to make it fit. The higher wattage should : >hold : >> >up against having the 120VAC line across it(!) : >>

: >> 120V, 6.8K will dissipate 2,11W If this is continuous duty you : >might want : >> to try something bigger. Are you sure that it was 6.8K? : >

: >It has to handle 56VDC, and possibly ringing current, which can be up to : >120VAC, 2S on 4S off.

Max telephone is usually -56VDC But the Ringer voltage is normally

86VRMS +/- 5% max, down to 45VRMS min.

Normally the Ringer AC is coupled via a capacitor in modern phones. (There is no real bell coil.)

: That sounds like a phone line (though 56V sounds higher than the : USA standard 48VDC onhook voltage), and the 120VAC ringing voltage is : superimposed on the 56V. That means it's an offset sine wave with : peaks of 56+120*1.414=225.68V and 56-120*1.414=-113.68V. If you do the : calculus, you can calculate the RMS equivalent. The power is MORE than : just 120VAC across the resistor, and if I envision the RMS equivalent : voltage correctly, the average power is MORE THAN the sum of the : powers of the separate AC (120VAC^2/6800) and DC (56^2/6800) : components, or MORE THAN 2.12+0.46=2.58 watts. : : >So I don't think it'll ever have to dissipate 2W. : >But then never say never. ;-) : : Woops... : >I'm fairly sure, it's the same as the other circuit on the board, and : >the resistor says 682 on it. I went by Fry's last night, and they were : >out of 6.8k 1W and 2W resistors, so I'll have to check my junque box. : >Or else put two 3.3ks in there, if they'll fit. : >> Boris Mohar : >

:

I read in alt.binaries.schematics.electronic that Ben Bradley wrote (in ) about 'Surface Mount Resistor Popped', on Fri, 9 Jul 2004:

If only that were true! You could generate power by adding AC and DC!

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

at

or

Trouble is, you don't know the whole circuit. I suspect the 120 v is current or duty cycle limited - there could be another resistance or reactance in the circuit that we don't know about. Has to be, if it was really a .33 watt unit.

[snip]

might want

was

It's a Progressive 77HP toner, and the circuit is a bidirectional red/green LED, and the resistor. As far as I know, there is no other DC circuit, only an AC signal coupled to the output thru a cap.

My problem is that it's used in a digital PBX environment where the telephone line can supply up to 5W of power(!) That's why the _constant_ 56VDC. There is no ring current on the digital PBX, only on a regular POTS line.

OK! That makes sense. At 54 volts (~2 v for the led) and with no ring, that's about .43 watts. You could probably go to a 10K resistor and still see the LED easily, and y7ou could use a .33 watt unit. Maybe it's worth trying, at least temporarily, to see if the LED is bright enough?

Thanks for the idea. I may just try a higher value, since I can't seem to find a 6.8k 1w flameproof resistor at my usual haunts. I found two

15k 1Watters, and if I can squeeze them in, I may parallel them. That's about 7.5k.