Heating of LM124AJ Quad OPamp

Hello Harish,

Yes. TI and National are one big happy family now...we are still here to help.

I do not see any obvious problems. Your resistor values are high enough to not cause large currents that would heat the device.

How "hot" is it getting? What is the temperature rise? At 30V it will get very slightly warm due to it's own quiescent current (~30mW), but it should be barely perceivable (maybe a 1° rise at most).

Even with it getting hot - are you getting the expected results? How much are they "off".

What are the voltage levels on the outputs while it is "heating"?

The LM124 is not a rail to rail input device, so as a follower the output can only swing to 2V below V+ (or 13V). The LM124 input can only go to within 1.5V of the rail (13.5V). Make sure none of the LM124 inputs have more than 13.5V on them. Are any of the outputs railed (>13V)?

Have you checked for oscillations? A part that seems to work but gets hot for "no reason" is usually a sign of an oscillation.

If you are using a digital scope, be sure to run the timebase faster than 1uS/Div to see oscillations in the 100KHz to 10MHz range. Turn off any bandwith limiting or averaging and/or filtering.

Follower configurations are the most unstable and more prone to oscillate due to capacitive loading. You need to look for places where the outputs are directly driving >20pF of capacitence (including stray). Directly driving coax cables are a notorious oscillation source, as well as long, unterminated PCB traces. The input capacitance of another device can also cause an oscillation.

I see the possibility with U2C directly driving the AD534's high impedance input X1, and/or U2B driving the "front Signal Facia" line. Is this a coax line or a wire run in a tight bundle of other wires to the front panel? Or a long PCB trace? THere should be an output isolation resistor (~100ohms) on any line that goes off the board to the "outside world" to protect against ESD, shorts or other external abuse.

Put a scope on each LM124 output and look for high frequency oscillations.

If the LM124 is in a socket - try lifting the output pins one-at a time, starting from pin 14 and working backwards.

Also try removing (or lifting the input leads of) U3 and/or U5 to see if the "heating" goes away.

Regards,