Electrolytic Capacitor Vs Ceramic Capacitor

[sabahan]

I build the circuit in the diagram that act as a time delay when the opamp which act as a comparator, when the condition is match, the opamp goes high which charge the capacitor instantly, when it goes low, the capacitor discharge through the resistor slowly, the signal is fed to another OpAmp, when I test the cirrcuit in the breadboard, I uses a aluminium electrolytic capacitor of 10uF, the time that it reaches the trigger point for next stage is about 10sec, but when I build the circuit on a PCB, I uses a ceramic multi layer capacitor with the same 10uF, but the time delay shorten to about 2 second, I wonder why this could happened, all the resistor value are the same on the breadboard and the PCB except that the bread board are using through hole resistor and capacitor and the PCB version are using surface mount capacitor,

BTW the capacitor on the bread board are polarized and the PCB version are non polarized

Will the surface mount version have a much lower actual capacitance value than the polarized version?

 

[Colin Mitchell]

Supply photo of 10u ceramic

 

[Harald Kapp]

Check the real capacitances of the capacitors in both circuits. Electrolytic capacitors often have -20%...+80% (!) tolerance. Ceramic capacitors often have a much lower capacitance than rated, especially when operated at >50% of the rated voltage.
Also check the resistors used. A ratio of 10s/2s is not easily explained by capacitor tolerances alone.

 

[sabahan]

all the resistor used are 1% tolerance metal film resistor, after solder on the resistor I have measure all the resistor and all are within tolerance, I have also check the ceramic capacitor value which is 9.8 uF, I than replace the smd capacitor with the through hole capacitor, it work as design, when I replace the capacitor with the ceramic capacitor, the time delay drop back to 2 second.
The working voltage of the circuit is 12 V, which is way below the rated voltage of the ceramic capacitor which is 50V, can someone explain to me why this happened?

The photo is the actual circuit on the pcb, I used a 16v 10uF capacitor for the small size, it actual is for a surface mount 1206 type capacitor of 10uF

 

[dorke]

What is the input resistance of the next op. amp stage?
Did you test it with both stages in the 2 cases- before and after PCB?

 

[sabahan]

It fed to a input of a opamp which is at least 1Mohm

 

[sabahan]

Now the problem is when I solder a electrolytic 10uF through hole capacitor on the pcb, the RC constant is larger, when I replace the electrolytic capacitor with a ceramic non polarized capacitor the RC constant shorten, the whole circuit parameter didn't change except the capacitor, why this happen?
I have measure the ceramic capacitor before I solder it on, it reads 10.3uF
The delay time shorten to a significant amount, about 30% of the electrolytic capacitor

 

[Harald Kapp]

I have also check the ceramic capacitor value which is 9.8 uF,

Did you also measure the electrolytic capacitor?
Imho the only way the observed effect can happen is when the real life capacitances of the two capacitors are very different.

 

[sabahan]

The tolerance of the electrolytic capacitor is +-20%, which means the largest value it can be is roughly 12uF, the discharge resistor is a 820K 1% metal film resistor, so the time constant on paper is 820000*0.00001, which give 8.2 second. If in the worst case, the capacitor is a 12uF, the time constant will be 9.84 second, and yes the delay time of the electrolytic cap is around 8 second, but why the ceramic cap time constant is around 2 second with a capacitance of 9.8 uF measured