|
John,
Every component has mixed characteristics of resistance, capacitance and inductance. Yes, a resistor has some amount of capacitance and inductance and a capacitor has a bit of resistance and inductance. Even a wire or circuit trace has resistance, capacitance and inductance. Diodes have some resistance, capacitance and inductance too. In fact, some diodes are created to enhance this effect of capacitance, they are known as "tuning diodes". Now, back to capacitors, especially the electrolytic caps. These capacitors, when charged will actually discharge by themselves out of circuit. This is because they have an internal "equivalent parallel resistance". When a capacitor charges to the rails, it should no longer draw any current, but it does draw a small amount of leakage current due to this EPR. We generally don't care much about the EPR, but the ESR is more significant because it reduces the amount of current the capacitor is capable of producing (or absorbing). You can see the ESR effect much better in a used disposable battery. Have you ever measured a battery voltage of 1.5V then put it in a flashlight to find it wouldn't work? It has 1.5V but the ESR prevents enough current to be passed. Now in a digital circuit, we need to keep the voltage on the rails steady, despite the mass quantity of transistors turning on/off. The capacitors do this by providing the extra current used in peaks and spikes as transistors turn ON, then re-absorbing the excess current when those transistors turn OFF. Ceramic disk decoupling capacitors 0.1uF have a very low ESR, much lower than Tantalum or Electrolytic. But they are meant for the fast peaks and spikes, when there are longer changes in current, the electrolytic capacitors need to step up to bat and provide larger quantities of current to stabilize and maintain the constant voltage. Here, the capacitor is limited in it's contribution due to this weird internal resistance. Look at that capacitor in the ESR meter test, it has an ESR of 0.58 ohms. If you charged that capacitor to 10V then put it on a 1 ohm resistor load, you would expect an instantaneous current of 10 divided by 1 = 10Amps if the capacitor was "Perfect", but in reality, you get an instantaneous current of 10 divided by 1.58 = 6Amps. Sometimes you may see several electrolytic capacitors in parallel, say you have 4 capacitors of 500uF each, they could have used a single capacitor of 2,000uF, but by using 4 capacitors, they are reducing the ESR of the group of capacitors. My last point about ESR is that it will generally increase as the Electrolytic capacitor gets older. I recently changed some capacitors on a friends amplified subwoofer because he could start hearing a 60Hz hum. I made sure to find and buy new capacitors of the same physical size and value with the lowest ESR possible. They were not cheap, but my friend didn't mind the expense and was pleased to get rid of that hum. I'm sure you can read a lot more about ESR, but at some point you want to get back to "life". Cheers, Josh From: mon...@vitasoft.org To: n8vem...@googlegroups.com Subject: RE: [N8VEM-S100:5170] Tantalum capacitor replacement? Date: Tue, 9 Sep 2014 03:07:43 -0700 Excuse my ignorance Ian, but what is an ESR meter. Do you have one you recommend John
From: n8vem...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of Ian McLaughlin
My experience with old tantalums is the opposite. I usually find at least one 1 shorted tantalum on anything from the 70s or older. If I acquire anything of that age, I go through all of the tantalum capacitors with an ESR meter and test them before powering on. The nice thing about an ESR meter is you can test a capacitor in-circuit without having to remove it from the board.
If the ESR is still within tolerance, I leave it in. If the ESR is out of spec, I will replace it before it decides to fail spectacularly.
I recently restored a NorthStar Horizon, and I had one pre-exploded tantalum and 2 shorted ones.
Ian
-- -- You received this message because you are subscribed to the Google Groups "N8VEM-S100" group. To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. |