... beyond the theoretical aspects of this issue and the excellent work of Conrad one may spent a few minutes to understand why an electrolytic is not made to last for ever.... ( apply to some people that said """if its not broken why bother to fix it ""
)
Imagine how a capacitor is made aluminum foil , paper foil and electrolyte ... aluminum housing and rubber as a seal ...
obviously mechanical leaks is an issue fluid may drop on the pcb from bad sealing or any other from of failure that can be chemical deformation of the rubber or other seal ...
high temperature is a big issue .... either is produced inside the machine the capacitor is working ... either is produced from the capacitor it shelf in the struggle... to filter the ripple, actually electrolyte is boiling and anything that boils need to evaporate some how
over voltage caused by any issue will have the same effect IE electrolyte will boil .... over voltage may also have to modes .... one might be high or higher voltage for a short period of time or just a bit more higher voltage but for a very long period of time ...
Surge is another issue ... not to be mixed with over voltage high pitch spark with 10's of times the voltage of the capacitor and for a fragment of a second can actually drill through the plates and/or boil the electrolyte
Calculate the above even worst if capacitor is exposed to excessive heat or moisture
there might other failure modes that i missed ... but any way under this logic if capacitors where properly or better sealed they will actually last for ever ..... well ... not true even elnas orange capacitors that are sealed with some short of resin still fail and beyond sealing issue there is all shorts of chemical things happening inside a cap ....
imagine a similar thing like a soda can that is just water and gas has an expire date ....
kind regards
sakis
) Imagine how a capacitor is made aluminum foil , paper foil and electrolyte ... aluminum housing and rubber as a seal ...
obviously mechanical leaks is an issue fluid may drop on the pcb from bad sealing or any other from of failure that can be chemical deformation of the rubber or other seal ...
high temperature is a big issue .... either is produced inside the machine the capacitor is working ... either is produced from the capacitor it shelf in the struggle... to filter the ripple, actually electrolyte is boiling and anything that boils need to evaporate some how
over voltage caused by any issue will have the same effect IE electrolyte will boil .... over voltage may also have to modes .... one might be high or higher voltage for a short period of time or just a bit more higher voltage but for a very long period of time ...
Surge is another issue ... not to be mixed with over voltage high pitch spark with 10's of times the voltage of the capacitor and for a fragment of a second can actually drill through the plates and/or boil the electrolyte
Calculate the above even worst if capacitor is exposed to excessive heat or moisture
there might other failure modes that i missed ... but any way under this logic if capacitors where properly or better sealed they will actually last for ever ..... well ... not true even elnas orange capacitors that are sealed with some short of resin still fail and beyond sealing issue there is all shorts of chemical things happening inside a cap ....
imagine a similar thing like a soda can that is just water and gas has an expire date ....
kind regards
sakis
Great article Conrad. I have an HP analyzer at work, an HP4194A and have done some basic ESR measurements on some caps and am surprised at how only a few inches extra wire will degrade the frequency plot. Most 'lytics over 35mm in diameter have resonance at < 20kHz and show a negative capacitance over the resonance frequency. Can you (or anyone) explain to me what exactly the negative capacitance means? I think I understand some of it; that as it gets negative the capacitance actually starts doing the opposite of what it is supposed to (be a low AC-impedance), but it still holds a charge, and thus functions as a power reservoir. But I am confused regarding what it means in practical terms.
and have done some basic ESR measurements on some caps and am surprised at how only a few inches extra wire will degrade the frequency plot. Most 'lytics over 35mm in diameter have resonance at < 20kHz and show a negative capacitance over the resonance frequency. Can you (or anyone) explain to me what exactly the negative capacitance means? I think I understand some of it; that as it gets negative the capacitance actually starts doing the opposite of what it is supposed to (be a low AC-impedance), but it still holds a charge, and thus functions as a power reservoir. But I am confused regarding what it means in practical terms.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.