• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Replacing capacitor LED Driver

Status
Not open for further replies.

skmdmasud

Member
Hi..
In my 220VAC LED lamp 40W the driver suddenly stopped working, after inspection i found out that the capacitor 400v 15uf was burned. The part of the circuit is like 220VAC input -->4 diodes ---> to 400v 15uf cap. and then the rest of the circuit which i do not understand but looks very simple with very few parts and very compact. something like attached image.

I replaced it with a 400v 10uf capacitor which i took from an old circuit which by dimension is bigger. My lamp is working again with same amount of light i do not notice any difference. SO the question is why/how.



Regards
Masud.
 

ci139

Active Member
it's an input filter capacitor that makes rectified sine ⋂⋂⋂⋂ more like direct current ⁀⁀⁀⁀
by lowering it's value you actually loose some power (also power passing that capacitor / e.g. the dynamic power(current) heating up the capacitor)
you may have a better circuit now unless the switching circuitry can't keep up with the modification

so hoping that the designer didn't pass a design flaw you should use the ® name value 33µF =
= 22+10(+1) = 3×[10(+1)] as for reducing heat load to capacitors
 

Nigel Goodwin

Super Moderator
Most Helpful Member
SO the question is why/how.
I don't really understand your question?, you replaced the faulty part so that cured the problem.

The capacitor is simply a reservoir capacitor, it's value is extremely non-critical, as long as it exceeds a certain minimum. In fact, I've even known power supplies work with no obvious problems with the capacitor completely O/C, or even removed entirely - I've also known plenty that self-destruct under such circumstances as well!.
 

schmitt trigger

Well-Known Member
Hmmmmmmmmm......
The photo clearly shows a 400V, 33 uF capacitor............not a 15 uF capacitor.

BTW, I like your phrase: It is better for a leader to make a mistake forgiving than make a mistake punishing
 

ci139

Active Member
BTW, I like your phrase: It is better for a leader to make a mistake forgiving than make a mistake punishing
-- indeed . . .
i wonder if the following has a translation error
The strong is not the one who overcomes the people by his strength, but the strong is the one who controls himself while in anger.
the hypothesis here is that the strong should be smart or wise -- the feeling of anger is disintegrating/chaotic/isolating e.g. the one feeling so will detach itself from "system" also from potential NRG sources and support groups + it gonna focus on the narrow*rejection rather than keeping it's system steady operational and open (to change for example) . . . . no matter than not surrendering to charm of anger may result in an image of a "stronger" (more capable) person or group of society the actual cause is me assume a wise or smart decision not that being strong automatically sets one to make the "right" choises . . .
 

skmdmasud

Member
Hmmmmmmmmm......
The photo clearly shows a 400V, 33 uF capacitor............not a 15 uF capacitor.

BTW, I like your phrase: It is better for a leader to make a mistake forgiving than make a mistake punishing
Hi,
Pic is from internet not the real driver, but its very similar.

The phrase is from Arabic glade you liked it.
 

schmitt trigger

Well-Known Member
got it!

To your question, cheap electrolytic capacitors have a wide capacitance tolerance: usually -20/ +85%

As Nigel mentions, many applications are completely non-critical.
 

ci139

Active Member
since the NEW 10µF capacitor causes the "direct current level" e.g. the average energy available for switching . . . . at the high side . . . . NOW to be lesser than it would be with higher value one - the switch has to operate more frequent or at higher duty ---- the energy been converted as you reported is principally the same as it was with your OLD 33µF one , ! BUT ! . . . . the longer ON / higher duty would put larger current to magnetic chain of the transformer (putting higher than designed load on TF and switch SOA) the higher switching frequency may cause switch to overheat and may fall off the "efficient" operating range of the TF e.g. you'll need more input watts per output watt than before . . .

. . . in other words -- your fix may do as temporary solution but not as the most reliable one
 

Nigel Goodwin

Super Moderator
Most Helpful Member
. . . in other words -- your fix may do as temporary solution but not as the most reliable one
Not at all, the massive tolerance on electrolytic's, and the non-critical nature of the component, means it's perfectly fine.

The fact it's physically larger (while having a smaller value) means it's very probably a better quality capacitor, and will last longer than the original.
 

ci139

Active Member
The fact it's physically larger (while having a smaller value) means it's very probably a better quality capacitor, and will last longer than the original.
you may be right here - but i still stick in that the OP should switch back to ® name value - there might be few 10-s or even fractions of percent difference on how the circuit behaves (applies also for supply I/O filter capacitors)
 

Nigel Goodwin

Super Moderator
Most Helpful Member
there might be few 10-s or even fractions of percent difference on how the circuit behaves (applies also for supply I/O filter capacitors)
Not at all, and even assuming the capacitor value was arrived at by calculation?, it was calculated for the cheapest - not for the best - and many of the capacitors fitted are incredibly poor quality (and remarkably small).
 

Nigel Goodwin

Super Moderator
Most Helpful Member
- i assume for the SMPS frequencies usually about (8kHz - ) 60kH - 400kH
Why?, it's the reservoir capacitor, it essentially runs at 100Hz or 120Hz only - hence they aren't usually low ESR, or even 105 degree, and it's almost impossible to find such specifications for the required values.

If you don't understand what the capacitor is doing?, why are you offering uninformed suggestions?.
 

hyedenny

Active Member
Exactly! It's sort of like someone who uses punctuation marks when they clearly have no idea how to correctly use them.
 

ci139

Active Member
If you don't understand what the capacitor is doing?, why are you offering uninformed suggestions?.
i'm just trying to decode what your hoax is all about - i haven't seen an electrolytic that does differ noticeably from it's printed label value - although that i have not found almost any ceramic one that matches it ??? . . .
i'm just wondering what are you talking about - you just say something that does not correlate with anything i know ???

as in fact i haven't seen you guys coming up with constructive sentences too often
most of the time you just bring any possible subject down to "find a loony" sketch
i don't consider such funny -- leaves me impression that you all never seen a clear day in your life ???
 
Last edited:

Nigel Goodwin

Super Moderator
Most Helpful Member
i'm just trying to decode what your hoax is all about - i haven't seen an electrolytic that does differ noticeably from it's printed label value
Obviously no hoax at all, if you read the label on the capacitor the tolerance is plainly marked, and can be absolutely HUGE - particularly for cheap ones.

I fail to see why you imagine the value in this case (of a simple reservoir capacitor) is in any way critical - essentially it's 'the larger the better', with any calculations (is indeed any were done) designed to give the cheapest they could get away with.
 

ci139

Active Member
QC. (the ceramic ones likely won't have 2 sig. digits matching)
el-ly. 10µF : 10.7 , 1.9% , 6.7% -- test-patch mean , mean deviation , offset from name value
el-ly. 10µF : 10.8 , 4.0% , 8.0% -- ! updated -- see '.txt
cer.-ic 1µF : 0.875 , 4.0% , -12.5% -- sap (actually it's better than i imagined , muhahaa)
_________
i likely are to switch to links only mode to avoid any futher pointless disputes in a future
 

Attachments

Last edited:
Status
Not open for further replies.

EE World Online Articles

Loading
Top