Recommendations for Failed Samsung Power Supply?

[KenO]

Received a Samsung UN32EH4003FXZA LED TV with starting problems. From experience with this model guessed it had bad cap(s) so checked the Power Supply and found the 47uF 160 V cap starting to bulge. The adjacent 47 200 V cap and all the other caps are OK.

To date, the best video I have found for this model TV problem is by radiotvphononut https://www.youtube.com/watch?v=9yLLAnjpTLI

I would like to upgrade this failed 47uF 160 V cap so would appreciate any suggestions.

Thanks

Ken

[pf...@aol.com]

Look for caps rated to 100 C. This greatly enhances longevity.
Raise the capacitance to the maximum that will fit in the space available, up to double OEM.
Bring up the voltage to up to double OEM.
Look for a decent brand - but if you follow the above, boutique caps are emphatically not necessary.

Peter Wieck
Melrose Park, PA

[c4urs11]

On Mon, 15 Aug 2016 10:24:19 -0700, pf...@aol.com wrote:

> Raise the capacitance to the maximum that will fit in the space available

Some caution though: raising capacitance also
raises peak current through rectifiers.
Rectifiers do have peak current limits.

Cheers!

[pf...@aol.com]

Hence the "Up to double the OEM limits".

Generally, and unless the OEM is _REALLY_ cheap, the rectifiers are not the limiting factor.

[Phil Allison]

c4urs11 wrote:
>
>
> Some caution though: raising capacitance also
> raises peak current through rectifiers.
> Rectifiers do have peak current limits.
>

** That idea is a common myth about line frequency PSUs and a nonsense about SMPSs.

Doubling the microfarads means the voltage drop is halved during the non conducting period - so when the diode conducts, it has to raise the voltage by half the previous amount. Doing this requires the same current to flow for the same time as before.

The peak current value depends on the transformer's winding resistances and any series inductance.

Increasing the capacitance does increase the start up surge current - but most diodes have large single cycle ratings. The 1N4004 has a 1amp average current rating with a * 30amp * single cycle rating.

http://www.onsemi.com/pub_link/Collateral/1N4001-D.PDF


.... Phil

[ohg...@gmail.com]

If it cycles like the one in the video, it could be that cap. It could also be the main board as Samsungs are notorious for having main board issues (including cycling). If it starts and has back light shut down but the TV doesn't cycle, it's most likely a flakey LED in the display. My notes show I did several of these and none of them had a power supply problem.

[KenO]

On Monday, August 15, 2016 at 11:45:36 AM UTC-4, KenO wrote:

Thanks everyone for your comments!

ohg... "If it cycles like the one in the video, it could be that cap." Yes it does. Is there some way I can post a JPG of bad cap?

pf... "Look for caps rated to 100 C. This greatly enhances longevity." The original bad SAMWHA was listed as 105 C.

"Look for a decent brand" Was thinking of using a NICHICON. Any comments?

Checked the cat http://www.nichicon.co.jp/english/products/alm_mini/index.html

Any suggestions concerning Series? Was thinking of upgrading from 160 V to 200 V? Any comments?

Am unsure if should keep original 47uF or ~ double it?

Are there any references as to difference it will make in cap life?

Am also wondering about normal vs low ESR cap?

Rechecked SAMWHA WL Series http://www.samwha.com/electric/product/list_pdf2/WL.pdf and failed cap was listed as Operating temperature range of 40 ~ +105 C, Extremely low impedance at high frequency, High reliability withstanding 5000 hours load life at 105 C

From checking seems SAMWHA has a poor reputation either that or Samsung did a poor job in designing this Power Supply!

It is a BN44-00554B Did a search and lowest price is $14.99 from https://gopartserv.com/products/samsung-bn44-00554b-power-supply?variant=12917092228&gclid=Cj0KEQjw88q9BRDB5qLcwLXr7_sBEiQAZsGja-bzrMPb6O-kmBOv4T-jbdqXsZDzILP3gEcO2tEB4gkaAh2V8P8HAQ

If you search "BN44-00554B problems" will find Samsung UN32EH4003FXZA - tries to start - problem found! http://www.badcaps.net/forum/showthread.php?t=40665 so do not have much faith in this Samsung PS at any price!

[ohg...@gmail.com]

Don't overthink this. This is simply a bridge bypass cap, not a PFC filter. For less than $4 you get two delivered:

http://www.ebay.com/itm/2pcs-200v-47uf-105-C-Radial-Electrolytic-Capacitor-US-SELLER-/252149403315?hash=item3ab5468eb3:g:ob8AAOSwdzVXk-eD

Unlikely you'll ever change this cap again.

[KenO]

ohg,

"Don't overthink this. This is simply a bridge bypass cap, not a PFC filter."

Am trying to use this as a learning experience.

I have a hard time believing that Samsung would continue to use poor quality caps in their LED TVs.

Instead am wondering if the Samsung BN44-00554B is a poor design? If this is correct then the 47uF 160 V 105 C cap should be upgraded. My problem is I have No knowledge or experience in this area, which is why am asking for suggestions.

Do you know if the schematic is available on line?

Thanks again for your help!

[pf...@aol.com]

On Thursday, August 18, 2016 at 12:38:31 PM UTC-4, KenO wrote:
> ohg,
>
> "Don't overthink this. This is simply a bridge bypass cap, not a PFC filter."
>
> Am trying to use this as a learning experience.
>
> I have a hard time believing that Samsung would continue to use poor quality caps in their LED TVs.
>
> Instead am wondering if the Samsung BN44-00554B is a poor design? If this is correct then the 47uF 160 V 105 C cap should be upgraded. My problem is I have No knowledge or experience in this area, which is why am asking for suggestions.

Samsung will use those caps that are most likely to survive the warranty period by sufficient a margin that the owner forgets when the item was purchased - about 2 years, or so. After which it is counter-productive to improve results.

If the OEM spec. is 47 @ 160 @ 105, stick with the 105, or look for 125C caps. You would do OK with these:

http://www.mouser.com/ds/2/420/-772867.pdf Mouser only to show what I mean - you do not need 1,000 of them!

[bitrex]

Phil Allison <palli...@gmail.com> Wrote in message:

A good point that I've seen Phil harping away on in the past;
maybe someday folks will get the picture.

I've seen the same error stated in ~40 year old textbooks...
--


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

[ohg...@gmail.com]

On Thursday, August 18, 2016 at 12:38:31 PM UTC-4, KenO wrote:

>

> I have a hard time believing that Samsung would continue to use poor quality caps in their LED TVs.
>

Oh please. Manufacturers will use the cheapest part they believe they can get away with, and Samsung is just as bad as anyone else. But more than that, this was most likely a bad run of caps, not just a crappy selection. If you replace this with the cheapest one you can find of the same value and voltage I promise you it will last a lot longer than your OEM cap did.

And if you really want to pick on Samsung, their LED TVs are killing LEDs in the display in wholesale numbers. The last one we did had only 11 out of 44 still working (after I jumped out the open one that caused the shutdown). 32 LEDs shorted and one open after 2 years of service. THAT is crappy design and/or part selection.

> Instead am wondering if the Samsung BN44-00554B is a poor design?

There's no "design" issue involved with your particular failure. It's a simple full wave bridge and bypass filter. If you want to say the cap was cheaply spec'd out, that's fair although I still believe it was just a shitty run of caps.

>>If this is correct then the 47uF 160 V 105 C cap should be upgraded. My problem is I have No knowledge or experience in this area, which is why am asking for suggestions.

If it makes you feel better, put in a 68uf/250v/105C and get it from Panasonic.

> Do you know if the schematic is available on line?

Not sure, but it probably isn't. Most service manuals these days give you block diagrams and parts lists but not much else. It's often easier to just get the datasheets of the ICs in the circuit and work with those than waste time looking for schematics.

[John Robertson]

> ..... Phil
>


I understand that one can calculate inrush current easily; divide input
voltage by ESR of the capacitor. So a 220ufd cap (rated at 25V or less -
0.4R ESR) run at 12V will have an inrush current of 12/0.4 = 30A...
However a capacitor rated at 35V may have an ESR of 0.2R (60A) and at
63V the ESR may be down to 0.15R (80A @ 12V)

Or am I doing something wrong here...

John :-#)#

--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

[Phil Allison]

John Robertson wrote:

>
> >
> > ** That idea is a common myth about line frequency PSUs and a nonsense about SMPSs.
> >
> > Doubling the microfarads means the voltage drop is halved during the non conducting period - so when the diode conducts, it has to raise the voltage by half the previous amount. Doing this requires the same current to flow for the same time as before.
> >
> > The peak current value depends on the transformer's winding resistances and any series inductance.
> >
> > Increasing the capacitance does increase the start up surge current - but most diodes have large single cycle ratings. The 1N4004 has a 1amp average current rating with a * 30amp * single cycle rating.
> >
> > http://www.onsemi.com/pub_link/Collateral/1N4001-D.PDF
> >
>
>
>

> I understand that one can calculate inrush current easily; divide input
> voltage by ESR of the capacitor. So a 220ufd cap (rated at 25V or less -
> 0.4R ESR) run at 12V will have an inrush current of 12/0.4 = 30A...
> However a capacitor rated at 35V may have an ESR of 0.2R (60A) and at
> 63V the ESR may be down to 0.15R (80A @ 12V)
>
> Or am I doing something wrong here...
>

** When you simply connect an electro to a DC source, your calc gives the peak surge value.

However, line frequency and SMPSs are way more complicated and the latter usually include a "soft" start systems for the input AC and DC output sides.




.... Phil

[KenO]

ohg,

"Samsung, their LED TVs are killing LEDs in the display in wholesale numbers. The last one we did had only 11 out of 44 still working (after I jumped out the open one that caused the shutdown). 32 LEDs shorted and one open after 2 years of service. THAT is crappy design and/or part selection."

Have not checked the display LEDs yet. Could bad LEDs have caused the 47uF cap to fail?

If yes seems like necessary to protect the display LEDs.

Did see mention about turning down the display level.

[John Robertson]

> ..... Phil
>
>

Hi Phil,

My point was one has to be careful with what cap you choose to replace
the original. Too low an ESR may over stress the diodes. If the input is
protected with a varistor or similar then indeed, I agree there isn't a
problem, however if the bridge rectifier is directly connected to line
(via xformer, etc.) then wouldn't lower ESR caps introduce a problem?

[ohg...@gmail.com]

NO! NO! NO! You're overthinking this again!!

It's simply a rectified and filtered AC supply. They've been building power supplies this way since the 1920s. It's simple and foolproof. Your rectifier is good, your cap is bad. The only external factor that could have caused your cap to fail is a weak or open neutral in your house's wiring that doubled the input voltage. But if that happened, you would notice other things, like light bulbs exploding..

Again, the cap in your TV was likely just a bad run of caps at least, and a poorly spec'd cap at most. Nothing caused it to fail, nor will the failed cap damage any LEDs. Further, shorted LEDs won't harm your AC bridge capacitor.

On a totally different subject, YES, lowering the back light in the menu (particularly in LED TVs) will extend the life of the back light exponentially. When I repair LED TVs for failed back lights, I always modify the LED drive circuit so that the LED array runs between 30 and 40 percent lower in wattage regardless of where the customer sets it or if it defaults back to full brightness. Customers don't notice and I don't have to worry about them coming back in warranty.

[Phil Allison]

John Robertson wrote:

> >>>
> >>> The peak current value depends on the transformer's winding
> >>> resistances and any series inductance.
> >>>
>

> Hi Phil,
>
> My point was one has to be careful with what cap you choose to replace
> the original. Too low an ESR may over stress the diodes.
>

** In the example given, the winding resistances of the switching tranny are gonna be way higher than the electro's ESR, overriding any effect changes in that value could cause.

Also, actual ESR is not clear from the published data as makers typically quote maximums at room temp. New caps may be half or less that value and in any case ESR drops with rising temp - by a factor of FIVE times more of the temp rises to 100C.

> If the input is
> protected with a varistor or similar then indeed,

** AC side thermistors or resistors help with inrush surges, but then normally have very little effect afterwards.

For practical purposes, the average value of the rectified current is all that matters, which does not change with C or ESR value.

Of course, the diode must be fast enough to handle the switching speeds with low loss.


.... Phil

[KenO]

ohg,

"There's no "design" issue involved with your particular failure. It's a simple full wave bridge and bypass filter."

Thank you for the clarification.

Since this is an inexpensive TV do not want to over do things so the Panasonic is not cost effective.

You have way more experience than I do so when you say "the cap in your TV was likely just a bad run of caps at least..." I believe you!

Am thinking about using a Nichicon replacement and have not found any problems with this brand.

Any experience with them? and last is it even cost effective to upgrade any aspect of original 47uF 160V 105C cap?

Thanks again for your comments!

[ohg...@gmail.com]

Cost effective? Ken, the cost to ship a capacitor of this type will far exceed it's cost.. If you're buying a few hundred thousand of them, then pennies matter. The cheapest Nichicon 47/160 radial I could find at Digikey that was in stock and could be purchased one at a time is 0.61c each. The cheapest Panasonic is 0.71c.

A 68/200v 105 Panasonic with 10K hour life is $1.14

http://www.digikey.com/product-detail/en/panasonic-electronic-components/EEU-ED2D680/P13526-ND/1086751

In any case, you're going to have to pay shipping. Does the cost of the cap actually matter?

[KenO]

ohg,

A Very Belated Thanks for doing my homework!!! Have been on extended holiday and now trying to get up to date.

"Cost effective? Ken, the cost to ship a capacitor of this type will far exceed it's cost.. If you're buying a few hundred thousand of them, then pennies matter. The cheapest Nichicon 47/160 radial I could find at Digikey that was in stock and could be purchased one at a time is 0.61c each. The cheapest Panasonic is 0.71c.

A 68/200v 105 Panasonic with 10K hour life is $1.14

http://www.digikey.com/product-detail/en/panasonic-electronic-components/EEU-ED2D680/P13526-ND/1086751

In any case, you're going to have to pay shipping. Does the cost of the cap actually matter?"

Agree with everything you said!!!

Wondered why I did not find that cap when dl the Pan cap catalog until dl the data sheet from Digi-Key http://industrial.panasonic.com/cdbs/www-data/pdf/RDF0000/ABA0000CE86.pdf and saw the red box with Discontinued.

Question: Is the 47uF 160 V cap common to Samsung(or other Mfrs) Power Supplies?

If yes am thinking about ordering a few more.

Thanks again for all your help!!!

Ken