Author Topic: [Solved] Repair / DC 3.3V 2.5A Open Frame Power Supply / Network Switch (Read 15244 times)

xan · « **on:** February 25, 2017, 01:38:35 pm »

Hi everyone,
i need some help and tips to fix an Open Frame Power Supply DC 3.3V 2.5A what i can do 1st ? fuse it's ok....

Leader Electronics
Input Voltage: 100 ~ 240VAC ~ 50/60Hz. 0.3A
Outputs: DC +3.3V === 2.5A
MODEL: SB10-033100-10

Thank's for support!

singapol · « **Reply #1 on:** February 25, 2017, 01:52:17 pm »

The cap arrowed in red has a bulged top. Check esr and capacitance.

xan · « **Reply #2 on:** February 25, 2017, 02:19:26 pm »

i don't have esr tool $:-\$ some workaround? i test capacity right now!

Thank's

bktemp · « **Reply #3 on:** February 25, 2017, 02:29:55 pm »

If you are not sure which capacitor is bad, replace all capacitors. The mains rated capacitor rarely fails, so it is probably fine. The secondary side capacitors and the small capacitor next to the SMPS controller ic fail quite often.

xan · « **Reply #4 on:** February 25, 2017, 02:46:02 pm »

Ok i got it! and with my multimeter can i do some cap measurements? or not?(i'm not so expert)

xan · « **Reply #5 on:** February 25, 2017, 04:34:44 pm »

ok green caps is gone it marks teapo 1000uF 10v.. can i use as replacement 1200uF 16v?

Bushougoma · « **Reply #6 on:** February 25, 2017, 04:59:44 pm »

A bulged cap is a replace on sight part no need to test it. It has indeed failed.

ESR and capacitance measurements are independent of one another I have measured bad capacitors that had good capacitance but high ESR and vice versa.

Electrolytics typically have a tolerance of +-20% so a 1200uF will be fine.

Just make sure it's a low ESR type and it's ESR is equal to or less than the original capacitors and it's max ripple current is equal to or greater than the originals. Check the datasheet for the original cap and your replacement cap to verify. General purpose caps can't be used on the high frequency output of a switch mode power supply.

I would replace both of the secondary caps and the small bootstrap (start) cap near the PWM / Switching transistor DIP IC that may be all you need to get it working.

If not these integrated supplies with the PWM IC and the switching transistor in one package are pretty easy to troubleshoot due to the low component count.

xan · « **Reply #7 on:** February 25, 2017, 05:12:24 pm »

Thank's really understood

for the green caps could you check sheets ?

original -> http://www.teapo.com/WebSiteFile/Products/Product_Data/SC.pdf
new -> http://www.nfjapan.com/datasheet/KZG.pdf

(sorry for ask but i'm really newbie in this kind of things and i want to understand!)

Bushougoma · « **Reply #8 on:** February 25, 2017, 05:28:36 pm »

No problem it can be a bit confusing.

On the Teapo datasheet provided ESR is listed as Z which is short for Impedance.

I've highlighted the critical specs you first find the working voltage of the cap at the top, then the capacitance on the left, then go to the proper column. You'll notice the ESR (Z) and ripple varies based on the capacitor case size so you'll have to measure the cap with a digital caliper or ruler and use those measurements to determine it's ESR and max ripple current.

Ripple on the replacement capacitor should be => original capacitor. ESR (Z) on the replacement capacitor should be <= original capacitor.

I can't find a 1200uF cap at 16 volts in the Chemicon datasheet you provided. In that series 1200uF only comes in 6.3 volts which won't work since the original capacitor has a working voltage of 10 volts.

xan · « **Reply #9 on:** February 25, 2017, 05:33:48 pm »

Ahhh understood.. i need to check more few things.. i keep you updated later... Thnak's again!

xan · « **Reply #10 on:** February 26, 2017, 01:58:37 pm »

Ok guys finally!!
old caps green and new cap blu but .... SAMXON cap have 85° instead of 105° (same 1000uF - 10V).. should be ok ? or?

can you check sheets for me?

Bushougoma · « **Reply #11 on:** February 26, 2017, 02:28:31 pm »

You can't go down in temperature rating from the original capacitors only up. You'll notice most caps will have a temperature in Celsius printed on the side (85C and 105C are the most common) that is the max operating temperature they are rated for. You can replace 85C rated caps with 105C but not the reverse.

I've never seen an 85C cap from a top tier cap manufacturer that was low ESR without even looking at the datasheet I would say that blue 85C capacitor was general purpose.

xan · « **Reply #12 on:** February 26, 2017, 02:30:10 pm »

damn ok... i will look for another... sorry for asking.. i keep you updated again!

Monkeh · « **Reply #13 on:** February 26, 2017, 04:28:08 pm »

Go to ebay, spend €2. Fixed.

Armadillo · « **Reply #14 on:** February 26, 2017, 04:50:59 pm »

Why don't you install the samxon Cap in first and see if it fully repaired the power supply or there are some other components to order at the same time, in the name of efficiency. However, there are also economics you should consider aside from electronics repair.

xan · « **Reply #15 on:** February 28, 2017, 01:25:08 am »

mm right! i will try, i think maybe i will order some other parts... tnx.

Armadillo · « **Reply #16 on:** February 28, 2017, 07:48:54 am »

Can you confirm that the IC is A6351. Please take a close up if you can.
What are the symptoms when you power up the power supply?

[The open frame exposes lethal voltages, do not touch when power is up, always unplug and discharge capacitor, so please be extremely careful with it as capacitor may be charged. The best is to power up with 100W incandescent bulb in series with the mains].

If you are not sure of what you are doing, please do not do it. thanks.

By the way, what instruments or meter do you posses?.

xan · « **Reply #17 on:** February 28, 2017, 01:16:13 pm »

Yes thank's i will not touch it!!!

if i'm not feel secure i will ask before! i have a cheap multimeter.. excel multimeter.. for now i still wait to find right capacitor for substitution, from old pcb.
The Symptoms was simply, network switch power led blinking green and network on LAN was Down!

Armadillo · « **Reply #18 on:** February 28, 2017, 02:03:44 pm »

There are effectively 2 capacitors at the output, so whether you install the one capacitor or not, it should still read 3.3vdc. To troubleshoot, please measure if only you can, as shown in the picture below.

Use multimeter, put to "DC voltage range" measure:

xan · « **Reply #19 on:** February 28, 2017, 02:25:09 pm »

Damn i just desolder bad caps... how i can test 3.3v now

? i will resolder it back but i think there isn't the right voltage there... obiviously!
IC on screen. Tnx

Armadillo · « **Reply #20 on:** February 28, 2017, 02:33:55 pm »

you do not need to solder the bad cap, just leave it, because there is already another cap on the board. It is also 1000uF 10V. We are just troubleshooting the problems now, to identify other areas causing the problems. You can install the new cap when it come. For now, don't need to, just leave it un-installed, vacant.!

Do you understand what I mean, the 2 capacitors are effectively in parallel across a choke. So one down, still not a big problem as far as DC is concerned.

xan · « **Reply #21 on:** February 28, 2017, 02:37:53 pm »

Ah.. Ok You welcome ! sorry for asking i'm newmbie

but i can't do measurements right now i'm not on my lab... right now i will do it later! i keep you updated ! Thank's for the tips.
Later

xan · « **Reply #22 on:** March 01, 2017, 11:25:23 pm »

i still wait to be back to my lab on this weekend so i can't test 3.3 lines ... but can i ask you somethings armadillo?
1st) what do you mean with:

Quote

[The open frame exposes lethal voltages, do not touch when power is up, always unplug and discharge capacitor, so please be extremely careful with it as capacitor may be charged. The best is to power up with 100W incandescent bulb in series with the mains].

2nd) why in this board there are 2 caps in parallel across the chocke? what are their functions if i can test it or use it without one ?

Thank's (i'm just so curious to understand!)

Armadillo · « **Reply #23 on:** March 01, 2017, 11:49:08 pm »

Quote from: xan on March 01, 2017, 11:25:23 pm

1st) what do you mean with:

Quote
[The open frame exposes lethal voltages, do not touch when power is up, always unplug and discharge capacitor, so please be extremely careful with it as capacitor may be charged. The best is to power up with 100W incandescent bulb in series with the mains].

2nd) why in this board there are 2 caps in parallel across the chocke? what are their functions if i can test it or use it without one ?

Reply #1: Please refer the load lamp safety at http://www.bristolwatch.com/load_lamp.htm . The purpose is to protect the device under test by limiting the current in case the device is shorted somewhere e.g. to minimize possibility of output transistor or mosfet from being damaged or burned. The power supply is normally segregated into 2 sections, the primary section and the secondary section. The primary section after the diode bridge is about 300vdc. Copper tracks underside of the board or the components leads exposes this high voltage. Accidental touching it can be lethal.

Reply #2: You can view it as 2 stage filters. The choke / inductor with the 2nd capacitor further improve the DC filtering of ripple voltage. The choke act as a resistor to high frequency switching ripple current because the inductive reactance is directly proportional to the frequency, stop high frequency noise / ripple while allowing DC to cross to the 2nd capacitor. After the choke, then the 2nd capacitor further filter it. For troubleshooting purposes like measuring the 3.3 vdc, single stage filter will do. In case you want to measure the performance of the power supply ripple voltage, then you can use a oscilloscope to measure it after the 2nd stage filter, for example.

Hope these clarify your questions.

xan · « **Reply #24 on:** March 18, 2017, 06:18:35 pm »

Ok Finally i was back to my Lab (after 10 days of full workload

) i change the capacitor with a new one and i made the measurements! here the screens... from power supplies come a strange sound like somethings it's activated and then deactivated .. "tik tik tik tik tik tik tik tik tik " still have blinking led ... -_- i measure at the -/+ terminals about oscillating from 2.6 to 2.9 V on both -/+ .... Ufff... some tips guys

??

!SCREEN DOWN!

Armadillo · « **Reply #25 on:** March 18, 2017, 07:45:39 pm »

Measure Resistance ohms across the IC pin 1 and 8, check whether it is shorted or low ohms.

xan · « **Reply #26 on:** March 18, 2017, 09:40:19 pm »

Ok Armadillo... why i should measure $\$\Omega\$$ across pin 1 and 8 can you explain?? (u know i'm newbie

)
is this my IC datasheet ? https://www.digchip.com/datasheets/download_datasheet.php?id=1913236&part-number=IRIS-A6351

(i will do measurement tomorrow morning.... do i have to take care about some high voltage during measurement??)

Thank's for suppport!

Bushougoma · « **Reply #27 on:** March 19, 2017, 12:32:05 am »

Pin 1 and 8 is the source and drain of the internal switching FET that drives the transformer. If it was open you wouldn't have any output if it was shorted the fuse would have blown but it couldn't hurt to check it.

It sounds like the supply is in hiccup mode. It is starting up, sensing a fault, shutting down for a predetermined time, and attempting to start again.

This can be due to an overload (a short on the output) or it could be that the primary winding dedicated to supplying the PWM controller isn't functioning.

It sounds like the bootstrap circuit is working but the primary dedicated to supplying the IC after the PWM controller starts isn't. Check the 1/2 watt resistor and diode to the left of the transformer it is responsible for rectifying the PWM controllers VCC also check the small capacitor right above the IC.

The bootstrap circuit consists of a rectifier diode connected to the AC mains, a resistor, and a capacitor. The resistor trickle charges the capacitor and it's voltage slowly rises once it hits a certain voltage the PWM IC comes to life and begins switching the transformer. The capacitor will maintain the voltage needed to run the IC for a short time once it starts switching it's supposed to supply power to itself through the dedicated primary shown in the datasheet. If it doesn't the voltage on the capacitor falls the IC shuts down and the process repeats again.

Armadillo · « **Reply #28 on:** March 19, 2017, 02:42:08 am »

Quote from: xan on March 18, 2017, 09:40:19 pm

Ok Armadillo... why i should measure $\$\Omega\$$ across pin 1 and 8 can you explain?? (u know i'm newbie )
is this my IC datasheet ? https://www.digchip.com/datasheets/download_datasheet.php?id=1913236&part-number=IRIS-A6351

(i will do measurement tomorrow morning.... do i have to take care about some high voltage during measurement??)

Thank's for suppport!

Turn off the AC power, pull out the plug. You MUST discharge C1 for your safety. After the proper discharge, then you can start to measure the ohms.

Please note that this power supply is HOT, even the IC is HOT, meaning it is connected to the AC mains, of high KVA, and can cause serious damage. So be very careful with it.

Because the circuit tried to start up, but there is a over-current situation that forbid it to start up [when pulse voltage at pin 1 exceed 0.7 volt]. Either the mosfet is shorted or the PWM frequency has degraded too low. You will most likely be looking at either the IC or the transformer. Either one will be hard to get. The IC is obsoleted. [ok one other measurement, you can measure the current sense resistor R7 that it is 0.9 ohms and is not open circuited, for double sure its not external circuit causing the problem]

And because the 3.3v power supply off ebay is getting so cheap, it is not worth the efforts to repair it, really.
But if this is just for your learning, I will say, please find yourself a more safer circuit to start off with.

Edit: you know why I ask you to test it without buying the capacitor? yeah! because there may be some other important and deterministic parts to consider, like the IC or transformer.

Bushougoma · « **Reply #29 on:** March 20, 2017, 05:45:55 am »

Switching transformers rarely fail and shouldn't even be considered until the usual suspects are ruled out.

Quote from: Armadillo on March 19, 2017, 02:42:08 am

And because the 3.3v power supply off ebay is getting so cheap, it is not worth the efforts to repair it, really.

As long as that supply has a proper EMI filter like this one does many cheap Chinese supplies don't and spit noise back onto the A/C line. Along with abysmal separation on the PCB between the primary and secondary don't even get me started on the construction of the transformers.

If you do go that route purchase your power supply from a reputable manufacturer.

Monkeh · « **Reply #30 on:** March 20, 2017, 05:48:05 am »

Elephant in the room: Which capacitor(s) did you replace and what did you replace them with?

james_s · « **Reply #31 on:** March 20, 2017, 05:53:16 am »

At least 90% of the time I work on power supplies like this, replacing the electrolytic capacitors is the only thing they need.

If it goes tick tick tick that strongly suggests a shorted part in the output. Check the diode(s) and also look carefully for solder bridges you may have created while working on it.

xan · « **Reply #32 on:** March 20, 2017, 07:38:40 pm »

ok guys so i have to measure resistence across this 2 pins ? right ? (be more specific ...)
how many Ohm on my instruments i need to set ?

Thank's!

xan · « **Reply #33 on:** March 21, 2017, 10:20:19 pm »

ok C1 discharged correctly with resistor and i have this reading across pin 1 and pin 8 on IC .632 $\$\Omega\$$ (scale 2k on my instrument)
(image on previous post is wrong... attached right pin configuration..)
so now, what's next ??

james_s · « **Reply #34 on:** March 21, 2017, 10:29:56 pm »

I think what is next is you need to study switchmode power supply design so that you understand how the circuit works, it's very difficult for someone to walk you through the whole diagnosis. If it's not working you have a bad component somewhere.

xan · « **Reply #35 on:** March 21, 2017, 10:33:13 pm »

don't worry let's try!!!!

Monkeh · « **Reply #36 on:** March 21, 2017, 10:33:36 pm »

I shall repeat: Which capacitor(s) did you replace and what did you replace them with?

xan · « **Reply #37 on:** March 21, 2017, 10:37:17 pm »

i changed this one in attachment! damaged: teapo 1000uF 10v -> new: Rubycon 1000uF 10v

Monkeh · « **Reply #38 on:** March 21, 2017, 10:38:02 pm »

And the other half of my question?

xan · « **Reply #39 on:** March 21, 2017, 10:40:15 pm »

sorry

refresh and read again up pls..

Monkeh · « **Reply #40 on:** March 21, 2017, 10:40:55 pm »

What type of Rubycon? They make quite a few different parts..

xan · « **Reply #41 on:** March 21, 2017, 10:42:22 pm »

Rubycon 1000uF 10v ZLH do need more specs?

Monkeh · « **Reply #42 on:** March 21, 2017, 10:44:04 pm »

Okay.

What value is the other large capacitor on the output? If it's quite close to 1000uF (it probably is), please replace it as well. If it still does not remain running, replace the small electrolytic next to the IC.

james_s · « **Reply #43 on:** March 21, 2017, 10:49:20 pm »

Without access to an ESR meter, the first thing I'd do on a power supply like this is replace *all* electrolytic capacitors on the secondary side. The bulk filter on the primary side is usually ok unless it's bulging or leaking.

xan · « **Reply #44 on:** March 21, 2017, 10:50:38 pm »

ahh ok.. i don't have ESR meter...

anyway... yes it is 1000uF 10 v Su'scon... the small one 22uF 50v Do You mean that above iC ?

Monkeh · « **Reply #45 on:** March 21, 2017, 10:51:40 pm »

Quote from: xan on March 21, 2017, 10:50:38 pm

ahh ok.. i don't have ESR meter... anyway... yes it is 1000uF 10 v Su'scon... the small one 22uF 50v Do You mean that above iC ?

Yes, that little 22uF one. Replace the other output cap first, if it still fails, replace the little one. If after that it fails, frankly throw it in the bin and go buy another.

Quote from: james_s on March 21, 2017, 10:49:20 pm

Without access to an ESR meter, the first thing I'd do on a power supply like this is replace *all* electrolytic capacitors on the secondary side. The bulk filter on the primary side is usually ok unless it's bulging or leaking.

The bulk filter we don't care about, the IC supply we do.

xan · « **Reply #46 on:** March 21, 2017, 10:54:25 pm »

ok got it.. can you help me to find on ebay an open frame pSu as a replacement?? or something provide an input like this ?

Monkeh · « **Reply #47 on:** March 21, 2017, 10:57:01 pm »

You've got a huge empty case, it's not hard to find something which fits. And frankly, switches are cheap.

xan · « **Reply #48 on:** March 21, 2017, 10:58:45 pm »

ok will look for some on ebay but i still want to repair this... just for understand..

Armadillo · « **Reply #49 on:** March 22, 2017, 12:17:30 pm »

If I recommend you the GBP$2 power supply, I know I will receive some comments from few free people here who just seems to enjoy on commenting others. For the cheaper ones, therefore you will have to look for yourself.

So, I will suggest you the UL TUV CB CE safety listed Meanwell made in Taiwan power supply. This is rated at 3A. So you don't need to spend nearly equivalent amount to repair the existing power supply to get quarter life or half life out of it. In my opinion, Meanwell power supply is known brand in my region. So you know you can get full life out of this all new power supply [new capacitors, new transformers, new diodes, new mosfets etc..]

http://www.ebay.com/itm/1pc-Switching-Power-Supply-RS-15-3-3-3-3V-3A-9-9W-62-5x51x28mm-Mean-Well-/141684943081?hash=item20fd14a8e9:g:fUYAAOSwrklVcHa-

I will let others to recommend to you the more perfect and better ones for you, "as usual".

You will have to be imaginative on how to do the mounting and how to hook this up to the existing "power socket" etc. However, please do this at own risks and own responsibility.
We cannot be walking you through every steps but remember this - "Safety First", still you don't know, just stop and ask!.

Bushougoma · « **Reply #50 on:** March 23, 2017, 11:11:32 pm »

I've used Meanwell before they make quality supplies for a reasonable price.

If you stick with the known brands you'll be ok just stay away from the dirt cheap no names they will likely have crap caps and no EMI filtering.

Phil_Tech · « **Reply #51 on:** August 30, 2017, 01:07:32 am »

I've just repaired one of these PSUs, the output was at 2.7v. Replaced C3 10uF 50v (the little one just above the 8pin controller chip A6351). Now output is back to 3.3v. The electro didn't look bad but did not register on my ESR meter.

xan · « **Reply #52 on:** June 27, 2018, 08:22:22 pm »

Well done guys!

thank's everyone for support! fixed with a new cap 50v 22uf! device back to life!


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: [Solved] Repair / DC 3.3V 2.5A Open Frame Power Supply / Network Switch (Read 15244 times)

Share me