Xtrend 9200 PSU repair / Choosing Electrolytics

[NivagSwerdna]

I am currently trying (and failing) to fix a PSU from a satellite receiver...

My assumption is that it is Electrolytic capacitor failure (and a couple of them do look physically very bulged etc...)

Anyway this got me to thinking....

Say I want to replace a 22uF 10V Electrolytic but in my parts drawer I only have a 22uF 16V Electrolytic.

What are the implications of making that swap?  Is that OK because 16V > 10V or does 16V imply a higher ESR (and possibly other parameters) so could be a problem.

Finally, it seems Electrolytics come in series from the manufacturer... e.g. Panasonic FC series.... Which characteristics are important when choosing a series to use for replacement PSU caps?

Thanks in advance
Nivag

PS
I came across this... https://industrial.panasonic.com/cdbs/www-data/pdf/RDF0000/DMF0000COL91.pdf so there seem to be a few dimensions to this... e.g. impedance, lifetime, size etc... Mind boggling.

[Kleinstein]

Usually taking a capacitor with slightly higher voltage rating is not a problem. The main point that changes is the physical size - so if it fits, it't OK.

With switched mode power supplies one might need to use low ESR caps because of the ripple rating. But usually tolerance on caps are not that tight, so that there is no need for an exact match. So usually a low ESR cap is OK.

Even for behind LDOs the ESR rating is usually not that critical. Still it might be a good idea to replace an low ESR cap with a low ESR one and a "normal" one with a normal.

[bob225]

Never had a issue, As long as the replacement are good quality branded items, I have come across where caps have been discontinued by manufacturers its often more time/cost effective to use the nearest spec (voltage wise)

tbh stb psu's are never built to a fine tolerance, they are built to a price for mass production - depending on the manufacture they may just be a off the shelf universal psu

What is the make and model ? there are know common issues with most brands

[NivagSwerdna]

What is the make and model ? there are know common issues with most brands

It is a PSU from an XTrend Satellite Receiver box.  The Caps are "Sam Young" whatever that means!

[tkamiya]

I have been repairing a lot of ham radio equipment from 70s and 80s.  Among dozens I went through, reason for failure being electolytic caps were/are very low.  Many times, failures were due to switches, relays, and cold solder joints.  So I wouldn't automatically assume I have to replace all caps. 

As far as spec goes, it's all depending on where in the circuit the cap is used.  For decoupling, anything will really do, provided you have sufficient headroom in voltage part.  Many manufacturers tend to make this too tight, as lower the voltage capability, cheaper the part cost.  I usually try to keep it at least twice as much of power supply voltage.

In short, what you suggest is perfectly acceptable.

[bob225]

sam young - Chinese caps, most stb suffer with heat issues the input side tends to go first that then takes out the output side caps, if its smps go through the sense circuit (flyback, etc) as well

I have done a few humax boxes the caps let go and they blow the main reg and sometime the diodes on the output too

[NivagSwerdna]

I don't have a schematic which makes things worse and in fact it could possibly not be a PSU issue at all but that is known as common failure mode in Xtrend 9200.

I can see some writing on the board for 5V and 3.3V and those voltages appear present but there are other voltages too... 12V and 24V on some pins but I can't tell which is correct or not.

Although not my picture this appears to be the same....



Other people seem to have issues with the 4700uF in the middle but mine seems fine; replaced it anyway and it didn't help.

I did remove and test the two small transistors towards  the bottom right and they appear sane.

Possibly beyond economic repair as the box, although quite expensive originally, is now 5 yr+

[NivagSwerdna]

Any idea on the principle of operation?  Does it have two modes controlled somehow by the 12H, 5H?

I checked the diodes... look OK.... the solid state components appear to be

FQP20N06L MOSFET which appears good, two BiPolars which appear good and two LDOs OK

There is a KA1M0565R but I don't know quite how to test that and I imagine I wouldn't be getting any output if it is broken. And a dual diode... not sure of make or part.

I will re-cap the whole board and try again.  Will need to wait for parts.

[bob225]

The 220uf 10v beside the heatsink may be a issue due to heat (right hand side 3rd cap down)

edit. the model number is on the left hand side IPS80W, the stb is also branded as Clarketech (one word)

[NivagSwerdna]

That was a stock picture; not mine.  But I agree!

I've been learning about SMPS and I hadn't appreciated that actually I get two SMPS for the price of one!

The top section has a OB2354AP; the bottom section 1M0565R. 

I have been through all the capacitors I removed and I think I am actually going to put most of them back as they test OK.  A few (4?) are out of spec so I will replace them.

I started to suspect the dual diode so de-soldered it and tested it and made a bit of a mess with the pads as I removed it which is very  anyway it tests OK

I'll try and reassemble when I have some 22uF 50V caps to replace the ones I'm unhappy with... currently in the post from Farnell.

[NivagSwerdna]

Replaced all the electrolytics with nice new FC series ones.... and it doesn't work... symptoms appear similar to when I started... in fact I'm not seeing 5V where I did before so I think it is worse 

I guess I don't have a future as a SMPS repair technician. 

... or maybe I do....  because for some reason it is now working (after a fashion)... the satellite receiver now boots and the voltages I understand are within tolerance.

One thing remains though the double diode is running VERY VERY hot... maybe I should swap that?  Is running hot a failure mode for double diodes?  or just an indication of load?

Do you think NXP BYV32E-200 Dual Ultrafast Power Diode 200V TO220AB would be a suitable replacement?

UPDATE: oh well it has stopped working again... I measure voltages on the secondary diodes I can safely get to but the box has stopped powering up... Guess I will have to rethink some of this.  I measured 48V on one of the secondary diodes which I thought was surprisingly high but I am unfamiliar with satellite receivers so maybe that isn't that surprising.

[Robinson]

Hi Everyone,
NivagSwerdna, have you solved it?
I seem to have the same issue with my Clarketech ET9000.
Only the blue diode is blinking.

I have already replaced most, if not all, capacitors on the PSU but it did not help. Perhaps something else is broken on the PSU or on the mainboard. The security check battery seems to be OK.

Does anyone still own 9000, 9200 or 9500 models and is willing to help?
If so, can you please measure voltages coming out of the PSU with the connectors dosconnected and connected?

I have the following values with connectors disconnected (my values in brackets):

1 - 5V (0V)
2 - 5V (0V)
3 - GND
4 - 3,3V (0V)
5 - 3,3V (0V)
6 - GND
7 - 12V (13V)
8 - 23V (0V)
9 - GND
10 - 32V (1,3V)
11 - 6V (0V)
12 - 3,3V (3,2V)
13 - 3,3V (0V)

And with both connectors connected:

1 - 5V (0V)
2 - 5V (0V)
3 - GND
4 - 3,3V (0,27V)
5 - 3,3V (0,27V)
6 - GND
7 - 12V (12,30V)
8 - 23V (0V)
9 - GND
10 - 32V (0V)
11 - 6V (0V)
12 - 3,3V (3,2V)
13 - 3,3V (3,15V)

I doubt the values are correct.
I also doubt the issue is due to security board dead battery because I have probably used the latest bootloader version with the security check removed.

[NivagSwerdna]

Nope. I gave up in the end as I had changed receivers.