Infomir MAG250 IPTV Box PSU problems

[dicky96]

Hi guys

This post is as much about desoldering techniques as fault finding but anyway....

I have two of these MAG250 boxes here, both faulty. After initial fault finding I determined one of them has a short (low resistance actually about 12ohm) on the 3.3V rail and the other has a similar problem on the voltage rail marked vCORE.

I'm just looking at the 3.3V fault box at the moment. I wanted to desolder the large inductor marked 3R6 to see whether the problem on the 3.3V rail is in the 1597A buck regulator or on the load side. The inductor measures too low resistance for me to tell from my DMM with it in circuit

Once removed I would power the 3.3V rail off my bench supply in CC mode and see what gets hot if the fault is on the load side..

What has me wondering is what is the best technigue to unsolder that large component without accidentally unsoldering smd components near to it? I have a hot air desolder station 858D and 60W temp controlled iron at my disposal. I'm wondering if it may be easier to remove the 1597A instead. Or just power the 3.3V rail with it still in circuit.

One of the smd caps from the 3.3V rail to 0V looks a little discoloured, it occurred to me that may actually be the fault anyway

Rich

[dicky96]

Re: Infomir MAG250 IPTV Box PSU problems

In lack of any suggestions here, I had a go and removed the inductor on the 3.3V rail no problem using my hot air desoldering station.

The low resistance was on the load side of the inductor

Using my bench PSU in CC mode and applying voltage to the 3.3V rail with the inductor removed I gradually ramped up the current to see what was getting warm. At the full 3A (V was now about 2V or so) the only thing getting warm is the big (STi7105-BUD) bga chip on the centre of the board.

I decided I would have a go at removing this with the hot air station just to see if that was really the cause of the low resistance but to be honest I couldn't even get it to come loose with any reasonable amount of heat applied. Also ther are a lot of smd components on the other side of the pcb below the STi7105 and I was worried they may fall off with too much heat

However what I did notice, once the board had cooled back down is the 12 ohm low resistance on the 3.3V rail is now about 7 ohm so that has changed.

Out of interest I googled the STi7105-BUD and found a couple of interesting things.

http://pdf1.alldatasheet.com/datashe...S/STI7105.html

Firstly it does not seem to be a programmable device but uses external Eeprom/Flash

Secondly I can get them for 7.50 euros inc postage
https://es.aliexpress.com/store/prod...3-d665507fd56b

The other MAG250 seems to have a similar problem, when powered from the bench PSU I can actually get this one up to 3.3V at about 2.0A in CC mode, the crystal even oscillates though nothing else seems to happen and the STi7105 gets hot

Sooooooooo....... I was thinking of having a go at fixing them, as much for the learning experience as anything else, and of course the ability to take on more similar repairs

Question is - what kit and techniques would I need to replace this chip?

Assuming I get the chip off, how do I get the replacement one positioned correctly to resolder?

This is all a bit new to me, but very interesting

Pics of the STi7105 and the underside of the board showing the components beneath it


Rich