[SOLVED] Talkswitch VOIP Phone 9133i repair

[edy]

Hi folks,

My office VOIP phone was acting up today and I decided to unplug it and plug it back in again (as I usually do) to reboot it, and when I plugged it back in it was DEAD. I figured it had something to do with the power. I swapped phones and my power supply was fine, so it was the phone itself.

Here's some pictures on what it looks like. I'll use a few posts in this thread to be able to upload all the photos. May be interesting to see the construction of this circa 2009 VOIP phone.

[edy]

Here are some of the processors on the board in case you were wondering. I see as follows:

Broadcom BCM1113KPBG CK0912 P30 914582 R2A
https://chipworks.secure.force.com/catalog/ProductDetails?sku=BRO-BCM1113KPBG&viewState=DetailView&cartID=&g=
(VoIP Ethernet CPE Engine)

Samsung 831 K4S641632N-LC75
http://www.datasheets360.com/part/detail/k4s641632n-lc75/-1298125115225187539/?alternatePartManufacturerId=0
DRAM Memory Chip

Based on the proximity of these two and the multiple connecting traces, I assume this is memory being directly used by the chip.

Next I see a SPANSION S29AL016D90TF102 832FF421 H, which according to Google and https://www.usbid.com/parts/S29AL016D90TF102 is a Boot-sector flash memory.
http://www.spansion.com/cn/Products/Documents/Quick_Reference_Guide_en.pdf

It's also joined directly into the Broadcom chip on the other side. So does one chip provide the boot-up firmware code to initialize the VOIP chip, and the other memory chip is used as operational memory? Not sure how this stuff works, maybe someone can explain. 

[edy]

A few more photos...

I see a DELTA LFE9214 0912 (Manufactured 2009, 12th week?). I can't figure out what this part is. I found it all over the web but it never explains what it does, just links to places you can buy them. Any clues? It's up near the Ethernet port so I figure it has to do with that?

The last 3 photos show the underside of the board and how they managed to connect it to the keypad. Looks like one of those elastomeric connectors or zebra strips. It connects to the board on one side, then pressed through that slot in the case to probably some board that houses the keys. Not fond of these types of strips, hopefully it will still work once I put it all back together.

[edy]

Ok, so basically it was a simple fault find... Like I mentioned earlier, it happened when I pulled the plug and then reinserted it. No power. As you can see by the photos, the connector appears to have lifted off together with the copper pad, and it ripped off the trace that it heading towards that SMD capacitor (what it looks like) to the right.

Now this is a bi&tch to fix, because I see a via in there as well, and now I have to scratch out the PCB mask/silkscreen/green coating to get down to the copper of that trace that is heading towards the SMD cap. Otherwise, I will have to bridge over to the cap.

[edy]

Here's my sh!t repair job, but it works.

First of all, let me explain what happened. The first photo shows that I picked up the pad to see what was underneath. I was checking to see if anything was going through the board, or if any other traces were heading off in other directions. Nope.... Just towards the right (towards that SMD capacitor pair which are in parallel.... the orange one and the black one). I scratched off the solder mask/green coating to try to get to the bare copper of the trace to solder to.

The second photo shows that I was ready to solder the pad from the plug to the trace. However, as I was heating up that tiny trace, the piece left of the via stitch hole broke off. F&ck!

So I decided to use a wire to join the power connector to the SMD capacitor directly. My thinking was that the wire would provide some flexibility since the connector can still move. Unfortunately I used a bit too heavy a gauge of wire which is pretty flexible in longer lengths, but this tiny little piece is too stiff to flex so it made the connector quite rigid.

Bottom line, I have to be very careful to plug in the power adapter and if I need to reboot the phone in the future, I'll unplug it from the wall instead. Hopefully this patch job will last a while. The trace may still be connected to the via stitch hole from the other side (the SMD capacitor side) so it is still getting voltage. Hopefully I didn't miss any connections.

The phone powers up and boots normally, so I'm sure it works fine. I haven't tested it fully because I'm not at the office and plugged into the network, but it's unlikely that there is anything else wrong with it. Also, if the power wasn't getting to the entire system I doubt it would boot. Something would definitely be wrong with it. I just hope my patch job doesn't melt or short something out. The wire gauge is fairly thick.

Any thoughts how this can be repaired differently? I'm not sure what options one has when a copper trace separates from the underlying board and rips, except to try and bridge the PCB traces by wiring directly across the connected components.