Computer Successfully COOKED!

I must say, this was one of the strangest repair jobs I've ever had: You may have heard the background: My wife's three year-old Sony Vaios 'puter suddenly came up with really serious graphics problems. A quick Google revealed that lots of computer have this problem (lots of brands... including Apple). We took the computer down to the Sony store and they got all legalistic on us. Yes, it was clearly the video card. Yes, Sony had agreed to fix the video cards. BUT unfortunately Sony had not put our model number on their list of computers that they would fix. So they wanted to charge us $400 to fix the problem. No thanks Sony! Goodbye Sony!

Inspired by the iFixit poster on the wall of my shack, I did some more Googling. Soon I'm reading messages from all around the world recommending that I -- in one way or another -- cook the computer. But why? Is the problem IN the GPU chip? Or is it in the solder connections between the chip and the board.

Gents, you'll be pleased to learn that this is a SOLDERING problem. Looks like NVIDIA used a bad mixture of soldering types. There is apparently little blobs of solder on the baord, and little blobs of solder on the chip. LOTs of little blobs. Surface Mount to the MAX! But they used two different solders and this causes the connections to fail far more quickly than they should have. Here are the details:
http://s0.2mdn.net/2994486/Polycom_inquirer.html?rfp=http://www.theinquirer.net/inquirer/news/1004378/why-nvidia-chips-defective
Enter the oven solution. The idea is to simply re-heat the chip and make the solder at the connections melt again!

YouTub presents lots of different ways of doing this. I like the idea of using a bright lamp to administer the heat. I used a 120 watt Halogen beam lamp. The kind that you see on lights for the backyard. I tested its heating properties on a chunk of 60/40 rosin core solder. At about 1/2 inch it melted the solder in 30 seconds.

I put the beam on the TOP of the chip and gave it about 30 seconds of burn. Then I administered some heat sink compound, vacuumed out the fan, and put the computer back together.

IT WORKED! That computer is now working just as it had before. This was a very satisfying repair, but there was no real troubleshooting, so it wasn't really up there with my favorite fixes.  You can check out www.computerrepairtraining.org if you want to learn more about fixing computers.

I'll put one of those cheap laptop coolers underneath it. I'm guessing that this thing will go for another few years.

Thanks to all who provided advice. Thanks to ifixit for the inspiration. Thumbs down on Sony. BOOOOO! HISSSSSS!

Should I Put My Sony Computer in the Oven?

In a recent podcast I asked for help with my wife's ailing Sony Vaios computer. The graphics card has gone toes up. Sony acknowledges that they had BIG problems with the NVIDIA graphics cards, but even though our computer has an NVIDIA graphics card and has all the symptoms that come with the NVIDIA graphics card problem, because our particular model is not on their list of affected computers, they won't fix it. (Mental note: Don't buy any more Sony computers!)

Anyway, KB1SNG recommends putting the mother board in the oven:

As crazy as it may sound, you could try putting the motherboard in the oven.
Many people (including myself) have tried this with success.
I did it a while ago, and I can't remember if this ( http://www.overclockers.com/forums/showthread.php?t=606658 ) was the guide I used or not, but it should suffice. If I find the other link, I'll post it here.73,Nick LaPointe, KB1SNG

Rogier provides this useful info:

NVIDIA lost a lawsuit with regards to a faulty graphics Chip. Caused by the fact that when they started using lead free solder. This new solder turned out not as heat resistant as thought and tends to break loose causing the chip to fail.
He continues:
Hmm that's a bold move to put the board in the oven.
On the other hand there is little to be lost and you might give it a try.
Thinking of it I rather apply the heat locally. Isolate the Graphics chip from the rest and apply a short blast of heat to the chip.
Perhaps with a hot air paint stripper....

So what do you guys think? Should I pop it in the oven? Or apply some heat? I kind of like the idea of fixing a problem caused by lead-free solder (yuck).

Immediate Gratification: WSPR shows results very fast

So you build or modify a rig and you put up a new antenna and you then spend some time wondering if you're really "getting out," right? Well, wonder no more my friends! WSPR provides almost instant feedback on your work! Here is what the display looked like for my station a short time after I connected my modified transceiver to my new 30 meter dipole. Pond crossings were rare events before.

In my last post I asked if there were others running homebrew gear on WSPR. I am not alone! Paul, M0XPD, has put together a rig far more sophisticated than mine. Paul writes:

You ask if there are any others running h/b in WSPR...

My 50mW multi-mode beacon runs WSPR, DFCW and S/MT Hell on a 10-minute, frequency-hopping cycle.

It is entirely stand-alone - severing the umbilical cord to the PC gave me a really good feeling!

I'm currently working on a set of filters (BP and LP) to allow me to run on other bands (currently I'm on 30m).

Description on my blog: http://m0xpd.blogspot.com

A good radio morning at N2CQR

Today I had one of those really good ham radio mornings in the shack. My objective was to add one more amplifier stage between the balanced modulator and the power amplifier in my Roman WSPR DSB Direct Conversion transceiver. First I had to take the balanced modulator (trifilar transformer and two diodes in a singly balanced configuration) off the board and move back, closer to the AF input jack -- I needed to make space for the new amplifier stage. For that stage I consulted Doug DeMaw's books and Wes Hayward's EMRFD. I went with a feedback amp using a 2N3904 in Class A.

With coffee brewed and with my Drake 2-B tuned to the very congenial DX-60 net (75 meters AM on Sunday mornings), I turned to the junk box. It was like meeting old friends! I pulled out parts that Michael, AA1TJ had sent me. I pulled out others that had been sent by Jim, AL7RV. I got out my box of isolation pads that Jerry Felts, NR5A had sent in. Soon the parts were glued and soldered to a board that already had an AF amplifier designed by Roger Hayward, KA7EXM and a PA that is my "Manhattan-ized" version of Tony Park's SDR rig PA. At the center of the board is little Colpitts oscillator that I took from a WSPR rig designed by Gene, W3PM -- earlier in the week Gene had posted a comment on this blog saying he was pleased to see my call on his WSPR screen. I wonder if Gene realized that he was seeing the signal from an oscillator from his design! The laptop was provided by a listener who prefers to remain anonymous -- thanks OM! Nearby, a copy of "Lid, Kid, Space Cadet" by Jeff K1NSS provided encouragement.

The rig passed the smoke test so I moved it over to the operating position and put it on the air. I got immediate gratification: the WSPRnet map right away showed my signal (now around 200 milliwatts) being received all round North America. You can watch this LIVE by going to http://wsprnet.org/drupal/wsprnet/map Just plug N2CQR into the "Call" box and hit UPDATE.

I know that WSPR is not everyone's cup of tea, but I like it. It was fun to build this rig. While WSPR is almost exclusively a mode that uses store-bought equipment, I get a kick out of being one of very few ops using a homebrew transceiver in this mode. (Are there any others?)

Next steps: I need to figure out how to set up automatic switching (by the computer) from transmit to receive. And I want to make some PSK-31 contacts with this rig.

Thanks to all who contributed!

73 Bill

Need audio reports for Podcast 128

How did 128 sound? I did drop off the lows, but I used the same filter I used in 127.

10% off on SolderSmoke The Book

Now through December 31. Use coupon code PEARTREE when checking out.

Saturn and Rhea

After a long spell of looking almost exclusively at Jupiter and our Moon, I was very pleased to see that Saturn is now visible. It is up in the south-east just before dawn, to the upper right of VERY bright Venus. The Sky and Telescope screen shot is a good representation of what I saw. I was using a 6 inch reflector telescope. I could see Titan very easily, and I could also see Rhea by using the averted vision trick.
Above is a much better view of Rhea. This one was taken by the Cassini spacecraft last November. Rhea has been in the news recently because scientists have discovered oxygen in its atmosphere:
http://www.astronomy.com/en/News-Observing/News/2010/11/Cassini%20reveals%20oxygen%20atmosphere%20of%20Saturn%20moon%20Rhea.aspx