Junebug Problem

[3v0]

I built several boards that worked with the same part number from the same supplier. The supplier switched parts/pinouts but kept the same number and datasheet. The computer printed label on the packages were the same but the parts inside were not. The only way around that would be to check every bag of parts.

The EE's were very good at what they did. But that was back when I worked for a living. We all worked for the same company.

 

[blueroomelectronics]

I hope you have a good solder sucker.

 

[3v0]

My method to remove transistors.

The only solder suckers I own are the spring powered jobs. I did not use them for this repair.

My method to remove transistors.

I clamped a hemostat to the transitor to be removed. With the board upside down I added enough liquid solder to bridge all 3 transistor pins. If the transistor did not drop from the weight of the hemostat, I rocked the hemostat gently. Do not pull or you will remove/ruin the hole plating material. If you remove the transistor before all the solder is melted on both sides of the PCB you will lift a top trace or pad. It should not take so long that you cook the PCB.

With the transistor removed I heated the solder in each hole from the top and knocked the bottom side of the board on the edge of the workbench. The board stops and the liquid solder keeps going. Some times it helps to add solder first. Each hole may take one or two tries. Use your head here. Use just enough force to get the job done.

I use a similar method to solder .05 inch pitch connectors. solder the two ends down and cover the rest with solder. Heat up a few pins and give it a wack If there are traces in the path of the flying solder cover them with a bit of blue tape.

How the transistors were installed makes a differance in how hard they will be to remove. If the leads do not come straight out of the holes the transistors will not easily come out. In this case cut the transistors loose from the board. The straighten any remaing top leads and remove.

The lesson here is that when you instal a component do it in such a way that it is easily removed.

If anyone want to use this method try it on a junk board first and use at your own risk. As far as I know it is not a recommended method.

 

[blueroomelectronics]

That's a method I've used for ages. I do like a solder sucker but to clean out the holes nothing beats velocity.

 

[Mickster]

...Some times it helps to add solder first....

How the transistors were installed makes a differance in how hard they will be to remove. If the leads do not come straight out of the holes the transistors will not easily come out. In this case cut the transistors loose from the board. Then straighten any remaing top leads and remove.

Great advice and considering the small price of a transistor, versus the potential of ruining a good, possibly expensive board, I would suggest using the hemostats clamped to a transistor leg on the underside of the PCB, and a solder-sucker at the ready topside. Removal & prep in one operation...

 

[mvs sarma]

Great advice and considering the small price of a transistor, versus the potential of ruining a good, possibly expensive board, I would suggest using the hemostats clamped to a transistor leg on the underside of the PCB, and a solder-sucker at the ready topside. Removal & prep in one operation...

It is is ok for a three legged device, but in case of higher pin count like ICs etc, we used to better follow the 3V0 method. Cut and remove the chip , then remove pinafter pin. afterall the board is many occasions costly and hard to get compared to a chip.

 

[Mickster]

Sarma,
the suggestion was made in reference to 3v0's method of firstly cutting out the main component, then removing the pins left behind. As you correctly state, this is much more beneficial with higher pin-count components, versus a costly PCB.

 

[Nigel Goodwin]

In the past I had to regularly replace 100 pin odd surface mount microprocessors in Sony CRT TV's. Couldn't do it with the Pace rework station as they are glued to the PCB by about 9 blobs of glue under the chip (overkill or what?).

My technique was to carefully cut all the pins off the chip with a sharp knife, then carefully remove all pins from the board with the soldering iron. This leaves the square legless chip sat glued to the board, and with the pins removed, no chance of damaging the PCB. Next I would get a wide bladed screwdriver and place it against one side of the chip - then hit the handle with a hammer - one or two blows would send the chip flying across the room, with the PCB undamaged.

Then simple scrape the glue reside off, flux the contacts of the PCB with liquid flux, and drop the chip in place - carefully tack it down at a couple of opposite corners, make sure it's lined up correctly, make sure (for the tenth time ) it's the right way round, then brush more liquid flux over the pins and draw the Pace 'spoon' bit full of solder along the pins.

Lastly clean the flux residue off, and test to make sure it all works.

Incidently, chip failure was usually caused by the LOPTX arcing internally - but almost always ONLY on 16:9 sets, the identical 4:3 chassis only rarely did this?.

 

[Mickster]

In the past I had to regularly replace 100 pin odd surface mount microprocessors in Sony CRT TV's...My technique was to carefully cut all the pins off the chip with a sharp knife, then carefully remove all pins from the board with the soldering iron... Next I would get a wide bladed screwdriver and place it against one side of the chip - then hit the handle with a hammer - one or two blows would send the chip flying across the room, with the PCB undamaged.

Surprised that you didn't omit the hammer, opting instead for a well-placed karate-chop!

Work, plus exercise at the same time...paid too.

 

[Nigel Goodwin]

Surprised that you didn't omit the hammer, opting instead for a well-placed karate-chop!

Well, at times, when I couldn't be bothered to walk and fetch the hammer, I have palm-heeled the screwdriver instead