It can be a real drag to fix a circuit board which has stopped working as intended, especially if you don’t have any reference material for the product. That’s the position that [Todd Harrison] found himself in when the controller for his mini-lathe gave up the ghost. He undertook and hefty repair process and eventually mapped out and repaired the driver board.
First off, we’re happy to report his success at the end of a year-long troubleshooting process; the entirety of which occupies six different posts. The link at the top is the conclusion, and you’ll find his final test video after the break. But as you can see from the image above, he was met with a lot of problems along the way. The first two segments show him reverse engineering the PCB, with a giant schematic coming out of the process. In part 3 he then started probing the board while it was live, with the smell of hot electronics causing him to disconnect the power every thirty seconds. One time he took too long and blew a resistor with the pictured results.
In the end it was a shorted PWM chip to blame. He tested a couple of different replacement options, dropped in the new part, and is now back in business.[youtube=http://www.youtube.com/watch?v=Z8h-0XiXVbc&w=470]
This is pretty amazing troubleshooting and one can’t fault Todd’s tenacity. But why didn’t he just buy a replacement board instead of spend a year fixing this one, unless he really, really enjoyed the learning process and thrill of fixing it himself?
Damn that’s some fast soldering at 7:35.
It looks sped up 😛
I used to repair DC motor controllers back in the day, often without prints. It was not unusual for me to test all the resistors and transistors and replace the Caps and ICs(I was paid by the hour ICs are cheap).
From my experiences if you blow the chips on the board but not the power components you need to check your speed pot and wiring to look for shorts. A speed pot terminal touching the machine case will blow the chips due to the fact there is no isolation between logic power and line. In fact since this is 240vac logic gnd and chassis gnd are at different potential.
A year to fix it? He could have joined the tech lathe Yahoo group and had the problem identified in minutes, with several offers to fix it for very low cost.
Good job, Todd. We have all been down that road before. Most of us give up, but not Tenacious T!
how can tel me what that band on his arm is for.
So basically, he just kept replacing parts, until it kinda worked, then just sent it back to the owner, and they finished the job. Awesome documentation, of extreme lack in skill. I would be shocked, if they ever ask him to fix anything again.