So I wanted to share this story cause I am actually happy with myself finding this out (it was a rather well hidden problem) and of course to help others
Everything started with me repairing a 12V power supply / UPS combo, nothing big of deal just replacing some shot capacitors leaking chiniseum.
After that I started testing it with various stuff to see if it works correctly and then having my STK600 laying around I plugged it in and got the red light!
Tested immediately with another power supply still the same thing! What was going on?
BTW the red LED fast flashing means normally that the power supply is overloaded (in the manual they don´t give any other definition or reason).
I plug it in my laptop and everything good! No red flashing light no nothing, everything hunky dory!
That of course means that nothing is shorted at the Vtarget side of things (I didn´t have anything connected but the board is supplying various chips on board with that rail).
So next though was that this thing probably had two separate current sensors and the one on the external power side was shot. But after a bit of investigation,
which I will not go fully through in this post, it used two switching regulators one powered from the USB and the other from the ext. jack, tied together to create an intermediate 6V rail.
This and only rail was being monitored from a LTC6101 current sense amplifier by one of the slave MCU. This rail in turn would power the rest of the switching regulators creating all the
other rails.
And I want to make a complain here! The STK600 has a current monitor and the data are not available to us! Granted they wouldn´t be very accurate as other stuff and not only the target
device are being powered but that could come with a warning and at least give us an idea of what our target is drawing! But I digress... back to the point!
Testing the current sense, given the values selected by the manufacturer and measuring it with my not so trusty multimeter I got a reading of about 80mA, which by all means in well in target.
So the current sense was working correctly.
Testing the rest of the power rails, also identified no problem (either testing with ext. power or USB power).
Next the logic conclusion was that it couldn´t really read a needed input. Was it the current sense, well... if it was that probably meant that the ADC was shot and that would have showed also
when powered from the USB which it didn´t!
So there must be something that tells it when it´s powered from the external jack or not! The STK600 can work either from an ext. power (to which it only powers the target with the programed
voltage), or with the USB, or with both, so there should be something that tells it when the ext. jack is connected so that is can provide the target with more power.
And there was it! A little BJT NPN transistor near the input side that had definitely seen better days. I went ahead and replaced it with a BC846T as shown in the picture and...
GREAT SUCCESS! Have I said with a Borat smile!
I replaced it using my soldering iron and not hot air due to the close proximity of the plastic header next to it. See the attached picture for details.
And so like that I was back to business, funny thing is that the power supply that I repaired didn´t do the damage. Having the STK600 for more than 10 years now, the problem
might as well have been hidden for years, but because I was powering it from my PC, I didn´t really get to experience it.
All in all, it was a happy ending (that dev. board was the first that I bough, years before all that arduino rubbish). Repairing my old friend gave me a smile,
and I hope that it might help someone, if they experience the same condition.
At the time I am writing this and have completed the repair, I haven´t found another post on the internet on this matter.
Best Regards, Lefteris
Greece
Home
Help
Search
Login
Register
