EEVblog Electronics Community Forum

Electronics => Repair => Topic started by: MathWizard on December 23, 2020, 09:00:07 am

Title: Thought I found a row of burnt caps, or was it dust ?
Post by: MathWizard on December 23, 2020, 09:00:07 am
I had a modern digital thermostat given to me, from a house that is said to have had the DTS appearing to lose power, so the screen would cut out and so was the 3kW baseboard heater. I'll find outmore on that later.

I opened the DTS, and right away saw what looked like a thumb-sized black ash/charred/blast area, centered over the corner of the ATMEGA169, and a row of little SMD caps. And on the power PCB, a few mm opposite that, there's a near thumb sized, coal black, blast mark, that was centered on a SMD 0.8A Thyristor/SCR (

It looked like a blast or big arc went from it, then under 2 big power resistors. on it's PCB, and covered the caps/MCU right arcoss from it, and the tops of stuff on the SRC PCB, looks clean, like 2 white mini fuses 1.5cm away, were still clean on top.

 But the cap's on the other side, looked complemely toasted, until I cleaned them. And everything seems fine. IDK about the small SCR for sure, but it's not shorted, it's big pin side, sort of looks like it's solder was melted weird, like on Mosfets I've seen fried.

Or maybe it's just soot? Electrostatically sticking ? The tenents used to burn candle's/incsene's??. I've seen cigarette tar get on these same exact same model thermostats before. But this stuff, litterally was like coal dust, and higly localized like a blast/high speed cooking, I should have taken a snap.


So I've been reverse engineering the power PCB, for the upto 3kW, 240V heater. And I sort of get it, but it's not really working in LTspice. Too late to go trying it on the variac, but I think I have all the power PCB.

I measured the 100Hz inductance of the little transformer, and all but 1 of the zener's, but I still can't get the 100uF cap to charge at all, it's for the LV Vcc on the other PCB.

I don't get how the SCR is triggering the Triac correctly , and the triac has 0.7amps coming from it's gate, so is that even possible ? The BJT network is over my head to calulate at AC, but I have that doing something useful looking for the gate triggering, and/or sensing for the MCU to control the SCR... I just assume have a better working model before I try powering it on the variac.

Title: Re: Thought I found a row of burnt caps, or was it dust ?
Post by: MathWizard on December 24, 2020, 10:54:33 am
I figured out 2 major errors in my schematic, I had the load in the wrong place, and some caps in the wrong place of the transformer based PSU. But also, the models for the triac and SCR were no good for my purpose. So I was lucky to find the datasheet's and the companies actual SPICE subcircuits to make models. And tada, it works just fine.

And so it really is just a resistor/BJT based LV PSU, When the MCU GNDs the base of a single BJT, there's no current for the SCR gate. But when the MCU PCB provides 3V to that BJT base, enough current can flow into the SCR gate to activate it and short the FWR. And so then the near full AC voltage is presented across the load, and the 24 ohm resistor between the Gate and cathode of the Thristor, and so it can get enough current to turn on/off at 120Hz.

And then with a lot of current flowing through the little transformer, it transfers the energy to the zener/diode bridge, and then powers the LV supply for the MCU PCB.

Title: Re: Thought I found a row of burnt caps, or was it dust ?
Post by: MathWizard on December 27, 2020, 11:02:56 pm
Indeed it was just black dust, electrostatically built up. I looked at another thermostat, and it had that household's dust built up in the excat same opposite spots on both PCB's

But at a glance, with black soot dust, I was hoping I something to work on,not just clean up