This version, bought from three vendors over the past 3-4 years, all get super hot, even when being powered via a USB programmer. It's the resistor between the voltage regulator and the programming pins. I have a small number of older units from 5-7 years ago that DO NOT GET HOT. The hot version is the only one I can find on the Web these days--do not buy it.
Are you sure you don't have a 5V programming adapter on a 3.3V Mini?
2 Likes
That's always a great question. I have screwed that up more than once. But in this case I'm fairly certain I'm talking about 5v devices. I've still got a few 5v devices in their anti-static bags--I'll put the pins on them and see what happens.
The issue is that the 3.3v devices are supposed to be able to handle 12v at Vin, so even if they are 3.3v devices, they ought to operate nominally if they are powered via Vin.
Bigger question: Are stupid vendors selling 3.3 devices labeled as 5v? Are there any 5V Pro Mini devices being sold today?
Oh yes, by SparkFun, 11 bucks but worth it to me.
https://www.sparkfun.com/products/11113
1 Like
Look into rolling your own with a DIP chip and however much board you need.
Minimal Arduino plugs into breadboards, has a use, failed obaka!
1 Like
That looks like a Schottky diode. It may get warm depending on what the arduino is driving.
** edit **
There isn't a Schottky diode on the original Sparkfun Pro Mini schematic. The Vendor may have added one to a clone design maybe for reverse polarity protection.
Thanks. Hot is an understatement. The component is so hot it smokes and unsolders itself.
I can see a resistor on either side of the regulator, but I couldn't imagine what that was or why it is hot. I've looked at my older Pro Minis and they don't have anything like it. Whatever the change is, these clones are not good. I'm going to test them ONE MORE TIME to make sure I'm not confusing a 3.3v device with a 5v device.
JCA34F, I should have been more specific. I was looking for 5v devices with the offset A4/A5 pins, with or without A6/A7 pins, and without the "hot" component above the regulator. I'll take a look at Sparkfun. Oops, the A6/A7 pins are in the "wrong" place for the PCB I've got installed. Thanks.
I am quite sure that is not a resistor, but a diode. Shown on schematics:
Why it gets hot though is the question, although the answer is usually "cheap substitution" (if not design error). If you don't need reverse polarity protection, just replace with a wire link.
ETA: actually maybe don't do that, the diode prevents back feeding of UVCC from the RAW input. It's ok if UVCC is unconnected I guess.
It seems that often there is an initial clone of an reference board, and then over time there is a race to the bottom and eventually you get highly substandard products or outright fakes.
Your board doesn't seem to follow the Sparkfun pin arrangement for the FTDI connector. Is it possible that you have VCC/GND reversed?
Interestingly, the FTDI header on that board does not even have the ground pin populated.
I'd certainly be wondering which pin the OP used instead.
It looks like there are two GND pins.
but a standard FTDI adaptor wouldn't plug right in.
Edit:
Or it would actually since CTS is GND as well.
Oh yes. That is indeed another variant. If the OP used a standard FTDI connector, there would be a 1 pin overhang right or left.
The device you point to is a Schottky diode. In normal operation that part should not even be warm.
I received a batch of 3 yesterday. Opened up one, applied 12 between raw and gnd. No issues. The diode was dead cold.
In my experience, "hot enough to melt the solder" would required a dead short to ground on the cathode of the diode. That would be the left side in your photo.
Could you better describe the connections when the diode is getting hot? Where is power coming in and what else is connected to the board?
My board:
UPDATE from my earlier post,
I opened up another board, Applied 10 volts to the RAW terminal ( only Raw and gnd connected no other pins were connected.
The board worked for a minute or so, the LED's when off and the diode got very hot.
The Datecode on the processor was "2043" ake the 43 week of 2020
Are you sure your power supply is DC, not AC?
Yes. I'm a retired EE.
And even if it was, the board would "kinda work" unless the AC peak surpassed the regulator input capability.
------------UPDATE ---------------------------------------------------------
On my board the voltage regulator is shorted out. My regulator is marked LR8103 which appears to be a "Lori Semiconductor" device which has a max input of 6.5 V (see red square).
In addition the series diode is a 1n4148 where normally it is a 1 Amp Schottky diode. Magenta oval.
The tantalum capacitors are different than the "normal" capacitors on these boards. I haven't been able to ID them.
In any case, I will not be using any of these boards.
1 Like
The marking S4 on that diode does tend to indicate a 1n5819 type Schottky diode. Has it got a higher than normal forward voltage ?
Certainly, a regulator shorted out due to overvoltage, or otherwise, is the most plausible cause of the OP's problems.
Your right (kinda). It seems S4 covers a number of devices so perhaps this is not part of the issue with this board.
My boards were getting hot when they were connected only to the USB-serial programming device. Not as hot as when powered on the PCB by 9v or 12v, but hot to the touch.