help with pic10f200, overheats when power is applied

[thepenguin77]

I am new to programming pics. My first task is to light a LED. The problem that I am having though is that whenever I attach power to the pic, it overheats within 5 seconds. Overheat as in ~150º F in 5 seconds.

I am using three volts. The negative side of the batteries is to vss and the positive side is to vdd.

Also when I have a LED attached with the proper code, the LED does not light.

Any ideas?

 

[betwixt]

Show a schematic and software please.

first guess though would be the series resistor in line with the LED - You do have one I hope! Value should be:

(VDD - (forward voltage of LED)) divided by LED current.

For example, if you use a red LED with 1.7V forward voltage and 5mA current, the resistor should be (3 - 1.7) / 0.005 which makes it 260Ω. Use 270Ω which is the nearest standard value.

Brian.

 

[IanP]

For a while forget about the LED and make sure that the PIC is really connected correctly to the battery, maybe you used it’s mirror image, or wrong pin configuration ..
If, however, it is connected correctly and still heats up fast – it must be damaged and you will have to carefully replace it with another unit ..
After this situation is sorted out you can think about re-connecting an LED (with reasonable resistor in series) ..

Rgds,
IanP

 

[thepenguin77]

Schematic:

----------------------------
|                           |
-(+3v-)-|vdd vss|-----------|
        |    gp0|-(100Ω)->|-|

Code:

list	P=10F200
nolist
	#include	<p10f200.inc>
list
	__config	_mclre_off & _cp_off & _wdt_off & _intRc_osc

org	0x0000
	movwf	osccal

	movlw	b'11111110'
	tris	gpio
	movlw	b'00000001'
	movwf	gpio
stop:
	goto	stop

end

I understand that the resistor is wrong. I do not have many parts and I just used this as a test. I also tried a 500Ω with the same result.

The problem I have though is that if I remove the whole gp0 part and just leave the PIC connected to the battery, it still overheats.

I am pretty sure that I am using the right pins as I had to rewire my programmer in order to program the chip.

 

[wp100]

Hi,

Well it sounds like your Pic has been totaly shorted.
Don't think using just 100R to a standard small led would have caused it either.

Looking down on the top side of the chip Pin 2 should be 0v and Pin 5 should be +3v

Would think at some stage you must have messed up the connections and blown the chip.

I would not use that chip again as it might cause problems for you programmer; is it a Pickit2, and why did you have to 'rewire' your programmer ?

Would suggest you use a test meter to check yor power supply, there is a fair chance its 3v regulator may be damaged - in situations like that I always replace it anyway, just in case.

 

[thepenguin77]

Hi,

Well it sounds like your Pic has been totaly shorted.
Don't think using just 100R to a standard small led would have caused it either.

Looking down on the top side of the chip Pin 2 should be 0v and Pin 5 should be +3v

Would think at some stage you must have messed up the connections and blown the chip.

I would not use that chip again as it might cause problems for you programmer; is it a Pickit2, and why did you have to 'rewire' your programmer ?

Would suggest you use a test meter to check yor power supply, there is a fair chance its 3v regulator may be damaged - in situations like that I always replace it anyway, just in case.

I use the 8pin version of the PIC10F200. From above, according to the datasheet, I have +3v on pin 2 and 0v on pin 7. I use a JDM programmer and I had to rewire it because the programmer connections did not match up with what the chips wanted. Basically i just use a bread board with it. I do not think that the chips are totally shorted because they can still be reprogrammed. When I take a brand new one, program it, and then hook it up, it overheats but retains its memory and programming functionality. The power supply is two AAs.

If it helps, what I am actually doing is trying to make xbox controller modchips. I noticed the problem because I used a computer simulator to test my mod program, which worked. But when I hooked the real life one up, it overheated and sent crazy signals.

 

[betwixt]

VDD should be 3.3V on pin2 and VSS should be 0V on pin 7 in the DIP package. I think wp100 was referring to the SMD package.

Lets try to eliminate the hardware, does it get hot in the programmer? Try repeatedly reading it so the power stays on. I'm not familiar with the working of the JDM programmer so I'm not sure if the power is turned off when not talking to the PIC.

The only other thing I can think of that might cause problems is if you have anything else connected to the PIC which runs off 5V. Putting 5V on it when powered from 3.3V will cause a latch-up condition and that will certainly cook it. Are you disconnecting the JDM when testing it with the LED?

Brian.

 

[wp100]

Hi,

Oopps - yes you are right for the power connections, I was looking at the SOT23
pinout not the PDIP.

From your description it really sounds as if something is shorted.

Have you actualy measured the current being taken by that chip with a MA meter, without anything else attached it should barely register 1ma ?

If as you say you can take a 'shorting' chip and read back its program and then reprogram it then that tends to imply that + - Mclre, PDat and PClk connections are good, but you do not say if it overheats when connected to the programmer ?

Assuming it does not, then that implies that you must have a shorting connection somewhere in you final circuit. - what does the current measure when it is in that circuit and getting hot ?

 

[thepenguin77]

The chips do not get hot in the programmer. I could not measure the current because my multimeter has a blown fuse, but it is apparently over 1ma.

I did some more tests. I took a brand new chip out of the tube, programmed it, and then hooked it up to 3v. I did not even include a LED. It was just battery to chip to battery. It over heated.

I did the same thing with a chip that I hadn't even programmed, same result.

 

[wp100]

Hi,

Well afraid I'm as lost as you as to the answer !

Have to ask the blindingly obvious, you are identifying the chip pins correctly ?
looking down at the black top of the chip, with the pins facing away from you, the little round indentation, by the notch in the end , is Pin1.

If you take a brand new chip and just apply power to that, without it touching the programmer, does that still short ?
EDIT - Opps - missed your last line...

 

[thepenguin77]

Yes, I am using the pins correctly lol. I can verify that myself because I had to identify the pins to program it.

 

[betwixt]

Hmmm....

I use lots of 10F200 series PICs in DIP and 6-Pin SMD packages and I've never had one get even slightly warm so something has to be wrong with your hardware.

Can you remove the IC, turn the power off and do a resistance check for shorts between any of the pin connections. Although it shouldn't have such a dramatic effect as you are seeing, try running it with the unused pins tied to ground through 100KΩ resistors so they can't 'float'.

Something is obviously sinking current into the PIC or drawing current from it.

Please confirm you are still running it from two AA cells and not a regulated power supply and that you have a decoupling capacitor between pins 2 and 7.

Brian.

 

[wp100]

Hi,

Sounds as though you have programmed other chips in your jdm, so must assume that works ok.

As it only takes a second to program that little chip I expect it would never have power applied to it for long enough to get hot.

When you say you connect the power to the chip - how are you doing this - in a breadboard, a wired ic socket ? - is it the same thing everytime you test one and get the apparent short ?

Think you need to get you ma meter working and check the actual current, with nothing connected it should be under 1ma according to the data sheet.

I would also be tempted to put a diode, almost any small type 1n914 - 1n4006, in your power line just before the Vdd pin and see what happens then - do you see approx 2.5v across Vdd and Vss - does the chip get hot ?

 

[thepenguin77]

There do not appear to be any shorts in the chip from checking the pins. Tying the unused pins down did not help. I am still using two AAs.

I am using a breadboard and it is always the same setup. Actually I don't even have a bread board. I am just using a piece of wood with holes in it and wires in the holes, but that is not a problem because I always check connections to the wires before I do testing.

Adding the diode made it work, but when I went and put a flashing program on it stopped working. Going back to the always-on doesn't work either. The diode does take away some heat though. I don't know what kind it is, but it is clear with a red core, and it says 4145.

There is 2.5v across the chip.

I got the ma meter working. With no diode the chip draws ~11 mA. With the diode about 4.5 mA. I did this by wiring the meter in series with the chip, correct?

 

[betwixt]

??? :?

11mA isn't enough to warm the chip up. It only amounts to 36mW.

Brian.

 

[thepenguin77]

Maybe I have my decimals mixed up. In line, right before the PIC: 11. A new AA battery, 62. So that's what I get, but I'm not sure of units.

 

[thepenguin77]

Sorry for double posting but this still isn't working.

So I figured out how the multimeter works.

The chip is running at 1.4 amps. With one diode it does 600 mA.

Edit:

Progress. I added a 3k resistor between vss and ground. The current through the chip was then 1mA. The LED turned on, but would not flash when I used a flashing program.

 

[betwixt]

At 1.4A I would forget the LED, the PIC will glow bright enough to see !

Something is wrong with your construction somewhere. Adding the resistor would have INCREASED the current even more yet it seems to have dramatically reduced it.

Can you post a photograph of your construction, it is difficult to visualize your problem which seems to defy the laws of physics!

Brian.

 

[thepenguin77]

I'm sure this looks ghetto compared to you guys' bread boards. But as I said, I never planned to make this.

 

[betwixt]

Don't worry about being tidy - I could show photographs that make your effort look like an artistic masterpiece!

Two things look wrong: I can't see a connection from the + end of the batteries to the PIC, you appear to have as resistor in line with the supply.

and

You should have a capacitor wired across the VSS and VDD pins as close to the PIC as possible. 100nF would do nicely.

Brian.