Help regulator burning up, which capacitor is best?

Hello, i have a circuit with two voltage regulators
MC33269D-5.0
MC33269D-3.3

The caps i had first were 100nf for Voltage in on both and 0.47µF on the outputs.
Everything looked fine but i kept getting random resets on pretty much any sketch i ran the code below which tells you how much power the chip is getting and saw that the power fluctuated quite a bit, from 4400-6000 when i looked at my original breadboard that i made i used 10UF thru hole electrolytic caps so i soldered one on and the power issue seemed fixed the output was a steady 5106. i thought all was fixed but later noticed the regulator was getting so hot i could feel through the enclosure, what is the best capacitor to use for this that wont overheat the regulator?

void setup() {
  Serial.begin(9600);
  Serial.write(" PWR ");
  delay(5000);
}

void loop() {
  Serial.println(readVcc(), DEC);
}

long readVcc() {
  long result; 
  // Read 1.1V reference against AVcc 
  ADMUX = _BV(REFS0) | _BV(MUX3) | _BV(MUX2) | _BV(MUX1); 
  
  delay(2); // Wait for Vref to settle 
  
  ADCSRA |= _BV(ADSC); // Convert while (bit_is_set(ADCSRA,ADSC)); 
  
  result = ADCL; 
  result |= ADCH<<8; 
  result = 1126400L / result; // Back-calculate AVcc in mV 
  
  return result; 
}

i should also note i am planning on powering with a 12 volt, from a vehicle but on the bench i have a 12v and 9v wall wart that ive been testing with.

What devices are connected to it? How much current do they draw? Linear regulators dump the difference between the input and output voltage as heat, while the current remains constant - so depending on your load, it may have to be getting hot.

Did you read the datasheet section on caps? Regulators always give some guidance on the caps. Your caps may not be appropriate (you haven't specified what the ESR of your electrolytic cap is - also your ceramic output cap may cause instability):

The MC33269 requires an external output capacitor for
stability. The capacitor should be at least 10 uF with an
equivalent series resistance (ESR) of less than 10  but
greater than 0.2  over the anticipated operating
temperature range. With economical electrolytic capacitors,
cold temperature operation can pose a problem. As
temperature decreases, the capacitance also decreases and
the ESR increases, which could cause the circuit to oscillate.
Also capacitance and ESR of a solid tantalum capacitor is
more stable over temperature. The use of a low ESR ceramic
capacitor placed within close proximity to the output of the
device could cause instability.

nmaison:
i should also note i am planning on powering with a 12 volt, from a vehicle but on the bench i have a 12v and 9v wall wart that ive been testing with.

You also need to realize that when a vehicle is running the voltage is 13.5V - 14.5V. Which means the regulator will get hotter. You really need to get a switching regulator that will produce almost no heat and are very efficient. You can get cheap ones off Ebay or if you want to assemble your own you can. But these are more complex in the design layout.

Actually, my new truck will occasionally run at 15.2 for a period (after a bunch of research and talking to the dealer, it appears that is some sort of programmed thing to help "equalize" the cells (dual batteries) ... at least that is the theory). So you may even be over 15 volts in the vehicle electrical system (and vehicle electrical systems are noted for spikes from things like starters etc) so it is a good idea to make sure you have some protection before the regulator and that you are not using a regulator that is spec'd for a max input of 15 volts or so - this is in addition to the heat issue from the linear regulator. For a linear regulator, if you have 15 volts in and 5 volts out at 1A, you are dropping 10 watts in the regulator (10 v * 1 A). That is a fair amount of heat.

Thanks for replying Drazzy, The devices connected are a bc118 ble module and 23lc1024 sram chip, i have the ble module level shifted via the bss138 and resistors.

sorry i dont know too much about capacitors, i breadboarded this all out and a board designer gave me a list of what was supposed to be correct components to use.

here is a link to the original ceramic

i am not sure on the specs of the electrolytic since i bought it at a local electronics shop and their packaged in homemade looking wrappers, it does say vht10m16 10MFD 16v 105c

the board i have has 0805 smd pads so i was thinking this capacitor? but im not sure if its better or not.

in the datasheet i dont see ESR mentioned.

DrAzzy:
What devices are connected to it? How much current do they draw? Linear regulators dump the difference between the input and output voltage as heat, while the current remains constant - so depending on your load, it may have to be getting hot.

Did you read the datasheet section on caps? Regulators always give some guidance on the caps. Your caps may not be appropriate (you haven't specified what the ESR of your electrolytic cap is - also your ceramic output cap may cause instability):

The MC33269 requires an external output capacitor for

stability. The capacitor should be at least 10 uF with an
equivalent series resistance (ESR) of less than 10  but
greater than 0.2  over the anticipated operating
temperature range. With economical electrolytic capacitors,
cold temperature operation can pose a problem. As
temperature decreases, the capacitance also decreases and
the ESR increases, which could cause the circuit to oscillate.
Also capacitance and ESR of a solid tantalum capacitor is
more stable over temperature. The use of a low ESR ceramic
capacitor placed within close proximity to the output of the
device could cause instability.

I am powering though the auxillary port which is still 12v in my truck (arent there regulators in cars for aux power?)

gpsmikey:
Actually, my new truck will occasionally run at 15.2 for a period (after a bunch of research and talking to the dealer, it appears that is some sort of programmed thing to help "equalize" the cells (dual batteries) ... at least that is the theory). So you may even be over 15 volts in the vehicle electrical system (and vehicle electrical systems are noted for spikes from things like starters etc) so it is a good idea to make sure you have some protection before the regulator and that you are not using a regulator that is spec'd for a max input of 15 volts or so - this is in addition to the heat issue from the linear regulator. For a linear regulator, if you have 15 volts in and 5 volts out at 1A, you are dropping 10 watts in the regulator (10 v * 1 A). That is a fair amount of heat.

Not in any of the ones I have seen - the "12 volt outlet" is typically wired direct (or through the ignition switch depending on the vehicle) to the "12 volt" bus so you will see whatever the batteries are at, which, typically is 13.7-14.3v with the engine running (although as I indicated, I have seen higher).

I would get one of these and power the board with 5V.
http://www.amazon.com/Charger-Eleckey-Portable-Adapter-Samsung/dp/B00P7VSRFC/ref=sr_1_1?ie=UTF8&qid=1457113239&sr=8-1-spons&keywords=car+usb+adapter&psc=1
Many smaller versions available if you don't need 4.8A

if choosing a tantalum cap 0805(thats the pads i have to work with) ive found this one that has 5ohms ESR. Will that help?

http://www.newark.com/avx/tacr106m016xta/tantalum-capacitor-10uf-16v-5/dp/54J2823?MER=PPSO_N_P_EverywhereElse_None

What the hell... How did I not realize that?

What crossroads suggested is obviously the right way to power it. Get the 5v from something like that, and use the 5v to power the 3.3v regulator, and bypass the 5v regulator entirely. My bad on not suggesting that.

The original ceramic capacitor is not appropriate for use with that regulator (because it's a crappy regulator - in my designs, I reject any regulator that says it isn't stable with ceramic caps. I like ZLDO1117 series - though they say you need 47uF on the output if using electrolytic caps - only 4.7 w/tant or ceramic). A tant with 5 ohm ESR and 10uf capacitance is suitable as a replacement. Alternately, the ZLDO is pin compatible with the one you're using - so you could switch to a ZLDO1117 regulator, and get a 4.7 or 10uF -0805 ceramic cap (any one that meets the voltage spec and X5R or better is fine) - probably slightly cheaper.

Your designer obviously didn't RTFDS for the regulator - and quite frankly, 0.47uF of supply filtering after the regulator normally just plain isn't enough, I almost never see output caps that small - most regulators want at least 4.7 on the output, and I usually go higher than the datasheet requires.

Thanks Drazzy! id like the device to have its own power regulated, you said the ZLDO1117 is pin compatible? i am trying to source one, they have an 8 pin?
the regulator version i have is the

if they do then i might switch to that and the ceramic 10 uf x5r http://www.newark.com/murata/grm21br61c106ke15l/ceramic-capacitor-10uf-16v-x5r/dp/35R0687

DrAzzy:
What the hell... How did I not realize that?
Alternately, the ZLDO is pin compatible with the one you're using - so you could switch to a ZLDO1117 regulator, and get a 4.7 or 10uF -0805 ceramic cap (any one that meets the voltage spec and X5R or better is fine) - probably slightly cheaper.

Oooooh, you have the SOIC-8 one. No, I assumed you were using the more common SOT-223 package (which also has better thermals). Odd choice.

If you're using a linear regulator to power it, you're back to the original issue of it potentially overheating. How much current does the stuff you're powering from that regulator require?

You may need a switching regulator for the 5v (as myself and crossroads have suggested), because for every watt you use at 5v, the regulator needs to dissipate 1.4 to 2 watts, which it does as heat... (ie, maybe it's working correctly, even with the current caps)