Author Topic: Made some magic smoke and I don't know why :/  (Read 3939 times)

0 Members and 1 Guest are viewing this topic.

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Made some magic smoke and I don't know why :/
« on: April 11, 2018, 06:16:18 am »
I'm making something similar to - one 12v fan in a box.  It worked well from a 12v wall wart for about 2-5 minutes then stopped, though the lights (lighted fan) blinked about once per second.  There was a hot component, so I let it cool.  When I turned it back on it burned up.  As far as I can tell the component was labelled "b62871"- my best guess is it is one of these.  It is part of a voltage step-up convertor from here.  However, the banggood page says input voltage from 2-24v, and output from 5-28v.  On that component pdf, it says input voltage is 11v max.  And yes, I have 4 more of them to try out. 

Eventually I plan to run it from an 18650 battery (~3.7v) to my desired fan voltage (12-15v), but atm I'm running it from a 12v wall wart (12.3v measured), to 16v out (to allow for any dropout voltage issues), which should be well under the 2amp rated for the boost converter.  I thought the heat and other issues would be minimized doing this.

I did accidentally hook it up backwards at first. :o  The wall wart may or may not be regulated, though I have the components (lm317 and so on) to regulate it before the boost convertor if you guys think that may help.  Could it be that the product page is wrong and the input voltage is out of range?  Could it be a current issue?  Thermal issue?  I don't think a single fan would draw too much current, probably only 1 amp max, which is half the rated current.

edit: sorry, I screwed up the html so half my post was hidden :p
« Last Edit: April 11, 2018, 06:27:38 am by MyHeadHz »
 

Offline Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #1 on: April 11, 2018, 06:24:32 am »
You didn't say what part released smoke.
 

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #2 on: April 11, 2018, 06:34:34 am »
I fixed the post, sorry.

Well, there are lots of people who have fans running from them, but mine died.  There are the possible issues above that I mentioned, though there could be others that I'm not aware of.  The part number is my best guess based on what the part number looks like, which could very well be wrong.  Also, 2 amps is not required.  My fan should probably only use about 1 amp max.

If it is a current issue, would I be able to add a heatsink to that tiny component?   I have heatsinks, thermal tape, and thermal compound.  The question for me is which rabbit hole is worth diving down.

also... Boner pot?

The part that smoked was what appears to be an SOT23 component.
« Last Edit: April 11, 2018, 06:45:11 am by MyHeadHz »
 

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #3 on: April 11, 2018, 06:54:52 am »


That's the component that smoked.  It looks like B62871.

Here is an aliexpress link of the same thing, with better pictures.
« Last Edit: April 11, 2018, 07:23:59 am by MyHeadHz »
 

Offline nrxnrx

  • Supporter
  • ****
  • Posts: 68
  • Country: ro
Re: Made some magic smoke and I don't know why :/
« Reply #4 on: April 11, 2018, 07:27:48 am »
It looks like this guy: https://datasheet.lcsc.com/szlcsc/SDB628_C77805.pdf
Also known as this: https://www.olimex.com/Products/Breadboarding/BB-PWR-3608/resources/MT3608.pdf

..and according to this thread you need to be careful with the pot, since it's possible to feed more than the allowed 6V to the feedback pin by accident.
« Last Edit: April 11, 2018, 07:35:40 am by nrxnrx »
 
The following users thanked this post: blueskull

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12806
Re: Made some magic smoke and I don't know why :/
« Reply #5 on: April 11, 2018, 01:16:52 pm »
If you hooked your boost module up backwards, you probably killed it right there right then.

Also the ratings of modules from Banggood and other Chinese eBazaars are almost invariably over-optimistic.   That 2A rating means that a friend of the manufacturer's 3rd cousin heard a rumour (while drunk) that it could do 2A - in a deep freeze on a Tibetan glacier - ONCE!    Personally I wouldn't expect more than 50% of its rating, so 1A peak, and not continuous - expect thermal issues if you try to run it at 1A continuous.  Competitor claims 1.5A for a maybe 1.2A module, pretty soon everyone is claiming 1.5A so they don't loose business.  Then some intrepid soul claims 1.6A and the cycle repeats until they reach the highest number that everyone can say with a straight face.

For the voltage rating you have to track down the IC manufacturer's datasheet - the IC actually used, not whichever IC it was cloned from, and allow a reasonable margin,  Also check the board for capacitors with an inadequate rating.   

N.B. if a regulator or converter board has ceramic input capacitors, connecting it to a supply of more than half its actual voltage rating, with the supply already on (or only switched off at its output) can kill it due to the wiring inductance ringing with the input capacitors, and momentarily doubling the applied voltage seen at the board input.
 

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #6 on: April 11, 2018, 11:49:08 pm »
It looks like this guy: https://datasheet.lcsc.com/szlcsc/SDB628_C77805.pdf
Also known as this: https://www.olimex.com/Products/Breadboarding/BB-PWR-3608/resources/MT3608.pdf

..and according to this thread you need to be careful with the pot, since it's possible to feed more than the allowed 6V to the feedback pin by accident.

In that thread, near the bottom the poster mausi_mick suggested adding a third resister ( schematic ) to make it safe.  Google translate seems to suggest a 5kohm resistor from the "top" (center?, since it is "above" the other two?) of the pot to vout (guessing vout+?).  Would this be correct?



Or would a capacitor help by smoothing out noise/peaks?  Would that resistor be a good suggestion to stop the burnouts?

Also, I measured the amperage.  At 12.3v the fan uses about 165mA (true RMS) once it settles, which is should be fine regardless of what the manufacturer claimed.

I also tried it out with a new boost converter without ever shorting it, overheating it, or reversing its polarity.  It died almost instantly.  It's OK, I still have 3 more.

I think this circuit is based off that second one you posted.  How did you find it?  The closest thing I found was what I linked in my first post, even though it wasn't very close.
« Last Edit: April 12, 2018, 12:02:20 am by MyHeadHz »
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12806
Re: Made some magic smoke and I don't know why :/
« Reply #7 on: April 12, 2018, 12:15:19 am »
A capacitor on the feedback pin would kill it by making the feedback loop unstable.

You could make it more idiot proof by adding a small 5.1V Zener to clamp the feedback pin + a small resistor in series with the top end of the pot to prevent the Zener blowing if the wiper is all the way to the top, + a resistor in series with the bottom end of the pot to limit the maximum voltage it can boost to, to something the IC can survive, but that's gilding a turd, and is only really required if you were going to fit a front panel pot, at which point you'd be better off with a boost-buck module using a smarter IC with better overload protection.

Its really quite simple to avoid electing a pope - initially set the pot to mid travel and  don't power it up with more than 5V in till you've rough-set the desired output voltage into a minimum resistive load.  Then you can connect your actual power source and load and fine trim the output voltage and lock the pot with a dab of nail varnish.  The minor inconvenience isn't worth the hassles of modding it.
 

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #8 on: April 12, 2018, 08:57:31 am »
OK, so I used an 18650 battery instead of a wall wart this time.  I added a 50 ohm resistor (had one handy) in series with my multimeter and adjusted the voltage accordingly.  By default it was approximately half-way through the range, about 50k ohms.  It seemed to work fine at first, but failed after about 20 seconds.  The theory about the fault at the extreme may be true, but it wasn't anywhere near either extreme-- so it was not the issue causing my problems.  It took about 2-3 full rotations to move the pot about 1 volt, for what it's worth.  Also, I tested the other pots from the dead circuits and they were all in the 40k-60k range (100k pots).

This is very frustrating.  I have two left that aren't fried.  :/  There must be something I'm doing wrong consistently, but I have no idea what.   :-//
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12806
Re: Made some magic smoke and I don't know why :/
« Reply #9 on: April 12, 2018, 09:55:22 am »
I suspect you are overloading it.  The IC's peak 4A  switch current rating results in 2A max, averaged over the On time of the internal switch, and its 90% duty cycle limit means it cant handle an input power at 3.7V of over  6.66W.    At low input voltages, the efficiency is down around 85%, so the maximum output power will be only 5.66W.  At 12V that's only 0.47A, and that's an optimistic, no safety margin Chinese rating.   Your 50R resistor which I presume was across the output, will have been drawing 240mA, 51% of the nominal rating, which is a bit optimistic  for a chip made of pure Chinesium.   At 3V in the absolute maximum 'Chinese rated' current it could deliver at 12V out would be 360mA, so IMHO you'd really be pushing your luck over about 150mA.

The fact it died after 20 seconds implies thermal issues.   Try loading the next one with 120 ohms and see if it survives outputting 100mA for a sustained test

Its obvious they are a reeking pile of s--t and haven't got a snow-flake's chance in hell of supplying 0.5A to the fan, especially as the battery gets low so the only question is how  deep and how ripe the s--t is.


 

Offline MyHeadHzTopic starter

  • Regular Contributor
  • *
  • Posts: 165
  • Country: us
Re: Made some magic smoke and I don't know why :/
« Reply #10 on: April 16, 2018, 04:40:26 am »
OK, after reviewing all of the above, it seems that I'm asking too much for the component.  Lowering the voltage wouldn't really serve the purpose for the project, so I checked out other strategies.

The obvious one was find a higher amp circuit.  There are mainly two IC's that power those devices (like these, powered by LM2587 clones, and these, powered by XL6009 clones.  Both usually claim min input voltage of 3v, but data sheets say otherwise.  Issues with overvoltage (output) are all over the internet because people are trying to run them at their chinesium specs.

So, I decided to change strategies.  Simply getting a second 18650 and running them in series will make life much easier for me. :o

Thanks for the help everyone- I learned a lot!
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Made some magic smoke and I don't know why :/
« Reply #11 on: April 16, 2018, 07:06:37 am »
FWIW, there are converters that beefy and small.  TI makes a 3A (buck? I forget) that small, and it really does deliver what it says. :)

Maxim should be okay at ratings, but... who knows what clone they stuffed in there?

Anyway, yay for shitty designed, unsafe circuits that can't handle adjustments without releasing magic smoke?

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Chris56000

  • Frequent Contributor
  • **
  • Posts: 896
  • Country: gb
Re: Made some magic smoke and I don't know why :/
« Reply #12 on: May 03, 2018, 12:51:17 pm »
Hi!

PTN78020ANZ/PTN78060ANZ?

This little module from Farnell will operate over 3–36V, uses one resistor to set the output voltage, and is about 25 X 40 mm size, and will give an output variable between 11.2 and 22.5 V!

They're listed about £18 but if you make your solder extractor with one of these it'll probably run forever!!!

There are a number of MLCC caps needed for stability on the I/P but it's all explained in it's datasheet!

I've just repaired a DC–DC converter module that's totally unobtainium
by digging out its remains and building a PTN78020ANZ in the original case!!!

Chris Williams
« Last Edit: May 03, 2018, 12:54:31 pm by Chris56000 »
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 
The following users thanked this post: MyHeadHz

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 6202
  • Country: de
Re: Made some magic smoke and I don't know why :/
« Reply #13 on: May 04, 2018, 07:08:14 pm »
Pardon the slightly off-topic question, but: Why does a solder fume fan have to be battery-powered?

On my bench, I have plenty of power outlets available. (I'll need at least one for my soldering iron!) If this is about soldering "on the go", with a battery or gas-powered iron, I would not bother with a fume extractor at all. Probably no room to set one up anyway in that scenario, and I don't want to carry it...

Personally, I would very much prefer not having to bother with recharging a battery, and would just power the thing directly from a wall wart. Preferably one with the correct voltage, so I don't have to bother with voltage converters either.  ;)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf