Pages: [1] 2 3   Go Down
Author Topic: A cautionary tale with cheap ebay Voltage Regulators.  (Read 2415 times)
0 Members and 1 Guest are viewing this topic.
Offline Offline
Edison Member
*
Karma: 29
Posts: 2455
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

ALL the cheapy voltage regulators on ebay using the switching regulator, out of about 10 ive bought they all share the same fault well not really a fault, but
a cautionary tale..


when low resistance is applied the the regulated voltage increases, so after running it for an hour, the temprature
slowly creeped up (at the IC end), with heat came a slow rising voltage.... take any switching regulator, get your multimeter out and then place a decent load (a couple of ohms) across
and watch the voltage rise....

So careful hooking a regulated 5v source directly to an arduino without any regulation, it could when over heating easily kill your board / device.

From now on, i'm going to use a 5v linear regulator from a switching power supply in future (unless you spend good money on a good regulator)  


[edited, futile LED conversation, if i had a linear regulator, the voltage would not have increased no matter how how, no drift]
« Last Edit: January 11, 2013, 07:29:42 am by cjdelphi » Logged

SE USA
Offline Offline
Faraday Member
**
Karma: 41
Posts: 3783
@ssh0le
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

it seems nothing runs directly from the wall correctly, but odd story..

I bought a bunch of 6 volt switching wall warts from china, to run though didode to LDO 5v regulator for a personal project

After some buyers remorse while they were in the mail I checked all my little switchers, cellphones, tablets, cameras, cable boxes everything

the little 1.20$ 6 volt warts I got were the only ones that even bothered to copy the UL logo, and on top of that provided a UL number (it could be for a toaster for all I know)

I have had one nailed across a very low resistance, high load for a while now, nearly maxing out its rating ... stupid thing still measures 5.99 volts, go figure
Logged


Offline Offline
Edison Member
*
Karma: 29
Posts: 2455
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I was refering to the switching regulators on ebay.

http://www.ebay.com.au/sch/i.html?_odkw=voltage+regulator&_sop=15&_osacat=0&_from=R40&_trksid=p2045573.m570.l1313&_nkw=step+down+voltage+regulator&_sacat=0




those kinds of boards....
Logged

SE USA
Offline Offline
Faraday Member
**
Karma: 41
Posts: 3783
@ssh0le
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

yea, add a bit of plastic and a plug, you essentially have a wall wart
Logged


SF Bay Area (USA)
Offline Offline
Tesla Member
***
Karma: 135
Posts: 6763
Strongly opinionated, but not official!
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
you essentially have a wall wart
Not really; they're designed for relatively low DC Vin (<35V), and don't even need to pretend to have any isolation or high voltage safety provisions (the way the plug-in supplies are supposed to have, and frequently don't.)

You'd think that a basic lm2596 "reference circuit" wouldn't have a lot that could be done wrong.  I wonder why you're seeing the behavior you see?  Bad caps?  Counterfeit chip?  Or real flaw in the chip?
Logged

Dubuque, Iowa, USA
Offline Offline
Faraday Member
**
Karma: 47
Posts: 2520
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

You're driving a LED based on voltage, then you're making it worse by using a regulator circuit that is specified at +/- 2.5% regulation. Grumpy Mike would kick your butt for doing it that way. There's nothing about a LED V/I curve that suggests anything but eventual failure.

It's also not necessarily a knock-off/counterfit LM2596 issue. Nat Semi specs their LM2596 at +/- 4% regulation:
Quote
Other features include a guaranteed ± 4% tolerance on out-
put voltage under specified input voltage and output load
conditions...
Logged

Valencia, Spain
Offline Offline
Faraday Member
**
Karma: 146
Posts: 5535
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I've blown several high power (>1amp cree) LED's with these, when they get hot (or becomes shorted) the regulated voltage increases, so after running it for an hour, the temp slowly creeped up, with heat came a slow rising voltage until death....

Sounds to me like you're using the wrong sort of board and overloading it. Anything will fail if it gets hot enough.

LEDs are supposed to be driven by current, not voltage. They make those boards with constant-current output (adjustable):

http://www.ebay.com/sch/i.html?_nkw=constant+current+driver

The boards with three adjusters on them also have a shutdown voltage, for charging batteries. ie. The board supplies a constant current then shuts down if the output voltage reaches a certain voltage level.

If you want to go the whole hog (and you should if you're using expensive Cree LEDs you can even get them with heatsinks, just in case.

Logged

No, I don't answer questions sent in private messages (but I do accept thank-you notes...)

Valencia, Spain
Offline Offline
Faraday Member
**
Karma: 146
Posts: 5535
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

what  i do and do not do with my LED is my own buisness...

Sure...

secondly... You're wrong what killed my LED's is not current at all but heat.... these power supplies got hot to drive the led which  increases the light, truth is even at 3 amps which i can assure you these can not deliver at 4.2v ... even at 12v and even if it could... it's well within max current specifications... but no all i required was a lot bigger heat sink... heat and heat alone caused the death as the regulator crept up in voltage so did the current death from heat NOT anything else

If the thing was getting too hot then you were overloading it.

What counts is the power dissipation (wattage), not the amps. The board may be rated for 2A max but 2A at 5V isn't the same as 2A at 20V. The voltage conversion is only 90% efficient at best, the rest turns into heat.
Logged

No, I don't answer questions sent in private messages (but I do accept thank-you notes...)

Offline Offline
Sr. Member
****
Karma: 3
Posts: 253
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

If you look at the board you posted there no way that board can handle a 20 volt 2 amp drop that's 40 watts
the chip should shut down, lower its voltage not increase it.
Logged

Offline Offline
Edison Member
*
Karma: 116
Posts: 2205
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
I've blown several high power (>1amp cree) LED's with these, when they get hot (or becomes shorted) the regulated voltage increases

If the regulated voltage increases when those devices are shorted, those are miracle devices.

What you are likely observing, without realizing it, is why a serial resistor is sometimes needed with leds: to prevent thermal runaway.
Logged

Offline Offline
Edison Member
*
Karma: 116
Posts: 2205
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
20 volt 2 amp drop that's 40 watts

This is a switching mode regulator.

Quote
There's onboard thermal regulation which prevents anything excess current being delivered to the LED.

That doesn't mean the less-than-excessive voltage from the regulator wouldn't cause excessive current / power dissipation on the led.

No, the board does not regulate current output at all: they call it a voltage regulator for a reason.
Logged

Offline Offline
Edison Member
*
Karma: 116
Posts: 2205
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
case closed.

I am actually still curious as to how its output voltage increased when  you shorted it.

It is so out of this world.
Logged

Offline Offline
Edison Member
*
Karma: 29
Posts: 2455
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

want a video?

I can show a stable voltage on video to the device you're powering, but decrease the resistance the voltage increases.
Logged

Offline Offline
Edison Member
*
Karma: 116
Posts: 2205
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
I can show a stable voltage on video to the device you're powering, but decrease the resistance the voltage increases.

That's probably quite different than saying that the voltage increases when you short them.
Logged

Offline Offline
Sr. Member
****
Karma: 3
Posts: 253
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
20 volt 2 amp drop that's 40 watts

This is a switching mode regulator.

Quote
There's onboard thermal regulation which prevents anything excess current being delivered to the LED.

That doesn't mean the less-than-excessive voltage from the regulator wouldn't cause excessive current / power dissipation on the led.

No, the board does not regulate current output at all: they call it a voltage regulator for a reason.


Read the data sheet it said if the output is shorted the chip decreases the voltage

you saying ohm 's law don't apply dherry if the voltage goes down the current can't go up.

20 volt 1 ohm load 20amps  5 volt 1 ohm load 5 amps

Now whats this say
Quote
(3) The oscillator frequency reduces to approximately 18 kHz in the event of an output short or an overload which causes the regulated
output voltage to drop approximately 40% from the nominal output voltage. This self protection feature lowers the average power
dissipation of the IC by lowering the minimum duty cycle from 5% down to approximately 2%.
« Last Edit: January 11, 2013, 07:55:03 am by be80be » Logged

Pages: [1] 2 3   Go Up
Jump to: