Author Topic: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault  (Read 1524 times)

0 Members and 1 Guest are viewing this topic.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 25886
  • Country: au
    • EEVblog
EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« on: May 25, 2018, 09:25:56 am »
An UNBELIEVABLE fault in Dave's Alesis M1 Active 520USB powered studio monitor speaker.
Can you discover the fault before Dave does?
This was supposed to be a boring trivial repair of a power LED, but it turned into something much more interesting.



Teardown:

 

Online TheSteve

  • Supporter
  • ****
  • Posts: 2410
  • Country: ca
  • GHz or bust
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #1 on: May 25, 2018, 11:56:12 am »
I've seen many LED's fail like that with LED Christmas lights.
VE7FM
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 25886
  • Country: au
    • EEVblog
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #2 on: May 25, 2018, 12:26:43 pm »
I've seen many LED's fail like that with LED Christmas lights.

But are were they current dependent or just intermittent contact?
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 2410
  • Country: ca
  • GHz or bust
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #3 on: May 25, 2018, 12:47:19 pm »
I've seen many LED's fail like that with LED Christmas lights.

But are were they current dependent or just intermittent contact?

I've seen variations of both. The fault is always internal to the LED itself.
VE7FM
 

Offline Harb

  • Regular Contributor
  • *
  • Posts: 239
  • Country: au
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #4 on: May 25, 2018, 01:02:43 pm »
I have only ever seen this once myself, and it was with a single colour blue led when they first came out......inspecting the led under a microscope you could see a very small arc on one of the bond wires as you cranked up the supply.....it was very repeatable and you would swear it was designed to work that way.
« Last Edit: May 25, 2018, 02:15:03 pm by Harb »
 

Online edpalmer42

  • Super Contributor
  • ***
  • Posts: 1333
  • Country: ca
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #5 on: May 25, 2018, 01:06:19 pm »
Wouldn't a leaky red led have the same effect?

Ed
 

Offline Harb

  • Regular Contributor
  • *
  • Posts: 239
  • Country: au
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #6 on: May 25, 2018, 01:08:48 pm »
Wouldn't a leaky red led have the same effect?

Ed

I don't think the colour would make any difference.......in my case I think it was probably due to the race to get blue led's out on the market and as a result maybe QC wasn't quiet up to standard.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 25886
  • Country: au
    • EEVblog
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #7 on: May 25, 2018, 01:43:04 pm »
Wouldn't a leaky red led have the same effect?

The blue LED failure scenario is much more plausible IMO
 

Offline Ivan7enych

  • Regular Contributor
  • *
  • Posts: 136
  • Country: ru
    • My astronomy projects
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #8 on: May 25, 2018, 06:12:02 pm »
It would be interesting to put this led in some curve tracer (signal generator + resistor + oscilloscope in XY mode), to see the voltage curve in low current (0-20mA) range. There could be some funny weird effects visible on this curve.
 

Offline rollatorwieltje

  • Supporter
  • ****
  • Posts: 566
  • Country: nl
  • I brick your boards.
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #9 on: May 25, 2018, 07:09:25 pm »
We have had issues with blue LEDs in a sensor setup (a device to measure the color of a liquid), they degraded much faster than other colors. We needed some serious compensation for it in software. The other LEDs would easily outlast the actual device and still be in spec. The blue one degraded by more than 40% in 10 years, while the others were less than 5%. It wasn't temperature related, the boards were kept at room temperature. They also ran at below maximum current.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 4172
  • Country: de
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #10 on: May 25, 2018, 07:49:47 pm »
I had an IR LED in an wireless headset (those old ones with IR transmission) to fail in a blinking way. For quite some time it did an about 1 second on, 1 second off.  The good think was that those 880 nm LEDs are still a little visible.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 6373
  • Country: nz
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #11 on: May 25, 2018, 10:21:26 pm »
Ive seen damaged leds behaving very randomly .

Getting brighter/dimer turning on/off etc..
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline W9GFO

  • Regular Contributor
  • *
  • Posts: 52
  • Country: us
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #12 on: May 26, 2018, 01:29:14 am »
The same thing happened to my Alesis M1 520USB, the blue led worked intermittently after a few days of use, then a few days later went out and never worked again.
 

Offline dmg

  • Regular Contributor
  • *
  • Posts: 88
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #13 on: May 26, 2018, 04:32:19 am »
I got a failure in a lab PSU screen backlight, which suddenly started to flicker. When you powered it on it stayed ON for a while and then it started flickering at a quite high frequency. The issue was some a bad contact inside the LED itself. When you applied power to it it turned on, but after several seconds it reached a certain temperature where thermal expansion interrupted the bad contact, the LED then turned off, cooled down, the contact engaged again and the cycle repeated. I had some fun experimenting with it, seeing how it changed frequency with temperature. If you submerged it in ice cold water it stayed on indefinitely, and at about 40ÂșC it stopped working.

It finally died eventually.
 

Offline thm_w

  • Frequent Contributor
  • **
  • Posts: 647
  • Country: ca
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #14 on: May 26, 2018, 05:45:49 am »
Wonder what the standby power consumption is on this. If its properly applying standby to the amplifier IC its probably not too bad, but still terrible compared to SMPS + Class D.
Couldn't find anything in their manual for specifications.

Reminds me of the night light LEDs that big clive looked at, where the transistor shunts out the LED during daylight. Very simple and cheap design.
 

Offline nuno

  • Frequent Contributor
  • **
  • Posts: 534
  • Country: pt
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #15 on: May 26, 2018, 05:55:49 am »
Ive seen damaged leds behaving very randomly .

Getting brighter/dimer turning on/off etc..

+1

If you stress them beyhond specs (I've done this on white ones) but take care not to blow them completely, they'll enter random mode.
 

Offline thm_w

  • Frequent Contributor
  • **
  • Posts: 647
  • Country: ca
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #16 on: May 26, 2018, 08:08:01 am »
If you stress them beyond specs (I've done this on white ones) but take care not to blow them completely, they'll enter random mode.

Surely 10mA is not beyond the specs?
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 25886
  • Country: au
    • EEVblog
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #17 on: May 26, 2018, 09:14:49 am »
The same thing happened to my Alesis M1 520USB, the blue led worked intermittently after a few days of use, then a few days later went out and never worked again.

Woah, so it's a thing. Interesting
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 25886
  • Country: au
    • EEVblog
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #18 on: May 26, 2018, 09:16:24 am »
Wonder what the standby power consumption is on this. If its properly applying standby to the amplifier IC its probably not too bad, but still terrible compared to SMPS + Class D.
Couldn't find anything in their manual for specifications.

I should measure it, but I don't have power meter at the office.
It's likely quite poor.
 

Offline Gary350z

  • Regular Contributor
  • *
  • Posts: 134
  • Country: us
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #19 on: May 26, 2018, 01:52:42 pm »
Wonder what the standby power consumption is on this. If its properly applying standby to the amplifier IC its probably not too bad, but still terrible compared to SMPS + Class D.
Couldn't find anything in their manual for specifications.

I should measure it, but I don't have power meter at the office.
It's likely quite poor.

Could you fly to the moon on that? ;D
 

Offline ggchab

  • Regular Contributor
  • *
  • Posts: 118
  • Country: be
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #20 on: May 28, 2018, 10:40:08 pm »
The Elektor magazine once reported the case of a LED that started to flash and then died soon after. I think there were some hypotheses (https://www.elektormagazine.fr/magazine/elektor-201109/11964). Unfortunately, a login is required  :-\
 

Offline Techoisseur

  • Contributor
  • Posts: 9
  • Country: gb
  • EE (Retd)
Re: EEVblog #1087 - UNBELIEVABLE Alesis Studio Monitor Fault
« Reply #21 on: June 18, 2018, 08:13:01 pm »
Certainly it is an interesting phenomenon.
It may be caused by dendritic growth within the moulding compound surrounding the leadouts of the Blue LEDs.
If the two leadout wires are close together, and there is some moisture content, or uncured resin or activator within the moulding, then the potential gradient may be sufficient, over a long period with applied voltage bias, to cause dendrites to form within the moulding by an electrochemical migration process.
Eventually, these dendrites may result in a resistive conductive channel across the LED, which will divert the current that was intended for the LED. When an increased current is supplied, this blows away part of the dendrite (like a fuse blowing), allowing the LED to work again. However, when the current is reduced, the dendrites grow back again so that the LED becomes intermittent and then stops working.
Another mechanism is the growth of tin whiskers. However, these are much smaller diameter, and will probably cause a direct short, which would have been detected by your multimeter tests. Also I believe that the tin whiskers would grow regardless of the applied voltage bias.
The dendrites may be visible (assuming a transparent moulding) with your microscope, as a dark channel between the leadout frame.
https://nepp.nasa.gov/whisker/background/index.htm
Hope this helps  :)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf