Author Topic: The brown corrosive gunk of death  (Read 4308 times)

0 Members and 1 Guest are viewing this topic.

Offline TerraHertz

  • Super Contributor
  • ***
  • Posts: 3721
  • Country: au
  • Why shouldn't we question everything?
    • It's not really a Blog
The brown corrosive gunk of death
« on: July 15, 2016, 02:10:04 pm »
Today I was given a bunch of wireless 'managed access points' that go with a 3COM POE switch I was also given. The switch was dead; that was an easy power supply fix. Of the access points, most were in a box marked "dead AP" so it's assumed they are dead. From the first two opened so far, I think I know why.

The manufacturer used that brown gunk to secure three components in each unit. It dries to a rigid solid. Who else recognizes that stuff, and knows that it has been ruining electronics for many decades? Or longer; I'm pretty sure I remember seeing the same stuff used in electronics I pulled apart as a child, and noticing it had the same destructive effect then.

Because it's *really* corrosive. Everything it touches becomes corroded. And always has for as long as this stupid crap has been around. Yet it's still used, even on high density surface mount PCBs like these. Which probably cost a fortune when they were made, only to be doomed by three dabs of this acidic poison.

When are manufacturers going to learn not to use this stuff?
Or is it a deliberate planned obsolescence death sentence?
Collecting old scopes, logic analyzers, and unfinished projects. http://everist.org
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2437
  • Country: us
Re: The brown corrosive gunk of death
« Reply #1 on: July 15, 2016, 05:13:44 pm »
Hi

Tough to tell without poking at it. It looks a lot like urethane. You can get versions that are electronics compatible.

Bob
 

Offline elecman14

  • Supporter
  • ****
  • Posts: 160
  • Country: us
Re: The brown corrosive gunk of death
« Reply #2 on: July 15, 2016, 05:34:01 pm »
It may not actually be corrosive. What may be happening is that the goop is causing electromigration. That in turn can cause conductive anodic filaments. They are really cool to look at under a microscope and keep me up at night with fears of product reliability  :=\. It appears the manufacturer probably should have done more environmental testing of their products. Then again it is a "consumer" product so that generally is not typical (depending on the market).  Out of curiosity have you lived in a humid, moist and/or hot environment during your observations of this failure?
   
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2437
  • Country: us
Re: The brown corrosive gunk of death
« Reply #3 on: July 15, 2016, 05:38:18 pm »
Hi

It could be as simple as:

Step one: Put on goop

Step two: Wash assembly contamination off the board for the first time

Reverse the order and you no longer trap a ton of crud under the goop.

Bob
 

Offline Tomorokoshi

  • Frequent Contributor
  • **
  • Posts: 911
  • Country: us
Re: The brown corrosive gunk of death
« Reply #4 on: July 15, 2016, 05:48:53 pm »
It may not actually be corrosive. What may be happening is that the goop is causing electromigration. That in turn can cause conductive anodic filaments. They are really cool to look at under a microscope and keep me up at night with fears of product reliability  :=\. It appears the manufacturer probably should have done more environmental testing of their products. Then again it is a "consumer" product so that generally is not typical (depending on the market).  Out of curiosity have you lived in a humid, moist and/or hot environment during your observations of this failure?
 

If that's the case, then I wonder if there is much corrosion like that with non-RoHS components.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 19218
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The brown corrosive gunk of death
« Reply #5 on: July 16, 2016, 07:35:16 am »
It's a semi hard and brittle porous foam and it's main claim to fame is its hygroscopic properties that promote conductivity and corrosion by attracting and holding moisture against componentry.
While that might not be too bad if there wasn't exposed dissimilar metals and various voltages in electronic equipment to allow active corrosion to propagate and also provide moisture enabled conductive paths.

When equipment is powered continuously it creates few problems, the few degrees of warmth that a standby or SMPS PSU provides can keep the moisture away, moisture being the main downfall of this stuff.

But leave equipment unpowered or place it in storage....do so at your peril.
Return unit to service....BANG !

TerraHertz's overly kind thread title does not do this brown muck justice, it is pure crap. Avoid it like the plague.
Avid Rabid Hobbyist
 

Offline Wirehead

  • Regular Contributor
  • *
  • Posts: 166
  • Country: be
    • Wirehead.be
Re: The brown corrosive gunk of death
« Reply #6 on: July 16, 2016, 08:38:55 am »
Common problem on audio amplifiers. Used to secure big caps etc. Eats through jumpers, resistors,.. Awfull stuff. It does have a purpose, and that's to lower mechanical stress. But.. Man, that stuff does suck.
"to remain static is to lose ground"
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 14037
  • Country: cn
  • Power Electronics Guy
Re: The brown corrosive gunk of death
« Reply #7 on: July 16, 2016, 08:48:42 am »
I do not think 3COM will do planned obsolescence. They are known for building high quality telecom quality gears.
I think it is just poor reliability investigation. Someone said: hey, let's use this compound, cheaper and stronger. Then the boss probably said okay without doing long term reliability research, nor hiring chemists to do composition analysis. Then 20 years later, the chemical reaction killed traces.
There are many competent EE companies out there, but only a handful of EE companies that can, and are willing to do thorough reliability analysis on mass produced products.
 

Offline Don Hills

  • Regular Contributor
  • *
  • Posts: 159
Re: The brown corrosive gunk of death
« Reply #8 on: July 16, 2016, 09:23:06 am »
That looks like a different "corrosive brown gunk" to that which I'm familiar.
The stuff I remember starts out white and somewhat flexible when new, with a smooth surface. It also has a distinctly different drying pattern than the stuff you show, which looks more like ADOS F2 or similar. As it ages, it darkens and hardens, hastened by heat. It eventually turns almost black, and extremely brittle - poking it with a probe causes it to crumble. In this state it also becomes very corrosive, I've seen thick through-hole component legs eaten clean through.
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 9773
  • Country: gb
Re: The brown corrosive gunk of death
« Reply #9 on: July 16, 2016, 09:42:34 am »
What an interesting thread. My thanks to the OP.

I must be lucky as I have only found this stuff on old CRT TV's  Hi-Fi systems, and not more modern kit. I a surprised that 3Com use it. The softer RTV mechanical support is what I see most often these days.

This brown gunk looks like it is very unsuitable for use with electronics. As stated, why use a material that has well known hydroscopic properties. Most odd I must say.

I will keep a lookout for this stuff in future and remove it, from any PCB's  that I work on.

Fraser
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 7808
  • Country: de
Re: The brown corrosive gunk of death
« Reply #10 on: July 16, 2016, 12:02:23 pm »
I found trouble with this gunk once in an audio amplifier, that was on most of the time. So I don't think humidity was the problem. If turned slightly conductive: measured 100 K between adjacent lines. Seemed to have been a rather messy stuff, lots of parts contaminated that don't need any glue, just like it was hard to apply in controlled way. 

The other problem such glue is it's rather difficult to get off to replace parts.

There are different types used - so they have more than one type of gunk to choose from. Some types stay silicone like even when older, some seem to crystallize and same are extremely hard. A hygroscopic one would be a really bad idea.
 

Online chris_leyson

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: wales
Re: The brown corrosive gunk of death
« Reply #11 on: July 16, 2016, 12:28:07 pm »
Being hygroscopic would certainly explain things, don't think I have ever seen a PCB in such bad condition will watch out for this in future. What I dont get is why they use it in the first place, I could understand it if you secure say, a few verticall mounted electrolytics, then you get some additional rigidity, but on a surface mount HC49-4H crystal package ? Ever tried taking one of a PCB it's very difficult you need a lot of heat and the pads are huge so there is a good bond between the copper and the PCB substrate. HC49-4H packages just do not drop off PCBs, a vertical mount leaded crystal is another matter.
 

Online george.b

  • Frequent Contributor
  • **
  • Posts: 269
  • Country: br
Re: The brown corrosive gunk of death
« Reply #12 on: July 16, 2016, 12:38:08 pm »
I do not think 3COM will do planned obsolescence. They are known for building high quality telecom quality gears.

Hmh, I'd think otherwise, that 3com would know better, if that wasn't intentional.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15425
  • Country: za
Re: The brown corrosive gunk of death
« Reply #13 on: July 16, 2016, 03:10:48 pm »
Closest glue I have seen to that is Pratley Pliobond, which is basically synthetic rubber, chalk and solvent in a tube. Thus the rubber degrades with heat and time after exposure to atmospheric oxygen, the filler chalk often is the lowest grade contaminated reject powder, which is bargain basement price because it is either contaminated with heavy metal, leachate from a mine runoff, cyanide or other organic chemicals or is simply a reground recycled concrete blend. Solvent is whatever would dissolve the rubber so could be almost anything or a blend of whatever.

Thus the glue will with time both decompose as the rubber dies, and the moisture in the air will be attracted to the filler and wash out the soluble ions there to corrode the board.

I guess they did a dab of glue to fix the caps, and the person doing the application rested the glue gun on the board while doing something, leading to the drop on the crystal. As well probably the original spec called for a certain manufacturers part and this was substituted by the assembler as the original spec was either expensive or they had a stock of the cheap junk already, used on the simple SRBP TV boards and PSU boards they normally make for others. There a slightly corrosive glue is not a worry on a single sided board with no voltage across the glue and only plastic sleeve and anodised aluminium oxide capacitors there.
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 6562
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: The brown corrosive gunk of death
« Reply #14 on: July 16, 2016, 05:20:02 pm »
If scraping off the glue doesn't fix it, just throw it in the parts bin. An old 11g AP isn't worth much nowadays. Even 11n is getting dated.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline Wirehead

  • Regular Contributor
  • *
  • Posts: 166
  • Country: be
    • Wirehead.be
Re: The brown corrosive gunk of death
« Reply #15 on: July 16, 2016, 06:29:48 pm »
These can be handy for some temporary point-to-point links. They may have their use. Not worth the hassle selling, but handy to have around..
"to remain static is to lose ground"
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3062
  • Country: us
Re: The brown corrosive gunk of death
« Reply #16 on: July 16, 2016, 06:39:45 pm »
The adhesive is applied to the larger parts which are only partly supported by their pads (big inductors and capacitors). That's because shear forces from dropped shipping packages can knock those components off. I expect it's a tradeoff where the manufacturer has a lower initial DOA rate when they use the goop. As you've seen, over time and humidity, the damage is done.
I don't recall seeing this brown stuff used very often. Usually in high-voltage areas like CRTs and PSUs, white RTV silicone is used. I recently repaired an automotive circuit, that used a quite hard, clear amber adhesive for mechanical support.
« Last Edit: July 16, 2016, 06:43:24 pm by helius »
 

Offline TerraHertz

  • Super Contributor
  • ***
  • Posts: 3721
  • Country: au
  • Why shouldn't we question everything?
    • It's not really a Blog
Re: The brown corrosive gunk of death
« Reply #17 on: July 16, 2016, 08:20:28 pm »
If scraping off the glue doesn't fix it, just throw it in the parts bin. An old 11g AP isn't worth much nowadays. Even 11n is getting dated.

Well, so far I was given seven of the AP2750 managed wireless access points 3CRWX275075A, and one 3COM WX1200 wireless LAN switch  3CRWX120695A   6-port POE that runs them.
But there'll be more: 3 more switches and perhaps a dozen more APs, all apparently working.

Two APs opened so far, both with identical gunk placement and corrosion. Looks like it's going to be the primary cause of failure.


Collecting old scopes, logic analyzers, and unfinished projects. http://everist.org
 

Offline xygor

  • Regular Contributor
  • *
  • Posts: 228
  • Country: us
Re: The brown corrosive gunk of death
« Reply #18 on: July 18, 2016, 05:29:09 pm »
How did you determine that that was the cause of the failure?
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3328
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: The brown corrosive gunk of death
« Reply #19 on: July 18, 2016, 06:05:46 pm »
Yes guys, 3com does plan obsolescence. I am sure with that.

Recently, I have repaired some older piece of 3com switch with two gbic ports.  The problem was in a power supply (who would expect that, huh?) Everywhere in the unit, Nichicon caps were used. Only in a single critical place, a shit chinese cap on the main SMPS controller IC supply was used and glued together with some power shunt resistor.  Dry as ... fucking dry!

So do not tell me, that 3com does not do planned obsolescence, if that was absolutely obvious, that was done deliberately to decrease the life expectancy considerably.

If I will manage to find the photos from repair, I'll post them.
 
The following users thanked this post: Kilrah

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15425
  • Country: za
Re: The brown corrosive gunk of death
« Reply #20 on: July 18, 2016, 08:01:05 pm »
Fixed up a few small switches that way as well, cheap caps in the power supply dried up, and as a result massive ripple on the power rail. The ST labs one as well had the plug in cheap and nasty SMPS blow up on the primary side, which from the split open capacitors on the secondary was a mercy.

Seems it died low voltage, as it worked well with a new power supply repurposed from an old cellphone brick, which gives 5V7 ( Nokia one from an old phone). I have seen these deliver over 50V when the feedback stopped working, which was not good for the stuff plugged in that was expecting a supply of 8V.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf