Searching \ for '[EE] To socket or not to socket' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=socket+not+socket
Search entire site for: 'To socket or not to socket'.

Exact match. Not showing close matches.
PICList Thread
'[EE] To socket or not to socket'
2006\07\01@150603 by Goran Hosinsky

picon face
I am placing small circuits with a PIC in my garden, in
plastic boxes. They are protected from rain but the
boxes are not hermetic, cables coming in from below,
so ambient humidity, ants, spiders etc can sometimes
come in.

Normally I always use sockets for mounting ICs but in
this case I wonder, if it would be better not to use
sockets to avoid the posibiity of oxidation and
faulty contact between the PICs pins and the socket.

Any recommendations?

Goran

2006\07\01@165141 by Kenneth Lumia

picon face
Goran,

Don't use the sockets!  Also, once the design is complete and
the board cleaned up, spray it with a sealer - I've used GC
electronics Acrylic Plastic, but there are other types of
coatings providing even better protection.  Try looking
for a urathane or silicone conformal type.  I think
Tech-Spray has a suitable line.

Ken
spam_OUTklumiaTakeThisOuTspamadelphia.net

{Original Message removed}

2006\07\01@170957 by Robert Rolf

picon face
If you don't have adequate lightening protection, you'll want
sockets.

Use silicon grease on the pins to make them watertight.
e.g. bead of grease on each row of pins, and then grease bead
after the chip is inserted. The mechanical action of a good
socket will wipe the grease off the pin where it makes contact.

You'd be better off sealing the boxes. Silcon sealant (RTV) applied
to the cable entry points works well, and is easily peeled off
if you need to service things.

But if you DO have good lightening protection on the wiring,
then do as Ken suggests and fully seal the units.
You can seal it with socketed parts, but it is a real PITA to
clean off the sealer to do repair work.

Robert

Kenneth Lumia wrote:

{Quote hidden}

> {Original Message removed}

2006\07\01@185316 by Carey Fisher

face picon face


Goran Hosinsky wrote:
> I am placing small circuits with a PIC in my garden, in
> plastic boxes. They are protected from rain but the
> boxes are not hermetic, cables coming in from below,
> so ambient humidity, ants, spiders etc can sometimes
> come in.
>
> Normally I always use sockets for mounting ICs but in
> this case I wonder, if it would be better not to use
> sockets to avoid the posibiity of oxidation and
> faulty contact between the PICs pins and the socket.
>
> Any recommendations?
>
> Goran
>  
If the boxes are not hermetically sealed, drill drain holes in the
bottoms. I'm not kidding!!  The boxes will breathe with changes in
temperature and you'll end up with liquid water trapped in the boxes.  I
know cause I've put tuning circuits on outdoor antennas for 30 years or so.

It is a good idea to coat the PCBds with a urethane spray to prevent
problems with humidity/moisture.  And don't use sockets due to corrosion
problems over time.

Carey
--

*Carey Fisher, Chief Technical Officer
New Communications Solutions, LLC
*careyfisherspamKILLspamncsradio.com <.....careyfisherKILLspamspam.....ncsradio.com>
Toll Free Phone:888-883-5788
Local Phone:770-814-0683
FAX: 888-883-5788
http://www.ncsradio.com <http://www.ncsradio.com/>

2006\07\02@010848 by Peter Todd

picon face
On Sat, Jul 01, 2006 at 06:53:23PM -0400, Carey Fisher wrote:
> It is a good idea to coat the PCBds with a urethane spray to prevent
> problems with humidity/moisture.  And don't use sockets due to corrosion
> problems over time.

Here's a related question...

If you do use sockets, are machine pin ones really all that much better
than the cheap wiper ones?

Tin plated vs. gold?

--
EraseMEpetespam_OUTspamTakeThisOuTpetertodd.ca http://www.petertodd.ca

2006\07\02@032029 by Vasile Surducan

face picon face
On 7/2/06, Carey Fisher <careyfisherspamspam_OUTncsradio.com> wrote:
{Quote hidden}

Even the boxes are hermetically sealed, what make you think will not be
moisture inside ?  Huge temperature variations will create sweat
inside a hermetical sealed box as well in an opened one.
Critical temperature is 7-8C when the temperature is increasing from 0C.

greetings,
Vasile

2006\07\02@035110 by Peter Todd

picon face
On Sun, Jul 02, 2006 at 10:20:29AM +0300, Vasile Surducan wrote:
> > If the boxes are not hermetically sealed, drill drain holes in the
> > bottoms. I'm not kidding!!
>
>
> Even the boxes are hermetically sealed, what make you think will not be
> moisture inside ?  Huge temperature variations will create sweat
> inside a hermetical sealed box as well in an opened one.
> Critical temperature is 7-8C when the temperature is increasing from 0C.

Not that I've ever done this sort of work but what's wrong with throwing
a couple bags of dessicant in the boxes? It's aparently good enough to
keep surface mount chips from going pop.

For something more complex, I'm sure a dry-nitrogen purge or some such
isn't so hard if you are willing to waste nitrogen. Flood a ad-hoc
glovebox and do the final assembly in it.

Just gotta have the nitrogen, and a regulator, well, and plenty of duct
tape!

--
@spam@peteKILLspamspampetertodd.ca http://www.petertodd.ca

2006\07\02@043041 by Xiaofan Chen

face picon face
On 7/2/06, Vasile Surducan <KILLspampiclist9KILLspamspamgmail.com> wrote:
> On 7/2/06, Carey Fisher <RemoveMEcareyfisherTakeThisOuTspamncsradio.com> wrote:
> > >
> > If the boxes are not hermetically sealed, drill drain holes in the
> > bottoms. I'm not kidding!!
>
> Even the boxes are hermetically sealed, what make you think will not be
> moisture inside ?  Huge temperature variations will create sweat
> inside a hermetical sealed box as well in an opened one.
> Critical temperature is 7-8C when the temperature is increasing from 0C.

I think Carey Fisher is saying that you can not drill drain holes
if the boxes are hermetically sealed. You are of course right moisture can
still occur even if it is hermetically sealed (eg: IP67).

This is not an easy question to be solved. I've seen solution which uses
a membrane. This is a level sensor with metal housing and it is used
in those beverage industry for CIP (Cleaning In Progress) application.

On a related issue, some IP67 sensors with plastic housing can be
swollen after temperature cycle and I do not know if there are
simple and cheap solution for this.

Regards,
Xiaofan

2006\07\02@071637 by Alden Hart

flavicon
face
Goran Hosinsky wrote:
> I am placing small circuits with a PIC in my garden, in
> plastic boxes. They are protected from rain but the
> boxes are not hermetic, cables coming in from below,
> so ambient humidity, ants, spiders etc can sometimes
> come in.
>
> Normally I always use sockets for mounting ICs but in
> this case I wonder, if it would be better not to use
> sockets to avoid the posibiity of oxidation and
> faulty contact between the PICs pins and the socket.
>
> Any recommendations?
>
> Goran
>  
If you can stand module being disposable, is there any reason not to
simply cast the entire module in acrylic resin - perhaps leaving the
ICSP lines accessible? A good seal around cable access would also be
necessary. You might also need a heatsink to protrude from the assembly
if you are dissipating any significant power.

Alden

2006\07\02@071838 by Bob Axtell

face picon face
There was a time- long, long ago and far, far away- when IC's were not
reliable.
Those days are gone. Sockets should be used ONLY if you have a device
exposed to lightning, power accidents, etc. If you MUST use a socket, it
needs to
be gold contact type.

Sockets also restrict normal IC cooling. Even optoisolaters, which will
need replacement
after a few years, the use of a socket is not recommended; just clip it
out and replace
it every 5 years.

--Bob
>> {Original Message removed}

2006\07\02@081938 by Xiaofan Chen

face picon face
On 7/2/06, Bob Axtell <spamBeGoneengineerspamBeGonespamcotse.net> wrote:

> Sockets also restrict normal IC cooling. Even optoisolaters, which will
> need replacement after a few years, the use of a socket is not
> recommended; just clip it out and replace it every 5 years.
>

Where do you get this 5 years data? Opto-couplers are widely used
in the field. For example, lots of the I/O modules for PLC/DCS are using
them because of the isolation requirement. Maybe this is why they need to
replace the I/O modules from time to time. I do not think they will open
the modules and replace the opto-couplers.

Regards,
Xiaofan

2006\07\02@094442 by Carey Fisher

face picon face


Xiaofan Chen wrote:
> On 7/2/06, Vasile Surducan <TakeThisOuTpiclist9EraseMEspamspam_OUTgmail.com> wrote:
>  
>> On 7/2/06, Carey Fisher <RemoveMEcareyfisherspamTakeThisOuTncsradio.com> wrote:
>>    
>>> If the boxes are not hermetically sealed, drill drain holes in the
>>> bottoms. I'm not kidding!!
>>>      
>> Even the boxes are hermetically sealed, what make you think will not be
>> moisture inside ?  Huge temperature variations will create sweat
>> inside a hermetical sealed box as well in an opened one.
>> Critical temperature is 7-8C when the temperature is increasing from 0C.
>>    
>
> I think Carey Fisher is saying that you can not drill drain holes
> if the boxes are hermetically sealed. You are of course right moisture can
> still occur even if it is hermetically sealed (eg: IP67).
>  

No.  I don't know how you got that from my post.   What I am saying is
that since you are not going to hermetically seal the boxes (and purge
them with dry air or nitrogen), you need to make large enough openings
in the box so that moisture doesn't accumulate because a closed but not
hermetically sealed box will breath with temperature variations.  That
means when the air temperature is high, air will leave the box; when air
temperature is low, air will enter the box.  And during this breathing,
moisture will get trapped in the box due to condensation.  Drain holes
help by reducing the amount of moisture trapped in the box and they let
any accumulated moisture drain out.
--

*Carey Fisher, Chief Technical Officer
New Communications Solutions, LLC
*careyfisherEraseMEspam.....ncsradio.com <EraseMEcareyfisherspamncsradio.com>
Toll Free Phone:888-883-5788
Local Phone:770-814-0683
FAX: 888-883-5788
http://www.ncsradio.com <http://www.ncsradio.com/>

2006\07\02@094717 by Bob Axtell

face picon face
Xiaofan Chen wrote:
> On 7/2/06, Bob Axtell <RemoveMEengineerEraseMEspamEraseMEcotse.net> wrote:
>
>  
>> Sockets also restrict normal IC cooling. Even optoisolaters, which will
>> need replacement after a few years, the use of a socket is not
>> recommended; just clip it out and replace it every 5 years.
>>
>>    
>
> Where do you get this 5 years data? Opto-couplers are widely used
> in the field. For example, lots of the I/O modules for PLC/DCS are using
> them because of the isolation requirement. Maybe this is why they need to
> replace the I/O modules from time to time. I do not think they will open
> the modules and replace the opto-couplers.
>
> Regards,
> Xiaofan
>  
Sorry, but its a fact.

All IR emitters degrade with usage. LED's degrade, too.The degradation
is dependent on how hard the
IR is driven. The output drops about 15% over 5 years, again dependent
on how hard they are driven.
I have a  few old LEDs that are so weak that its hard to decide if they
are red or not, and they are only 20
years old. You can get about 10,000 hits if you Google "LED Degradation".

I've designed with opto-isolators, too. But their lifetime is limited,
unlike most other components
on the shelf. This limitation  is simply dealt with, like replacing  wet
electrolytics  (which also has about
a 5 year effective lifetime).

I think the power module makers deliberately overdrive them and average
out their lifetime, expecting
them to be replaced in 10 years.

--Bob

2006\07\02@101729 by Dave Lag

picon face
Bob Axtell wrote:
> There was a time- long, long ago and far, far away- when IC's were not
> reliable.
> Those days are gone. Sockets should be used ONLY if you have a device
> exposed to lightning, power accidents, etc. If you MUST use a socket, it
> needs to
> be gold contact type.

> --Bob

Interesting Bob, any exceptions?
Hobby users?, prototypes?, single sided boards-holes without plate-thru?

I vaguely recall some argument about tin sockets vs gold ones to avoid a
dissimilar metal junction?
D

2006\07\02@103947 by Xiaofan Chen

face picon face
On 7/2/06, Carey Fisher <RemoveMEcareyfisherspam_OUTspamKILLspamncsradio.com> wrote:

{Quote hidden}

Yes this is exactly my understanding from your post. Maybe it is
just a problem with my English expressions (which is quite okay
as a non-native speaker but not so good compare to native speakers).

What I want to say is that drain holes will normally make the product
not hermetically sealed any more.

Regards,
Xiaofan

2006\07\02@104301 by Spehro Pefhany

picon face
At 06:45 AM 7/2/2006 -0700, you wrote:


>All IR emitters degrade with usage. LED's degrade, too.The degradation
>is dependent on how hard the
>IR is driven. The output drops about 15% over 5 years, again dependent
>on how hard they are driven.
>I have a  few old LEDs that are so weak that its hard to decide if they
>are red or not, and they are only 20
>years old. You can get about 10,000 hits if you Google "LED Degradation".

Yes, it's an issue that's received a lot of attention and many improvements
have been made in semiconductor processes.

>I've designed with opto-isolators, too. But their lifetime is limited,
>unlike most other components
>on the shelf.

The thing is, the lifetime is a function of the quality and performance
of the parts, and how conservative the design is. Degradation is
accelerated at high LED current and at high temperatures. They don't
degrade just sitting there. I am a big fan of really conservative designs,
because of the kind of work I do.

Using a good quality optoisolator, even at 60°C, the CTR drops only
20% in 100,000 hours of operation. That's 11 years with the LED on
at 5mA 24/7. If your circuit is designed with
100% margin, then you've got essentially unlimited lifetime, even at
relatively high temperature. At 25°C, the decrease is only half of that. And
often the LED duty cycle is much less than 100%.

You should not ship a design that is sensitive to only a 20% or 30% decrease
in CTR, for one thing it can drop 20%-35% just at high or low temperature,
and aging is on top of that. Cheap generic optos can be worse. And using
a crummy low-CTR opto with a high LED current (like 20-30mA rather than
1mA-5mA)
just compounds the problem.

These things are used in 'infrastructure' products such as PLCs and
telephone exchanges with expected lifetimes in the decades.

> This limitation  is simply dealt with, like replacing  wet
>electrolytics  (which also has about
>a 5 year effective lifetime).

Depends on operating temperature, ripple current, temperature rating and
so on. I'm sure we all have plenty of 20+ year old equipment with
aluminum electrolytics that is operating fine. Operating right at the
ratings yields significant failure rate within only a 2000 hour lifetime,
which is just a few months, so we don't typically design that way!

>I think the power module makers deliberately overdrive them and average
>out their lifetime, expecting
>them to be replaced in 10 years.
>--Bob

The Japanese engineers for a line of industrial controls told me they designed
for 5 years 24/7, but there were only 2-3 components that had such wear
mechanisms
*and* were stressed ( power supply electrolytics and relays). Still they
didn't socket the relays. The other electrolytics and optos were used very
conservatively and would last 20-30 years most likely. We tend to keep
$50-100K
machines running longer that 5 years in North America. Their stuff typically
worked for at least 10 years without problem, even at high temperature, but
then those few parts might need to be replaced to get another 10 reliably.

2006\07\02@105336 by Xiaofan Chen

face picon face
On 7/2/06, Bob Axtell <RemoveMEengineerTakeThisOuTspamspamcotse.net> wrote:
{Quote hidden}

I know this. Typical data from the vendors is 50,000 to 100,000 hours but
it is subjected to the temperature, driving current and other factors. Maybe
you get this data from the 50,000 hours.

> I think the power module makers deliberately overdrive them and average
> out their lifetime, expecting them to be replaced in 10 years.
>

No comments here...

Regards,
Xiaofan

2006\07\02@120159 by Gerhard Fiedler

picon face
Dave Lag wrote:

>> Those days are gone. Sockets should be used ONLY if you have a device
>> exposed to lightning, power accidents, etc. If you MUST use a socket,
>> it needs to be gold contact type.
>
> Interesting Bob, any exceptions?
> Hobby users?, prototypes?, single sided boards-holes without plate-thru?
>
> I vaguely recall some argument about tin sockets vs gold ones to avoid a
> dissimilar metal junction?

Not sure about the dissimilar metal junction. Are you talking about
corrosion? In my experience, tin sockets often don't last for many
insertion cycles. Depending on the situation, 10 or 20 may be all you get
reliably.

Gerhard

2006\07\02@124731 by Peter

picon face

To socket in gold plated sockets and to solder the chips into the
sockets if anything seems amiss later.

Peter

2006\07\02@130356 by Peter

picon face

On Sun, 2 Jul 2006, Vasile Surducan wrote:
> Even the boxes are hermetically sealed, what make you think will not be
> moisture inside ?  Huge temperature variations will create sweat
> inside a hermetical sealed box as well in an opened one.
> Critical temperature is 7-8C when the temperature is increasing from 0C.

A small bag of silica gel and conformal coating over sensitive parts of
the board fix this problem usually. VERY few long term devices for
outdoor use are hermetically enclosed and purged with dry nitrogen (the
'book move' for this kind of thing).

Peter

2006\07\02@140542 by Bob Axtell

face picon face
Dave Lag wrote:
> Bob Axtell wrote:
>  
>> There was a time- long, long ago and far, far away- when IC's were not
>> reliable.
>> Those days are gone. Sockets should be used ONLY if you have a device
>> exposed to lightning, power accidents, etc. If you MUST use a socket, it
>> needs to
>> be gold contact type.
>>    
>
>  
>> --Bob
>>    
>
> Interesting Bob, any exceptions?
> Hobby users?, prototypes?, single sided boards-holes without plate-thru?
>
>  
Oops. Prototypes and hobbyists can do whatever is needed. I always
assume a production use of
everything.

> I vaguely recall some argument about tin sockets vs gold ones to avoid a
> dissimilar metal junction?
> D
>  

2006\07\02@141157 by Bob Axtell

face picon face
Peter wrote:
> To socket in gold plated sockets and to solder the chips into the
> sockets if anything seems amiss later.
>
> Peter
>
>  
I never thought of that.

--Bob

2006\07\02@143245 by Bob Axtell

face picon face
Spehro Pefhany wrote:
{Quote hidden}

I have noticed significant improvements in recent years in optos.

{Quote hidden}

I did a lot of military stuff, so I still think that way, sorry. I try
to never design in wet electrolytics,
using tantalum or other solid types whenever I can.

--Bob
--Bob

>> -

2006\07\02@144307 by William Chops Westfield

face picon face

On Jul 2, 2006, at 6:45 AM, Bob Axtell wrote:

> All IR emitters degrade with usage. LED's degrade, too.  The
> degradation is dependent on how hard the IR is driven. The
> output drops about 15% over 5 years, again dependent on
> how hard they are driven.

Is that five years of ON time, or just 5 years?  If the former,
which sounds more reasonable, I'd expect that many optoisolator
applications have such a low duty cycle that 5 years of on time
is essentially "forever."

BillW

2006\07\02@145524 by Dave Lag

picon face
Gerhard Fiedler wrote:

>>I vaguely recall some argument about tin sockets vs gold ones to avoid a
>>dissimilar metal junction?
>
>
> Not sure about the dissimilar metal junction. Are you talking about
> corrosion? In my experience, tin sockets often don't last for many
> insertion cycles. Depending on the situation, 10 or 20 may be all you get
> reliably.
>
> Gerhard
>

I wasn't sure...goggle google...seems like its a corrosion issue?

http://www.ethanwiner.com/St-Synth.html
June 1979 issue of Recording-engineer/producer magazine

By the way, if you decide to use IC sockets, never use a gold plated
socket if the ICs have tin plated leads or vice versa. Many people don't
realize this, but over the years, corrosion can result from the
dissimilar metals if even the tiniest impurities (like sweat or other
moisture) gets on the mating surfaces.

http://www.dsp.dla.mil/sustainment/PEMs_best_practices.pdf

Sockets are unreliable due to several things. These areas include
micromotion
causing fretting corrosion. This is corrosion formed by
dissimilar metals at the board/socket/microcircuit interfaces and open
circuits due to insufficient contact pressure or from vibration and
shock. If the sockets have to stay on the board, request that there is a
common metal finish on the sockets contacts and the microcircuit
leads.

2006\07\02@155148 by Bob Axtell

face picon face
William Chops Westfield wrote:
> On Jul 2, 2006, at 6:45 AM, Bob Axtell wrote:
>
>  
>> All IR emitters degrade with usage. LED's degrade, too.  The
>> degradation is dependent on how hard the IR is driven. The
>> output drops about 15% over 5 years, again dependent on
>> how hard they are driven.
>>    
>
> Is that five years of ON time, or just 5 years?  If the former,
> which sounds more reasonable, I'd expect that many optoisolator
> applications have such a low duty cycle that 5 years of on time
> is essentially "forever."
>
> BillW
>  
5 years of ON time. Just doing nothing doesn't degrade the emitter.

--Bob

2006\07\02@163908 by Charles Craft

picon face
Lots of comments and suggestions but nobody asked the really interesting question.

What are you doing with PICs in the garden?
(I got a late start with my watermelons so am interested in automating their growth to increase yield) :-p


{Original Message removed}

2006\07\03@021653 by Vasile Surducan

face picon face
On 7/2/06, Peter Todd <EraseMEpetespamspamspamBeGonepetertodd.ca> wrote:
> On Sun, Jul 02, 2006 at 10:20:29AM +0300, Vasile Surducan wrote:
> > > If the boxes are not hermetically sealed, drill drain holes in the
> > > bottoms. I'm not kidding!!
> >
> >
> > Even the boxes are hermetically sealed, what make you think will not be
> > moisture inside ?  Huge temperature variations will create sweat
> > inside a hermetical sealed box as well in an opened one.
> > Critical temperature is 7-8C when the temperature is increasing from 0C.
>
> Not that I've ever done this sort of work but what's wrong with throwing
> a couple bags of dessicant in the boxes? It's aparently good enough to
> keep surface mount chips from going pop.

Sure, but as I said, the moisture is still there, now in the silica-gel pack.
I've just want to point to a problem which exist, no matter what
you're doing, except if some radical solutions are used, which are far
away too difficult and too expensive.

A hermetically sealed box with a PCB inside, running one day in open
sun could have even water inside in the next morning.

One example (nothing in common with the topic).
I've used a 250W 220V bulb as a flooding device for opening and closing a valve
(something more or less identical with the classical closet reservoir).
The water temperature insidethe tank  was between 10C and 90C. The
bulb works in this way one year. After one year inside the bulb I
found water, filling about 10% from the bulb volume. The bulb WORKS
after that when I supply it (with the bulb kept in a position where
water does not touch the filament. Conclusion: the porosity of the
bulb's glass allows bidirectional gas/fluid transfer. More or less on
this principle, the glass PH-meter
is built using a very porousive glass.

Why a hermetically sealed plastic box can't have the same behaviour?
:)

greetings,
Vasile

2006\07\03@022839 by Goran Hosinsky

picon face
Open and shut the electrical taps to water the different parts
of the garden. As for your water melons I really do not have
a tested solution but I read somewhere that cows milk better
if you play music for them and that they react very different
for different types of music. Perhaps you could try vibrating
the earth at different frequencies ...
Goran

Charles Craft wrote:
> Lots of comments and suggestions but nobody asked the really interesting question.
>
> What are you doing with PICs in the garden?
> (I got a late start with my watermelons so am interested in automating their growth to increase yield) :-p
>
>
> {Original Message removed}

2006\07\03@062107 by Peter Todd

picon face
On Mon, Jul 03, 2006 at 09:16:51AM +0300, Vasile Surducan wrote:
> > > Even the boxes are hermetically sealed, what make you think will not be
> > > moisture inside ?  Huge temperature variations will create sweat
> > > inside a hermetical sealed box as well in an opened one.
> > > Critical temperature is 7-8C when the temperature is increasing from 0C.
> >
> > Not that I've ever done this sort of work but what's wrong with throwing
> > a couple bags of dessicant in the boxes? It's aparently good enough to
> > keep surface mount chips from going pop.
>
> Sure, but as I said, the moisture is still there, now in the silica-gel pack.
> I've just want to point to a problem which exist, no matter what
> you're doing, except if some radical solutions are used, which are far
> away too difficult and too expensive.

True, and come to think of it, I wonder what it takes to get the
moisture *out* of a silica-gel pack... Could really screw up there.

> A hermetically sealed box with a PCB inside, running one day in open
> sun could have even water inside in the next morning.
>
> One example (nothing in common with the topic).
> I've used a 250W 220V bulb as a flooding device for opening and closing a valve
> (something more or less identical with the classical closet reservoir).

Flooding device? You mean, you used it for it's buoyancy, not for any
electrical properties?

> The water temperature insidethe tank  was between 10C and 90C. The
> bulb works in this way one year. After one year inside the bulb I
> found water, filling about 10% from the bulb volume. The bulb WORKS
> after that when I supply it (with the bulb kept in a position where
> water does not touch the filament. Conclusion: the porosity of the
> bulb's glass allows bidirectional gas/fluid transfer. More or less on
> this principle, the glass PH-meter
> is built using a very porousive glass.

What can I say, WOW! Actually I've been doing a fair bit of research
into making hermetically sealed containers as I'd like to eventually
make some vacuum tubes for fun. They always say glass (and some metals)
are the gold standard in lack of permiability, but even they let helium
though. They also didn't say exactly what types of glass to use...

I'd be interested to try that experiment again.

> Why a hermetically sealed plastic box can't have the same behaviour?
> :)

Well above mentioned research said over and over again that plastics are
generally useless if you want a good hermetic seal.

--
RemoveMEpeteKILLspamspampetertodd.ca http://www.petertodd.ca

2006\07\03@062729 by Peter Todd

picon face
On Sun, Jul 02, 2006 at 02:55:33PM -0400, Dave Lag wrote:
{Quote hidden}

Very interesting document!

So given that almost all ICs, especially in DIP format, are tin and or
lead finish where are all the gold-plated sockets that I see absolutely
everywhere in catalogs and surplus pins used?

--
peteSTOPspamspamspam_OUTpetertodd.ca http://www.petertodd.ca

2006\07\03@070157 by Alan B. Pearce

face picon face
>They always say glass (and some metals) are the gold
>standard in lack of permiability, but even they let
>helium though.

The problem here is that a helium molecule is the smallest molecule there
is, so it is going to find its way through the cracks and crevices in any
other molecule come what may. There is just no way to stop this, unless you
can work out how to make a solid vessel out of helium ...

2006\07\03@084913 by Gerhard Fiedler

picon face
Peter Todd wrote:

> So given that almost all ICs, especially in DIP format, are tin and or
> lead finish where are all the gold-plated sockets that I see absolutely
> everywhere in catalogs and surplus pins used?

Not in high reliability applications. I'm not sure, but my view of this is
that I usually don't use sockets for anything that's made to last. I use
sockets where I expect to exchange a component a lot. (To say the truth,
it's been a while; I don't have much use for sockets anymore. I may stick
something together quickly on a breadboard... but once I'm at the PCB
prototyping stage, things get usually soldered. Most components are SO or
below anyway.)

In such a scenario, I think that the gold-plated contacts have the
advantage of more insertion cycles. Tin-plated contacts wear out pretty
quickly. And since the components don't stay in for a long time, corrosion
is not that much of an issue.

Gerhard

2006\07\03@124914 by Peter

picon face

On Mon, 3 Jul 2006, Peter Todd wrote:

> True, and come to think of it, I wonder what it takes to get the
> moisture *out* of a silica-gel pack... Could really screw up there.

A power resistor, a lightbulb or sun heat.

Peter

2006\07\03@130050 by Bob Axtell

face picon face
Peter wrote:
> On Mon, 3 Jul 2006, Peter Todd wrote:
>
>  
>> True, and come to think of it, I wonder what it takes to get the
>> moisture *out* of a silica-gel pack... Could really screw up there.
>>    
>
> A power resistor, a lightbulb or sun heat.
>
> Peter
>
>  
Bake it in a low temp oven (180F) for 30 minutes.

--Bob

2006\07\03@143257 by Spehro Pefhany

picon face
At 06:41 AM 7/3/2006 -0400, you wrote:


>True, and come to think of it, I wonder what it takes to get the
>moisture *out* of a silica-gel pack... Could really screw up there.

One manufacturer (below) recommends a vented oven at approximately 300°F for
at least three hours (sometimes more time is required, they say)

http://www.dehumidify.com/HydrosorbentReactivation.htm

It's also possible to use a microwave oven.

You can buy silica gel crystals relatively cheaply in quantity at crafts
stores.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spamBeGonespeffSTOPspamspamEraseMEinterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->>Test equipment, parts OLED displys http://search.ebay.com/_W0QQsassZspeff


2006\07\03@185942 by olin piclist

face picon face
Vasile Surducan wrote:
> Conclusion: the porosity of the
> bulb's glass allows bidirectional gas/fluid transfer.

How do you know it's not the seals?  That sounds more likely to me, although
I don't have any direct knowledge one way or the other.

Amusing side story:  A group of us in high school did a bunch of pranks.
One of them was to take a few bulbs from a particular room, and carefully
file a tiny hole in the glass close to where it meets the screw base.  We
then carefully heated the bulbs, then dunked the end in water.  As the air
inside shrunk as it cooled, some water was sucked in.  We then put the bulbs
back, which hung from the ceiling with the glass down and screw base up.
The bulbs worked fine for a while, and eventually someone noticed the water.
We floated the "explanation" that there had been a flood, and surprisingly
people were a bit dumbfounded but then accepted the explanation.  Nobody
stopped to think about it, since the high school was on a hill and this room
was on the fourth floor.  Eventually the bulbs we modified burnt out, much
quicker than normal and everyone apparently forgot all about it when they
were replaced.  Moral of the story: People are stupid and are easily led
around like sheep with a little imagination.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products

2006\07\03@190142 by olin piclist

face picon face
Goran Hosinsky wrote:
> As for your water melons I really do not have
> a tested solution but I read somewhere that cows milk better
> if you play music for them and that they react very different
> for different types of music. Perhaps you could try vibrating
> the earth at different frequencies ...

And waving a dead fish over them during a full moon.  When you're done
dispose of the smelly fish by burying it by the plant.  It really works, try
it.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products

2006\07\04@142346 by Peter

picon face

On Mon, 3 Jul 2006, Bob Axtell wrote:

> Peter wrote:
>> On Mon, 3 Jul 2006, Peter Todd wrote:
>>
>>> True, and come to think of it, I wonder what it takes to get the
>>> moisture *out* of a silica-gel pack... Could really screw up there.
>>
>> A power resistor, a lightbulb or sun heat.
>>
>> Peter
> Bake it in a low temp oven (180F) for 30 minutes.

If the box gets daily sun exposure the problem is fixed.

Peter

2006\07\04@142725 by Peter

picon face

Speaking of silica gel, I once bought a kilo bag of a desiccant product
hoping to cure a mildew problem in a cupboard. The substance came in a
troughlike plastic tray (think cake loaf form). I expected it to swell
or something. Instead it turned entirely liquid (with water from the air
I presume). That was unexpected. Does anyone know what this substance is
? It was a Swedish product, and originally came in whitish crystals
about 2mm each.

Peter

2006\07\04@170241 by Gerhard Fiedler

picon face
Peter wrote:

> Speaking of silica gel, I once bought a kilo bag of a desiccant product
> hoping to cure a mildew problem in a cupboard. The substance came in a
> troughlike plastic tray (think cake loaf form). I expected it to swell
> or something. Instead it turned entirely liquid (with water from the air
> I presume). That was unexpected. Does anyone know what this substance is
> ? It was a Swedish product, and originally came in whitish crystals
> about 2mm each.

I think it's a hygroscopic salt. Not sure which one.

Gerhard

2006\07\04@180105 by William Chops Westfield

face picon face

On Jul 4, 2006, at 2:01 PM, Gerhard Fiedler wrote:

>> I once bought a kilo bag of a desiccant product hoping to
>> cure a mildew problem in a cupboard. ...it turned entirely
>> liquid (with water from the air I presume). That was
>> unexpected. Does anyone know what this substance is?
>
> I think it's a hygroscopic salt. Not sure which one.

Sounds like Calcium Chloride, marketed under a variety of
names.

BillW

2006\07\04@182357 by William Chops Westfield

face picon face

On Jul 2, 2006, at 11:55 AM, Dave Lag wrote:

> if you decide to use IC sockets, never use a gold plated socket
> if the ICs have tin plated leads or vice versa. Many people
> don't realize this, but over the years, corrosion can result
> from the dissimilar metals if even the tiniest impurities
> (like sweat or other moisture) gets on the mating surfaces.

I heard the "don't use gold with tin" argument WRT SIM sockets,
where it was described as a mechanical issue; wiping vs piercing.

I wouldn't think that the "dissimilar metal" issue would apply
much to "noble metals"; might as well complain that your solder
contains both tin and lead, or that the IC leads are tin/lead
plated while the sockets are just tin plated.

I don't trust mil specs all that much; some of it was probably
one man's opinion that got written down and became LAW, without
much data to support the original opinion.  It was quite some time
ago that I heard a story that the test regime that mil-spec INSISTED
on applying to some form of RAM put such a stress on the parts that
the resulting "passed" chips had easily measured LESS reliability
than OTS consumer chips of similar size...

Or put another way, if Augat, milmax, and other socket vendors say
that it's worthwhile to have gold plated sockets for my ICs, then
I'm more inclined to believe them than some other spec...

BillW

2006\07\05@125744 by Peter

picon face


On Tue, 4 Jul 2006, William Chops Westfield wrote:

{Quote hidden}

Thanks for that. Would CaCl2 turn entirely liquid with just moisture
from the air ? What would be a household test for CaCl2 ?

http://en.wikipedia.org/wiki/Calcium_chloride


Peter

2006\07\05@151410 by Gerhard Fiedler

picon face
Peter wrote:

>>>> I once bought a kilo bag of a desiccant product hoping to cure a
>>>> mildew problem in a cupboard. ...it turned entirely liquid (with
>>>> water from the air I presume). That was unexpected. Does anyone know
>>>> what this substance is?
>>>
>>> I think it's a hygroscopic salt. Not sure which one.
>>
>> Sounds like Calcium Chloride, marketed under a variety of names.
>
> Thanks for that. Would CaCl2 turn entirely liquid with just moisture
> from the air ?

I think so. They use that here for places where you store clothes and
stuff, in small plastic containers with a membrane on top. You buy them
sealed with the dry salt inside, and after removing the seal, the container
fills slowly with water until there seems to be no salt left. I haven't
opened one yet and can't tell you whether it all dissolves, or just most of
it. Anyway, about the same volume the salt had in the container is later
filled with liquid.

Gerhard

2006\07\05@170023 by Wouter van Ooijen

face picon face
> Thanks for that. Would CaCl2 turn entirely liquid with just moisture
> from the air ?

AFAIK yes

> What would be a household test for CaCl2 ?

Test Ca2+ : dissolve a little in water, bubble your breath trough it.
The solution should become opaque white.

Testing Cl- is more difficult, silvernitrate would be suitable but not
exactly a household chemical. You could rub some solution on your hands.
If it feels soapy the negative ion is a weak acid, Cl- is strong. If it
does not feel soapy the most likely negative ions are Cl- and NO3-. I
don't think Ca(NO3)2 is hygroscopic, at least not as much as CaCl2.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\07\05@173820 by Carey Fisher

face picon face
SOCKET 2ME !!!!!!

Sorry...couldn't resist...one of those moods...

Google "Laugh-in"

2006\07\06@143204 by Vasile Surducan

face picon face
On 7/3/06, Peter Todd <KILLspampetespamBeGonespampetertodd.ca> wrote:
{Quote hidden}

I've tried it two consecutive years on the same reservoir but I never
reproduce it again ! It was a defective bulb for sure.

Vasile

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...