AVR-Chat

--- In AVR-Chat@yahoogroups.com, "Steven Hodge" <stevehodge@...> wrote:

> ... the Mouser catalog ... it's driving me nuts.

Don't rely on specifications in the Mouser catalog.  Obtain and read 
the manufacturer's data sheets.  Quite often there are transcription 
errors in Mouser's data.  Also, Mouser tries to force specifications 
into a pattern to drive their on-line part selection tool (which thinks 
that 200 mA, 0.2 A and 0.200 A are diffent ratings).

On the other hand, unless you are switching an inductive load, yes, a 
switch rated at 300 mA AC will be able to handle 200 mA DC.

The reason that a switch might be able to handle more AC current than 
DC is that AC drops to zero every half cycle, which would quench any 
arc across the terminals.  This doesn't matter for a non-inductive load.

Graham.

I may have posed this question some time ago, but I'm still confused.  I'm
trying to determine if a pcb slide switch will meet my current requirements,
which in this particular case are 200 mA at 12 VDC.   So I'm looking at a
page (1646) of the Mouser catalog, for such switches all from the same
manufacturer (Alcoswitch) and I see one series that says:

"gold, 0.4 VA; silver 300 mA @ 115 VAC"

So if I divide 0.4 VA by 12 V I get 33 mA.  How come this seems so
drastically less than 300 mA at 115 VAC (300 mA x 115 V = 34.5 VA)?  I know
there is an AC/DC thing here but the difference in this "power" is almost
two orders of magnitude!

Another series is similar:

"gold, 0.4 VA @ 20 VDC; silver 300 mA @ 115 VAC"

Why was the "@ 20 VDC" added here?  How do I convert the rating to my 12 VDC
level?

Then another series says only:

"300 mA @ 125 VAC"

Does that mean the switch is not usable with DC?

These seem like incredibly stupid and elementary questions to be posing
here, but it's driving me nuts.  Can somebody help me out, please?  My
intuition tells me a switch that can take 300 mA at 115 VAC ought to be able
to handle 200 mA at 12 VDC.

Thanks, Steve


 

__________ Information from ESET Smart Security, version of virus signature
database 3842 (20090210) __________

The message was checked by ESET Smart Security.

http://www.eset.com

What they are talking about there, is the difference between an
reactive load and a resistive load.

http://en.wikipedia.org/wiki/Volt-ampere

Reactive loads can cause a lot more wear on a switch during make or
break  than resistive loads.

> What they are talking about there, is the difference between an
> reactive load and a resistive load.
>
> http://en.wikipedia.org/wiki/Volt-ampere
>
> Reactive loads can cause a lot more wear on a switch during make or
> break  than resistive loads.
>

There are other issues than just reactive load. A switch that passes DC
arcs differently than on that is handling AC. This tends to be more of a
problem at higher voltages (say, above 60V or so).

Also, there is the old "dry circuit" problem. If there is not ENOUGH
current, oxide films on the switch contacts won't be broken down and the
switch will be erratic.

At 12V DC, and several hundred milliamperes, almost any switch should be
OK except the very lightest dry circuit ones and the very heaviest ones.
Slide switches should be fine. Those ratings are MAXIMUM ratings.

Jim Wagner

On Tue, Feb 10, 2009 at 11:33 AM, Graham Davies
<Yahoo37849@ecrostech.com> wrote:
>  Also, Mouser tries to force specifications
> into a pattern to drive their on-line part selection tool (which thinks
> that 200 mA, 0.2 A and 0.200 A are diffent ratings).

Digikey has similar problems.. But things are cleaning up..

> The reason that a switch might be able to handle more AC current than
> DC is that AC drops to zero every half cycle, which would quench any
> arc across the terminals.  This doesn't matter for a non-inductive load.

Similarly with fuses, and a 32VDC fuse is NOT guaranteed to be able to
open the circuit in a 120VAC application!

On Tue, Feb 10, 2009 at 08:13:54AM -0800, Steven Hodge wrote:
> I may have posed this question some time ago, but I'm still confused.
> I'm trying to determine if a pcb slide switch will meet my current
> requirements, which in this particular case are 200 mA at 12 VDC.   So
> I'm looking at a page (1646) of the Mouser catalog, for such switches
> all from the same manufacturer (Alcoswitch) and I see one series that
> says:
> 
> "gold, 0.4 VA; silver 300 mA @ 115 VAC"
> 
> So if I divide 0.4 VA by 12 V I get 33 mA.  How come this seems so
> drastically less than 300 mA at 115 VAC (300 mA x 115 V = 34.5 VA)?  I
> know there is an AC/DC thing here but the difference in this "power"
> is almost two orders of magnitude!

Is because "there is an AC/DC thing here." In general a switch can not
handle as much DC power as AC power.

> Another series is similar:
> 
> "gold, 0.4 VA @ 20 VDC; silver 300 mA @ 115 VAC"
> 
> Why was the "@ 20 VDC" added here?  How do I convert the rating to my
> 12 VDC level?

They are saying one should not use more than 20 VDC or nominal 115 VAC
with that switch. That with DC one should stay under 0.4 VA. They don't
say the same thing about AC because one is either going to use 100, 115,
or 220. That rating at 115 is close enough for the oddball places in
Japan running 100 VAC.

> Then another series says only:
> 
> "300 mA @ 125 VAC"
> 
> Does that mean the switch is not usable with DC?

Means it is not rated for DC.

You need stouter switches for your 200 mA.

-- 
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.

--- In AVR-Chat@yahoogroups.com, "Steven Hodge" <stevehodge@...> 
wrote:

> ... the actual data sheet ... says
> ... "gold: 0.4 VA max @ 20 VDC or
> peak AC".   Does that seriously mean
> the switch can only handle a pathetic
> 33 mA at 12 VDC (or 20 mA at 20 VDC)?

Once you're reading the data sheet, your beef is now with the 
manufacturer.  Either use another switch, or ask the manufacturer to 
clarify.  My guess is that "VA" is a misprint and the switch can 
handle 0.4 amps.  But, don't act on my guess.

> In my case the switch is input to a
> regulator which ... has the usual
> input capacitor on it.  Does this
> qualify as a resistive load?  It
> would seem to me that it does.

How can a capacitor qualify as a resistive load?  But, it isn't an 
inductive load, which gets rid of the biggest problem.  With a large 
capacitive load, the weird stuff happens when the switch makes and 
the break action is benign.  You'll get a large surge current on the 
make.  My opinion is that for your microcontroller application you 
can ignore this.  All other things being equal, just choose a switch 
with a slightly higher current rating than you need.

I don't think anyone is in disagreement here.  Someone said that if 
the manufacturer chooses not to give a switch a DC rating you should 
not use it in a DC application.  That's a fairly strict position to 
take and I would be a bit more relaxed myself, but it is the 
technically most defensible thing to do.

Graham.

Well now I'm even more confused.   If VA is a measure of reactive/apparent
power, then why is there a separate rating for "VAC" (which looks like a
reactive rating to me)?   In other words, where is the resistive/real power
rating?   

 

It is definitely correct about looking at the actual data sheet.
Unfortunately, that only added to my confusion.  All it says is "gold: 0.4
VA max @ 20 VDC or peak AC".   Does that seriously mean the switch can only
handle a pathetic 33 mA at 12 VDC (or 20 mA at 20 VDC)?  It also seems that
spec is confusing the accepted use of "VA" versus "watts".

 

In my case the switch is input to a regulator which is supplying DC voltage
to a microcontroller and its peripheral chips.   The regulator has the usual
input capacitor on it.  Does this qualify as a resistive load?  It would
seem to me that it does.

 

And judging from the other responses that came in after this one, I don't
seem to be the only one confused.   One of you says the switch is just fine,
but another says no way!

 

Steve

From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf
Of David VanHorn
Sent: Tuesday, February 10, 2009 8:19 AM
To: AVR-Chat@yahoogroups.com
Subject: Re: [AVR-Chat] switch rating

 

What they are talking about there, is the difference between an
reactive load and a resistive load.

http://en.wikipedia.org/wiki/Volt-ampere

Reactive loads can cause a lot more wear on a switch during make or
break than resistive loads.



[Non-text portions of this message have been removed]

> In my case the switch is input to a regulator which is supplying DC voltage
> to a microcontroller and its peripheral chips.   The regulator has the usual
> input capacitor on it.  Does this qualify as a resistive load?  It would
> seem to me that it does.

definitely not!

Calculate your inrush current, assuming the cap is at 0V, and using
its ESR as it's impedance, assuming that closure happens at the peak
input voltage, with the transformer DC resistance in series with it as
a voltage source.  (Worst-casing it)


The first few uS will greatly exceed the switch ratings.

In practice will this matter?  Maybe not, but these effects are real,
and should be looked at when designing a system.
The inrush current may pit the switch contacts and cause premature
failure after a few thousand cycles.

Also, you might look at your diode current ratings vs the size of the
regulator filter capacitor.
When the cap is mostly charged, and the system is under full load, the
diodes only conduct part of the time. (conduction angle)
This means that the peak current is much higher than the average, if
too large a filter cap is used!
You can blow your diodes if their repetitive peak rating isn't up to
what is actually happening.

In the hobby world, a given solution "works", if you have a loose
enough definition of "works".
:)

Yeah, I stupidly blundered into that one on the capacitor, all right.   But,
I'll bite.   I don't have any diodes.   The power source is not an AC power
supply but a 12 VDC battery (a large 700 amp-hr boat system) and I don't
bother with reverse polarity protection (I prefer being careful to suffering
the voltage drop).  It's just battery to fuse to switch to regulator, with
cap on the input to regulator.    Would a current-limiting resistor in
series with the cap be a good idea to help protect the switch?   Steve

From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf
Of David VanHorn
Sent: Tuesday, February 10, 2009 10:18 AM
To: AVR-Chat@yahoogroups.com
Subject: Re: [AVR-Chat] switch rating

 

> In my case the switch is input to a regulator which is supplying DC
voltage
> to a microcontroller and its peripheral chips. The regulator has the usual
> input capacitor on it. Does this qualify as a resistive load? It would
> seem to me that it does.

definitely not!

Calculate your inrush current, assuming the cap is at 0V, and using
its ESR as it's impedance, assuming that closure happens at the peak
input voltage, with the transformer DC resistance in series with it as
a voltage source. (Worst-casing it)

The first few uS will greatly exceed the switch ratings.

In practice will this matter? Maybe not, but these effects are real,
and should be looked at when designing a system.
The inrush current may pit the switch contacts and cause premature
failure after a few thousand cycles.

Also, you might look at your diode current ratings vs the size of the
regulator filter capacitor.
When the cap is mostly charged, and the system is under full load, the
diodes only conduct part of the time. (conduction angle)
This means that the peak current is much higher than the average, if
too large a filter cap is used!
You can blow your diodes if their repetitive peak rating isn't up to
what is actually happening.

In the hobby world, a given solution "works", if you have a loose
enough definition of "works".
:)





__________ Information from ESET Smart Security, version of virus signature
database 3843 (20090210) __________

The message was checked by ESET Smart Security.

http://www.eset.com



[Non-text portions of this message have been removed]

On Tue, Feb 10, 2009 at 1:44 PM, Steven Hodge <stevehodge@speakeasy.net> wrote:
> Yeah, I stupidly blundered into that one on the capacitor, all right.   But,
> I'll bite.   I don't have any diodes.   The power source is not an AC power
> supply but a 12 VDC battery (a large 700 amp-hr boat system) and I don't
> bother with reverse polarity protection (I prefer being careful to suffering
> the voltage drop).  It's just battery to fuse to switch to regulator, with
> cap on the input to regulator.    Would a current-limiting resistor in
> series with the cap be a good idea to help protect the switch?   Steve

It would help.

Running "naked" on a vehicular bus has its own set of problems.
Alternator load dump transient, etc can kill you really easily.
They do make regulators designed to cope with that, check for
"automotive" types.

Another fun failure mode that most three-terminal regs have, is when
the input is shorted to ground.
If the output has any significant capacitance, this can kill the reg.
It's documented in the older motorola sheets on the 340 series and
nationals on the 78XX series, but most three terminal regs have this
weakness.  A diode connected from input to output, such that it is
normally reverse-biased, solves the problem.