>>> I was going to use a 1N914 as the back EMF diode.
>>
>> We have a design where an 18F is directly driving a reed relay, with
>> a
>> 1N4148 across the coil. Devices have been in use for two years.
>>
>> I'm curious, what was your reasoning for going with a reed relay, as
>> opposed to a solid state device?
>>
>> Vitaliy
>
>
> Some reed relays even include the clamp diode inside the relay! A reed

> is good where you need to do floating dry switching. Doing that solid
> state can be expensive...

> Relay contacts are not wetted in any way, such as mercury.
>
> -----Original Message-----
> From: RemoveMEpiclist-bouncesTakeThisOuTmit.edu [spamBeGonepiclist-bouncesspamBeGonemit.edu] On Behalf
> Of Vitaliy
> Sent: Monday, January 12, 2009 5:23 PM
> To: Microcontroller discussion list - Public.
> Subject: Re: [EE] Reed relays from PIC port pin
>
> PAUL James wrote:
>>> What's "floating dry switching"? :)
>> Switching that isn't referenced to ground.
>
> Why is it "dry"?

> At 04:22 PM 1/12/2009, Vitaliy wrote:
>
>>PAUL James wrote:
>> >> What's "floating dry switching"? :)
>> > Switching that isn't referenced to ground.
>>
>>Why is it "dry"?
>
> "Wet" and "Dry" are old telephone-industry terms.
>
> "Wet" means that there is (DC) voltage present.  I think that it used
> to mean the DC supply and load of a telephone handset (48~54Vdc,
> 20mA~100mA) but is now generally referred to as pretty much any
> supply and load.
>
> "Dry" means that the (relay or switch) contacts are isolated and not
> connected to any power supply inside the device where the relay is
> located.
>
> This may seem not obvious, but any contact that is referred to as
> "Dry" is usually a low-current-capable contact and not normally
> suitable for controlling heavy loads directly.  Do note that I said
> "usually".  Low-current-capable means that the contact is gold-plated
> or otherwise capable of handing low current signals (does not require
> wetting current).
>
> dwayne

>
>
> I think that pretty much summarizes it. If you try to switch low current
> loads with relays that are designed for power loads, the results will not
> be reliable. I learned this years ago when I made a "Yodeling Clock" for
> the Madonna Inn in San Luis Obispo, CA. There were two volume controls,
> one for day, one for night. A clock drove a relay that switched between
> the two. The relay I used was rated for something like 1A at 120VAC. Half
> the time there was no audio. A relay rated for dry switching would have
> handled it fine. These often are "bifrucated gold cross point" contacts.
> There are two contacts in parallel, and each is a gold wedge that make
> contact with another gold wedge at 90 degrees so the contact pressure is
> high.
>
> Back on wet and dry, telephone cable splices are quite often unreliable
> unless a DC "sealing current" is run through them.
>
> Harold
>

>Just curious (and because I have a background project to identify information
>transfer issues in engineering education) - how many folks were taught this
>(below) in their degree program, either in the lab (taught in the lab), or
>formal class, or in theory resulting in this bit of knowledge? Also,
>did anyone
>stumble across it during labs in trying to get something to work and got
>'experience' this way instead (not formally taught but needed the
>knowledge to
>complete the course)?
>
>-Skip
>
>Harold Hallikainen wrote:
> >
> > I think that pretty much summarizes it. If you try to switch low current
> > loads with relays that are designed for power loads, the results will not
> > be reliable. I learned this years ago when I made a "Yodeling Clock" for
> > the Madonna Inn in San Luis Obispo, CA. There were two volume controls,
> > one for day, one for night. A clock drove a relay that switched between
> > the two. The relay I used was rated for something like 1A at 120VAC. Half
> > the time there was no audio. A relay rated for dry switching would have
> > handled it fine. These often are "bifrucated gold cross point" contacts.
> > There are two contacts in parallel, and each is a gold wedge that make
> > contact with another gold wedge at 90 degrees so the contact pressure is
> > high.
> >
> > Back on wet and dry, telephone cable splices are quite often unreliable
> > unless a DC "sealing current" is run through them.
> >
> > Harold