Wetting current

[Klaus Vestergaard Kragelund]

Hi

I am working on a design where we are monitoring a relay, if it has been
energized. That is done with high ohmic resistor and a source voltage,
detecting a current when drawn

But, contact wetting could be a problem, so have added a small capacitor
on the contacts, so when they are closed a relatively high current runs
for a short period of time to punch through the oxide layer of the contacts

I am having difficulty finding any information on the subject, how much
current is needed for punch through and how much time.

I need a general input, since we do not know if the relay has gold
plated contacts or not

Anyone got a clue?

Regards

Klaus

[Joe Gwinn]

There are books written on electrical contacts and their design for
various conditions. Unless the current is very small gold tends to
weld. Rhodium is better in such cases.

Joe Gwinn

[John Larkin]

On Fri, 18 Jun 2021 00:56:55 +0200, Klaus Vestergaard Kragelund
<klau...@hotmail.com> wrote:

We have a similar issue now. We will sell a general-purpose relay
module as part of a modular test system, and we are concerned that
once a relay is used to switch power, it may not be good for signals
any more.

That may be urban legend. It would be awkward to test.

[legg]

On Fri, 18 Jun 2021 00:56:55 +0200, Klaus Vestergaard Kragelund
<klau...@hotmail.com> wrote:

Aren't you actually interested in the existence of practical contact?

Any voltage detected indicates no contact, ie failure, even if it's
supposed to be closed. Ensuring contact shouldn't be the job of the
contact monitor - it's the job of the application designer.

Your measuring method shouldn't degrade the degree of isolation
expected when the contacts are open, however. So high z sensor
with lots of common-mode range.

RL

[piglet]

On 17/06/2021 23:56, Klaus Vestergaard Kragelund wrote:
>
> But, contact wetting could be a problem, so have added a small capacitor
> on the contacts, so when they are closed a relatively high current runs
> for a short period of time to punch through the oxide layer of the contacts
>

Do not wire the capacitor directly across the contact - add a series
resistor else the peak current could weld delicate contacts. Even tens
of pF capacitance can weld reeds. 10-50mA peak should be safe.

Too small a test current could also be problematic. Keep the DC current
above 100uA. also too low an open circuit voltage can be insufficient to
break down oxides, should be above 3-5V. Min switching of 5V 100uA are
figures often quoted in switch/relay datasheets.

Old time telegraph/telephone books from 19th cent-early 20th often had
information on wetting as folk were just discovering this stuff then.

piglet

[piglet]

Definitely not an urban legend. I have read a warning about gold plated
contacts in some manufacturer's datasheet, possibly Tyco?

piglet

[Steve Wilson]

I wonder if mercury-wetted contacts are still available?

Are reed relays available in nitrogen-filled enclodures? This might help
avoid contamination and oxidation of the contacts.

--
The best designs occur in the theta state. - sw

[Clive Arthur]

I'm pretty sure all reeds are either inert gas filled or for HV types,
evacuated, so it would be a good solution.

--
Cheers
Clive

[jla...@highlandsniptechnology.com]

On Fri, 18 Jun 2021 06:38:55 +0100, piglet <erichp...@hotmail.com>
wrote:

Got a link to a data sheet or appnote? It would be nice to find a
general-purpose relay contact material. Maybe not down to
thermocouples, but amps to mA maybe.



--

John Larkin Highland Technology, Inc

The best designs are necessarily accidental.

[jla...@highlandsniptechnology.com]

On Fri, 18 Jun 2021 06:36:05 +0100, piglet <erichp...@hotmail.com>
wrote:

>On 17/06/2021 23:56, Klaus Vestergaard Kragelund wrote:
>>
>> But, contact wetting could be a problem, so have added a small capacitor
>> on the contacts, so when they are closed a relatively high current runs
>> for a short period of time to punch through the oxide layer of the contacts
>>
>
>Do not wire the capacitor directly across the contact - add a series
>resistor else the peak current could weld delicate contacts. Even tens
>of pF capacitance can weld reeds. 10-50mA peak should be safe.

Reed relays can weld themselves with zero contact current. They are
expensive and flakey.

>
>Too small a test current could also be problematic. Keep the DC current
>above 100uA. also too low an open circuit voltage can be insufficient to
>break down oxides, should be above 3-5V. Min switching of 5V 100uA are
>figures often quoted in switch/relay datasheets.
>
>Old time telegraph/telephone books from 19th cent-early 20th often had
>information on wetting as folk were just discovering this stuff then.
>
>piglet

[Phil Hobbs]

ISTR that somebody makes gold-over-rhodium contacts, so that they're
good for dry circuits as well as power.

Of course as you say, once you blow the gold off it's not a dry-contact
relay anymore.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

[John Walliker]

Pickering make a wide range of reed relays and provide a lot of advice on how to choose and use them.

www.pickeringrelay.com

John

[John Larkin]

They made an RTD simulator with a heap of reed relays and resistors.
It was a horror. We do it electronically, which avoids all sorts of
interesting effects.

[Dimiter_Popoff]

My experience is not very extensive but does cover things close to both
extremes you seem to be looking at.
In the netMCA-3 signals (gain, switches opamp stages/feedbacks, fairly
wideband so the currents are not unthinkably low) get switched by
latching DPDT relays (no longer in production and alas, our stocks are
getting low - replaceable but no drop-in) and these work for years
without causing any issues at 14 bit overall precision.
At the other extreme I have relays switching on/off ventilator heaters
here so they maintain a constant temp. About 2kW each, 220V AC.
The relays last for 1-2 winters and then begin to stays stuck,
not too badly welded as a kick to the housing unwelds them for some
time (usually has to be done during the night when you wake up
and notice it...).
I'd definitely not trust these relays for signal switching even
after a very short use with the ovens. (Obviously the relay types are
quite different, the power ones have been different over time,
same behaviour though).

Dimiter

======================================================
Dimiter Popoff, TGI http://www.tgi-sci.com
======================================================
http://www.flickr.com/photos/didi_tgi/

[Cydrome Leader]

What's some weirdness you see with reed relay logic?

[Piglet]

On 18/06/2021 15:31, jla...@highlandsniptechnology.com wrote:
>
> Got a link to a data sheet or appnote? It would be nice to find a
> general-purpose relay contact material. Maybe not down to
> thermocouples, but amps to mA maybe.
>
>
>

I think I saw it SDS Relais-now Panasonic or Omron but am away from my
bookshelf. I think it is also in AoE X-chapters.

See page 4 of this from Phoenix Contact:

<https://www.phoenixcontact.com/assets/downloads_ed/global/web_dwl_technical_info/105396_en_00.pdf>

piglet

[Piglet]

On 17/06/2021 23:56, Klaus Vestergaard Kragelund wrote:

This Axicom databook looks helpful on different contact material
properties (circa pages 13-20):

<http://www.farnell.com/datasheets/105262.pdf>

piglet

[John Larkin]

In the RTD case, horrible transient values when a value is changed.
like spiking to 500 deg C for some milliseconds, going from 48 to 49.
Engine control computers don't like that.

Reeds have huge thermoelectrics too. Lots of intermetallic splices
with a heater roughly in the middle.

I discovered reed twang some years ago. The reed wiggles like a struck
bell when it's energized, very complex waveforms lasting many
milliseconds. In the coil mag field, that makes biggish voltages.

[Tom Del Rosso]

John Larkin wrote:
>
> We have a similar issue now. We will sell a general-purpose relay
> module as part of a modular test system, and we are concerned that
> once a relay is used to switch power, it may not be good for signals
> any more.

Just make them all the same and spec them differently, with different
markings on the terminals.

You could even make a whole assortment of models, with half the channels
spec'd for power, or 25% or 10%.


--
Defund the Thought Police

[Klaus Vestergaard Kragelund]

I seen many examples listed, but no real accurate numbers

[Klaus Vestergaard Kragelund]

Nice table, but are the currents representative of wetting keepouts

[Klaus Vestergaard Kragelund]

Any idea if the current could be ac?

I have the possibility to inject ac current, with more than 1mA, a lot
easier than DC current

[Cydrome Leader]

you can hear them ring mechanically too when you pull a magnet away from
them.

maybe you need some lovely mercury wetted relays.

[John Larkin]

On Sat, 19 Jun 2021 21:21:56 +0000 (UTC), Cydrome Leader

We use lots of Fujitsu FTR-B3G relays... almost 100K so far.

DPDT, surface mount, small and reliable. We use the latching version
at thermocouple levels. They are good up to 3 GHz too.

Reeds are expensive junk.

--

John Larkin Highland Technology, Inc trk

The cork popped merrily, and Lord Peter rose to his feet.
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"

[Bill Sloman]

On Sunday, June 20, 2021 at 11:26:16 AM UTC+10, John Larkin wrote:
> On Sat, 19 Jun 2021 21:21:56 +0000 (UTC), Cydrome Leader
> <pres...@MUNGEpanix.com> wrote:
>
> >John Larkin <jlarkin@highland_atwork_technology.com> wrote:
> >> On Fri, 18 Jun 2021 19:08:37 +0000 (UTC), Cydrome Leader
> >> <pres...@MUNGEpanix.com> wrote:
> >>
> >>>John Larkin <jlarkin@highland_atwork_technology.com> wrote:
> >>>> On Fri, 18 Jun 2021 09:18:19 -0700 (PDT), John Walliker
> >>>> <jrwal...@gmail.com> wrote:

<>snip>

> >maybe you need some lovely mercury wetted relays.

They are nice but you do have to mount them so they stray vertical. Orientation-insensitive mercury wetted reeds did get touted from time to time, but never lasted.

> We use lots of Fujitsu FTR-B3G relays... almost 100K so far.
>
> DPDT, surface mount, small and reliable. We use the latching version at thermocouple levels. They are good up to 3 GHz too.
>
> Reeds are expensive junk.

Of course Bell Labs develped them because they needed something more reliable than traditional relays for the telephone network.

They were invented in Russia in 1922, but Bell Labs developed them into something useful in the 1930s - there's a 1941 US patent.

If John Larkin doesn't like them, it probably means that he didn't use them as carefully as he should have.

https://en.wikipedia.org/wiki/Reed_switch

--
Bill Sloman, Sydney

[Piglet]

If your open circuit ac voltage is in the 5-15V range I guess that
should work. The worst switch contact oxidation film I saw took about
12V before reliable contact was made. Since you don't know what contacts
your circuit may be exposed to I would steer on the high side (10-15V?).
I think once the current is above a few hundred micro amps you should be OK.

Is there a reason why you cannot see the relay coil voltage instead?

piglet

[Spehro Pefhany]

On Fri, 18 Jun 2021 00:56:55 +0200, Klaus Vestergaard Kragelund
<klau...@hotmail.com> wrote:

>Hi
>
>I am working on a design where we are monitoring a relay, if it has been
>energized. That is done with high ohmic resistor and a source voltage,
>detecting a current when drawn
>

>But, contact wetting could be a problem, so have added a small capacitor
>on the contacts, so when they are closed a relatively high current runs
>for a short period of time to punch through the oxide layer of the contacts
>

>I am having difficulty finding any information on the subject, how much
>current is needed for punch through and how much time.
>
>I need a general input, since we do not know if the relay has gold
>plated contacts or not
>
>Anyone got a clue?
>
>Regards
>
>Klaus

For a reliable bulletproof general solution, I think it would be best
to use something like 24V/10mA and pulse the current for the
measurement to minimize the wasted power.

You can probably get away with less..

--
Best regards,
Spehro Pefhany

[Piotr Wyderski]

John Larkin wrote:

> In the RTD case, horrible transient values when a value is changed.
> like spiking to 500 deg C for some milliseconds, going from 48 to 49.
> Engine control computers don't like that.

No Gray-or-compatible encoding? That's too bad.

Best regards, Piotr

[Klaus Vestergaard Kragelund]

Yes, the relay is in another product, so we run wires from our box to
the relay contacts, to detect the state of the external product

Cheers

Klaus

[Klaus Vestergaard Kragelund]

On 20/06/2021 12.12, Piglet wrote:

The solution we have now is a capacitor charged to a small voltage. Then
when the relay closes, high current runs to discharge said capacitor

In this case the voltage is low, but peak current is high (for less than
100ns)

Cheers

Klaus

[John Larkin]

It's hard to do that with relays switching resistors, especially
relays that bounce a lot. I guess it could be done, but it would take
too many relays and resistors.

We do resistor simulation electronically. That can be kept monotonic.



--

[David Lesher]

You might look into the BSTJ archives. I recall chatting
20 years ago with someone from the Labs who had studied it
intensively.

A shortcut to them is <http://etler.com/docs/BSTJ/archive.html>
--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

[Cydrome Leader]

Well it's not 1930 anymore and the applications for reed switches are few
and far in between. The wiki article even mentioned reed switches being
used as sensors in old brushless motors. Awful.

[Clifford Heath]

Bosch are still using reed switches to make their dishwashers unreliable

So much for "German engineering"

CH

[Cydrome Leader]

I've never buy anything from bosch but a few power tools.

european appliances are just pure trash when it comes to design and
repairability. Oh, the cars too.

[whit3rd]

On Friday, June 18, 2021 at 2:11:47 PM UTC-7, John Larkin wrote:
> On Fri, 18 Jun 2021 19:08:37 +0000 (UTC), Cydrome Leader
> <pres...@MUNGEpanix.com> wrote:

> >What's some weirdness you see with reed relay logic?

>... I discovered reed twang some years ago. The reed wiggles like a struck

> bell when it's energized, very complex waveforms lasting many
> milliseconds. In the coil mag field, that makes biggish voltages.

That's why one 'powers' a relay coil, instead of signalling it.
The motion of the contacts means the device is an electric motor,
with all the back-EMF you expect, generated by moving
magnetic parts in the field of the coil.

[whit3rd]

On Saturday, June 19, 2021 at 6:26:16 PM UTC-7, John Larkin wrote:

> We use lots of Fujitsu FTR-B3G relays... almost 100K so far.
>
> DPDT, surface mount, small and reliable. We use the latching version
> at thermocouple levels. They are good up to 3 GHz too.
>
> Reeds are expensive junk.

Do your FTR-B3G things work while wet with solvents, hot, and in a corrosive atmosphere?
Can you position them to serve as coil-current or B-field detectors?
How's the explosion-proofing on those make/break events, in case you need 'em in a
petroleum-handling facility?
Did you ever pot an assembly with relays?

Reeds have their uses.
I'm puzzled how there's a thermocouple problem; other than length of the capsule,
there's nothing to encourage thermal gradients, and both glass-seal metal elements
ARE the same alloy. Contact plating is a few microns, not a problem at all.

[John Larkin]

On Wed, 30 Jun 2021 11:30:46 -0700 (PDT), whit3rd <whi...@gmail.com>
wrote:

>On Saturday, June 19, 2021 at 6:26:16 PM UTC-7, John Larkin wrote:
>
>> We use lots of Fujitsu FTR-B3G relays... almost 100K so far.
>>
>> DPDT, surface mount, small and reliable. We use the latching version
>> at thermocouple levels. They are good up to 3 GHz too.
>>
>> Reeds are expensive junk.
>
>Do your FTR-B3G things work while wet with solvents, hot, and in a corrosive atmosphere?

They work great in our products. DPDT, nanovolt offsets, 3 GHz
bandwidth, $1.32 for the non-latching surface-mount version.

>Can you position them to serve as coil-current or B-field detectors?

Coil current, sure. They have coils. But I want signal switching, not
mag field detectors.

Reed relays are fun B_field sensors. If one is energized, it can keep
an adjacent one from dropping out. The open coil field leaks all over
the place.

>How's the explosion-proofing on those make/break events, in case you need 'em in a
>petroleum-handling facility?
>Did you ever pot an assembly with relays?
>
>Reeds have their uses.
>I'm puzzled how there's a thermocouple problem; other than length of the capsule,
>there's nothing to encourage thermal gradients, and both glass-seal metal elements
>ARE the same alloy. Contact plating is a few microns, not a problem at all.

Nice in theory, but they have terrible thermocouple offsets in real
life. The coil gets hot and the leads are not the same metal as the
reeds.

[John Larkin]

On Wed, 30 Jun 2021 11:00:01 -0700 (PDT), whit3rd <whi...@gmail.com>
wrote:

Why do some people love and defend reeds? They are expensive,
unreliable, usually just SPST, twang, have big thermals, are giant
compared to a modern telecom relay, and have to be kept apart from one
another.

You and sloman love reeds and love China. You're probably the same
person.

[whit3rd]

The leads are part of the magnetic circuit, they ARE the same metal in
all the examples I've seen. Are you talking about RELAY leads, or reed leads?

Coil dissipation is less of an issue in latching relays, but you can make those with reeds
if you care to. It's an easy design with a cruciform pole piece.

[John Larkin]

On Wed, 30 Jun 2021 15:16:26 -0700 (PDT), whit3rd <whi...@gmail.com>

wrote:

>On Wednesday, June 30, 2021 at 1:40:53 PM UTC-7, John Larkin wrote:
>> On Wed, 30 Jun 2021 11:30:46 -0700 (PDT), whit3rd <whi...@gmail.com>
>> wrote:
>> >On Saturday, June 19, 2021 at 6:26:16 PM UTC-7, John Larkin wrote:
>
>> >Reeds have their uses.
>> >I'm puzzled how there's a thermocouple problem; other than length of the capsule,
>> >there's nothing to encourage thermal gradients, and both glass-seal metal elements
>> >ARE the same alloy. Contact plating is a few microns, not a problem at all.
>
>> Nice in theory, but they have terrible thermocouple offsets in real
>> life. The coil gets hot and the leads are not the same metal as the
>> reeds.
>
>The leads are part of the magnetic circuit, they ARE the same metal in
>all the examples I've seen. Are you talking about RELAY leads, or reed leads?

In a reed relay, the tangs that come out of the capsule are welded od
soldered to the copper pins that solder to the PCB. That make two
thermocouple junctions. There is stress relief involved, because
stress on the reeds can make the relay malfunction.

Really, get a reed relay, connect it to a good DVM, apply coil power,
and watch.

>
>Coil dissipation is less of an issue in latching relays, but you can make those with reeds
>if you care to. It's an easy design with a cruciform pole piece.

We use latching versions of the Fujitsus for microvolt stuff. They
cost a bit more.

DPDT is often handy.

[Anthony William Sloman]

But they do have unique properties which are sometimes useful. Nobody sane expects to use the same part for every job, and anybody sensible keeps track of stuff that can be bought off-the shelf to deal with specialised problems.

> You and Sloman love reeds and love China. You're probably the same person.

We don't "love" reeds, or China. We just object to dim-wits like you posting your fatuous misapprehensions about them and pretending you know what you are talking about.

I'm pretty confident that I haven't posted anything that ended up as being shown as coming from whit3rd at gmail.com. John Larkin doesn't seem to know how these things work.

--
Bill Sloman, Sydney

[DecadentLinux...@decadence.org]

Joe Gwinn <joeg...@comcast.net> wrote in
news:3mpncgleb1jbu7339...@4ax.com:

> On Fri, 18 Jun 2021 00:56:55 +0200, Klaus Vestergaard Kragelund
> <klau...@hotmail.com> wrote:
>
>>Hi
>>
>>I am working on a design where we are monitoring a relay, if it
>>has been energized. That is done with high ohmic resistor and a
>>source voltage, detecting a current when drawn
>>

>>But, contact wetting could be a problem, so have added a small
>>capacitor on the contacts, so when they are closed a relatively
>>high current runs for a short period of time to punch through the
>>oxide layer of the contacts
>>

>>I am having difficulty finding any information on the subject, how
>>much current is needed for punch through and how much time.
>>
>>I need a general input, since we do not know if the relay has gold
>>plated contacts or not
>>
>>Anyone got a clue?
>

> There are books written on electrical contacts and their design
> for various conditions. Unless the current is very small gold
> tends to weld. Rhodium is better in such cases.
>
> Joe Gwinn
>

I thought that the best contact surface/media was platinum.

[DecadentLinux...@decadence.org]

John Larkin <jlarkin@highland_atwork_technology.com> wrote in
news:fqqncg50hpi43anlh...@4ax.com:

snip

> We have a similar issue now. We will sell a general-purpose relay
> module as part of a modular test system, and we are concerned that
> once a relay is used to switch power, it may not be good for
> signals any more.
>
> That may be urban legend. It would be awkward to test.

High freq relays with SMA I/O ports are supposed to address small
signal long term reliability.

A GP relay would if the right contact media is used and if it is
environmentally sealed. Then each physical connection node needs to
be gas tight as well Use an anti-oxidant paste like the cable TV
guys use on outside F-fittings. Air exposed contacts can build a
film if the atmosphere they are in.

The old reed switch operated pinball machines were a joy to work on
because the contacts had to be 'burnished' individually because they
used the cheap contact materials back then that would rub off if you
cleaned them wrong. A poor reed switch was always the first thing
one would search for when working on a faulty machine (that contained
several dozen). And they switched higher currents and voltages then
than today's modern pinballs, so the contacts could go bad pretty
quickly. Especially since their closure usually involved the firing
of a solenoid making for back EMF current and arcs at the contacts,
etc.

[DecadentLinux...@decadence.org]

Piglet <erichp...@hotmail.com> wrote in
news:saisfb$e1g$1...@dont-email.me:

> On 17/06/2021 23:56, Klaus Vestergaard Kragelund wrote:
>> Hi
>>
>> I am working on a design where we are monitoring a relay, if it
>> has been energized. That is done with high ohmic resistor and a
>> source voltage, detecting a current when drawn
>>
>> But, contact wetting could be a problem, so have added a small
>> capacitor on the contacts, so when they are closed a relatively
>> high current runs for a short period of time to punch through the
>> oxide layer of the contacts
>>
>> I am having difficulty finding any information on the subject,
>> how much current is needed for punch through and how much time.
>>
>> I need a general input, since we do not know if the relay has
>> gold plated contacts or not
>>
>> Anyone got a clue?
>>

>> Regards
>>
>> Klaus
>
> This Axicom databook looks helpful on different contact material
> properties (circa pages 13-20):
>
> <http://www.farnell.com/datasheets/105262.pdf>
>
> piglet
>

Hahahaha... A quote...

ENG (gas-tight plastic sealed relays filled with insulting
inert gases)


They have to be sealed to keep the insults inside.

[DecadentLinux...@decadence.org]

Piglet <erichp...@hotmail.com> wrote in
news:saisfb$e1g$1...@dont-email.me:

> On 17/06/2021 23:56, Klaus Vestergaard Kragelund wrote:
>> Hi
>>
>> I am working on a design where we are monitoring a relay, if it
>> has been energized. That is done with high ohmic resistor and a
>> source voltage, detecting a current when drawn
>>
>> But, contact wetting could be a problem, so have added a small
>> capacitor on the contacts, so when they are closed a relatively
>> high current runs for a short period of time to punch through the
>> oxide layer of the contacts
>>
>> I am having difficulty finding any information on the subject,
>> how much current is needed for punch through and how much time.
>>
>> I need a general input, since we do not know if the relay has
>> gold plated contacts or not
>>
>> Anyone got a clue?
>>
>> Regards
>>
>> Klaus
>
> This Axicom databook looks helpful on different contact material
> properties (circa pages 13-20):
>
> <http://www.farnell.com/datasheets/105262.pdf>
>
> piglet
>

Very nice. Page 11 up until they start going over their actual
products is very informative about contact science.

[DecadentLinux...@decadence.org]

Clifford Heath <no....@please.net> wrote in
news:168b11ac909b0362$1$2193376$28dd...@news.thecubenet.com:

Pretty goddamned good piece of gear anyway.

[Clifford Heath]

On 2/7/21 3:33 pm, DecadentLinux...@decadence.org wrote:
> Clifford Heath <no....@please.net> wrote in
> news:168b11ac909b0362$1$2193376$28dd...@news.thecubenet.com:

>>> Well it's not 1930 anymore and the applications for reed switches
>>> are few and far in between. The wiki article even mentioned reed
>>> switches being used as sensors in old brushless motors. Awful.
>>
>> Bosch are still using reed switches to make their dishwashers
>> unreliable
>>
>> So much for "German engineering"
>>
>> CH
>>
>
> Pretty goddamned good piece of gear anyway.

50% rate of intermittent or total failure in a $1600 appliance because
of a badly installed $0.01 part does not constitute a "goddamned good
piece of gear".

[Phil Hobbs]

We live in 'opes. ;)

Cheers

Phil Hobbs

[John Walliker]

On Friday, 2 July 2021 at 05:58:09 UTC+1, DecadentLinux...@decadence.org wrote:

> The old reed switch operated pinball machines were a joy to work on
> because the contacts had to be 'burnished' individually because they
> used the cheap contact materials back then that would rub off if you
> cleaned them wrong. A poor reed switch was always the first thing
> one would search for when working on a faulty machine (that contained
> several dozen). And they switched higher currents and voltages then
> than today's modern pinballs, so the contacts could go bad pretty
> quickly. Especially since their closure usually involved the firing
> of a solenoid making for back EMF current and arcs at the contacts,
> etc.

I thought that reed switch contacts were always sealed in a glass tube,
so how were these pinball machine reed switches constructed?
John

[John Larkin]

On Fri, 2 Jul 2021 04:58:02 +0000 (UTC),
DecadentLinux...@decadence.org wrote:

>John Larkin <jlarkin@highland_atwork_technology.com> wrote in
>news:fqqncg50hpi43anlh...@4ax.com:
>
>snip
>
>> We have a similar issue now. We will sell a general-purpose relay
>> module as part of a modular test system, and we are concerned that
>> once a relay is used to switch power, it may not be good for
>> signals any more.
>>
>> That may be urban legend. It would be awkward to test.
>
>
> High freq relays with SMA I/O ports are supposed to address small
>signal long term reliability.

But they are absurdly expensive and don't mount on PC boards.

There are surface-mount RF relays. The Fujitsu FTR-B3GA relays are
pretty good up to 3 GHz.

https://www.dropbox.com/s/se162xpw86hpmzs/DSC06884.JPG?raw=1

[DecadentLinux...@decadence.org]

Clifford Heath <no....@please.net> wrote in

news:168deb1da11cc477$1$1356503$70dd...@news.thecubenet.com:

The one my boss had has lasted for years. The one he has in his new
house is still running as well.

[DecadentLinux...@decadence.org]

John Walliker <jrwal...@gmail.com> wrote in
news:678ed96f-d6f0-4573...@googlegroups.com:

Open air designs throughout the cabinet of the machine. They get
actuated by the pinball hitting a bumper that has a lever or rod behind
it that depresses the reeds together.

[Clive Arthur]

They're not 'reed switches' as we know them today, but that's what some
people called them before 'reed switches' were as we know them today.
They're a bit like ordinary relay contacts, but without the coil and
operated by physical movement or impact.

--
Cheers
Clive
(former amusement arcade technician)

[Rheilly Phoull]

On 3/07/2021 4:07 am, DecadentLinux...@decadence.org wrote:
> Clifford Heath <no....@please.net> wrote in
> news:168deb1da11cc477$1$1356503$70dd...@news.thecubenet.com:
>
>> On 2/7/21 3:33 pm, DecadentLinux...@decadence.org wrote:
>>> Clifford Heath <no....@please.net> wrote in
>>> news:168b11ac909b0362$1$2193376$28dd...@news.thecubenet.com:
>>>>> Well it's not 1930 anymore and the applications for reed
>>>>> switches are few and far in between. The wiki article even
>>>>> mentioned reed switches being used as sensors in old brushless
>>>>> motors. Awful.
>>>>
>>>> Bosch are still using reed switches to make their dishwashers
>>>> unreliable
>>>>
>>>> So much for "German engineering"
>>>>
>>>> CH
>>>>
>>>
>>> Pretty goddamned good piece of gear anyway.
>>
>> 50% rate of intermittent or total failure in a $1600 appliance
>> because of a badly installed $0.01 part does not constitute a
>> "goddamned good piece of gear".
>>
>
> The one my boss had has lasted for years. The one he has in his new
> house is still running as well.
>

We have had 10 years so far out of ours.

[Jasen Betts]

I'd call those leaf switches.

--
Jasen.

[Clifford Heath]

The problem I'm referring to is the models with a water flow sensor
containing a reed switch to detect the rotor moving. The reed switches
are soldered on to a rigid PCB with no provision for thermal expansion,
so depending on the ambient temperature at the time it was soldered, a
temperature excursion or/combined-with a slight shock cracks the glass
at one end, and the reed becomes unreliable. It might be always on,
always off, or might work normally. It might work at some ambient
temperature and not at others. It will probably pass factory QA, under
the same climate conditions where it was manufactured, but become
unreliable as soon as the machine is put on a truck or a boat.

The effect of intermittently reporting water flow problems is very hard
for service techs to diagnose. In our case the dishwasher is in a rental
and we live in another city, so I wasn't there to chase it down myself -
but it caused us dramas multiple times with tenants complaining about
flakey behaviour. There are many reports online of the same kind of
failure, and no clear "here's what's wrong, here's how to fix it" - so I
know this is not just a one-off issue with our unit.

All for not soldering the damn thing down with a tiny allowance for
thermal expansion. Now that I've fixed it, I'm sure the washer will be
good for ten years. But this was just to point out that 99% good
reliable engineering is still 1% unreliable, and that can spoil the
whole party.

Clifford Heath

[gray_wolf]

I worked on pinballs until ca 1977 and I never saw a reed switch in one.
Jukeboxes had a maybe a few.

[DecadentLinux...@decadence.org]

gray_wolf <gray_wolf@howling_mad.com> wrote in
news:iQzEI.104771$Hy1....@fx43.iad:

What do you call them then? The early pinballs had several open
contacts in them. Hell the left and right flipper actuator switches
were (and still are) open contact devices.

[John Larkin]

On Fri, 18 Jun 2021 09:18:19 -0700 (PDT), John Walliker
<jrwal...@gmail.com> wrote:

>Pickering make a wide range of reed relays and provide a lot of advice on how to choose and use them.
>
>www.pickeringrelay.com
>
>John

Yikes. The rep says prices on the surface-mount parts is

"$73.96 - $8.28, depending on the quantity and the part number."