Isolated variable resistor function?

[Bill Sloman]

On Wed, 29 Aug 2012 13:36:15 -0700 (PDT),BillSloman

On 2012-08-29 19:21,BillSlomanwrote:
On Aug 29, 6:30 pm, John Larkin
21:27 -0700, John Larkin
[snip]
I first invented a Manchester decoder (and encoder, and a supervisory
control system that sold megabucks) when I was an EE undergrad.
Deriving the receive clock is easy: a transition detector followed by
a one-shot of 0.75 times the bit period.
"Invented?"  Bwahahahahaha!  That's like Gore claiming he invented the
Internet.  Your "invented" method can be found in ancient textbooks.
                                       ...Jim Thompson
I didn't say I invented it first, just that I invented it. Beforethe
internet, it was hard to access prior art. It's not shocking thatlots
of people independently discovered the same circuits.
It's a bit depressing when an undergraduate with free access to a
university library declares that he couldn't access prior art.
Oh nonsense! Some things are so simple that reinventing them costs
less than looking them up. OK, so it's been done before. So what?
And some things look that simple, and aren't. When you look them up
you find out about the less obvious gotcha's.

More often, you find a circuit in RSI or somewhere and discover that
it's fundamantally unreliable: beta dependent, bad corner cases, stuff
like that. RSI is better read for amusement than guidance.

Too true, which is why instruments papers are wayyyyy below the salt in
academic circles.  I wrote one RSI paper early in my career and then
wised up.

Sturgeon's Law says that 90% of everything is rubbish, and the papers
that get published in RSI don't do any better. The remaining 10% can
be very useful.

This doesn't explain why the instrument literature isn't highly
regarded. The rest of the scientific literature has a similar ratio of
dross to treasure. The instrument literature reports what ought to be
practicable solutions to practical problems, and science is - in the
end - about building theories. Testing those theories is the process
of murdering beautiful theories with mundane facts, and that side of
the business is just less glamorous, if no less important.

Gizmo-building is sort of like cooking--it's close enough to everyday
life that lots of folks have opinions worth listening to, and there are
lots of people who are good enough to make up recipes of their own.
What's original about Aunt Millie's meatloaf recipe?  Who cares?  It
tastes great, fills the gap, and reminds us of good times.

More to the point, it can be made from the ingredients at hand, and
doesn't blow the budget.

Physicists aren't known for economical circuit design - they
persistently use more expensive parts than they need to, and often use
obsolete parts long after better components are freely available. My
1996 comment took a couple of RSI authors to task for waxing lyrical
about the speed advantage of of 10K ECL over TTL at a time when
ECLinPS was freely available and offered a similar speed advantage
over 10K.

 

[Jure Newsgroups]

Joerg,

i don't remember if this part was mentined in this thread,
but here it goes anyway.: H11F1M

http://www.fairchildsemi.com/pf/H1/H11F1M.html
General Description

The H11FXM series consists of a Gallium-Aluminum-

Arsenide IRED emitting diode coupled to a symmetrical

bilateral silicon photo-detector. The detector is electrically

isolated from the input and performs like an ideal

isolated FET designed for distortion-free control of low

level AC and DC analog signals. The H11FXM series

devices are mounted in dual in-line packages.





Thanks, Jure Z.

 

[Bill Sloman]

On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?
Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.

OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.

Regards,
Jeroen Belleman

I haven't found reeds to be reliable.

Bell Labs invented them back in 1936 to get a more reliable relay for
use in the telephone system, and they worked pretty reliably for them
for many years.

Properly used, a dry reed switch is good for about 10^7 operations. A
mercury-wetted reed lasts about ten times as long.

If you haven't found them to be reliable, you probably haven't been
taking enough care to prevent arcing across the contacts

And they have nasty thermal EMFs
and bounces and electroacoustic twangs.

Mercury-wetted reeds don't twang or bounce.

And they are big.

You can buy "sub-minature reeds", but they need bigger magnetic fields
to work.

We've used
about 15,000 of the Fujitsu parts so far, and they are very reliable.
The latching versions have no measurable thermal EMFs, so are good for
thermocouple circuits and such.

Latching reed relays are pretty good on thermal EMFs too. If you don't
have to keep on running current through the coil to keep them closer,
you don't have the thermal gradients that generate the thermocouple
voltages in the first place.

The DPDT Fujitsu parts average about $1.50 each.

In what sort of volume?

 

[Bill Sloman]

Joerg,

i don't remember if this part (H11F1M) was mentioned in this thread,

It was. It was the first thing that Joerg though of, and he spelt out
why it wasn't a good solution for his problem in his original post.

 

[[email protected]]

42, John Larkin wrote:
On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?
Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.
OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.
Regards,
Jeroen Belleman
I haven't found reeds to be reliable.

Bell Labs invented them back in 1936 to get a more reliable relay for
use in the telephone system, and they worked pretty reliably for them
for many years.

Properly used, a dry reed switch is good for about 10^7 operations. A
mercury-wetted reed lasts about ten times as long.

I've seen data sheet claims of 1e9 cycles. In practise, they sometimes
stick closed, or make hi-z closures. The physics is bad: no wiping
action, possibility of residual magnetism latching them closed,
metalurgical bonding sticking contacts closed, magnetic interaction
with adjacent relays. I suspect that getting the reeds exactly
aligned, as molten glass cools and hardens, is a trickly production
challenge.

assembly starts around 4:30

-Lasse

 

[Joerg]

On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?

Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.

OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.

Regards,
Jeroen Belleman

I haven't found reeds to be reliable. And they have nasty thermal EMFs
and bounces and electroacoustic twangs. And they are big.

Bounces? On reeds? I've never seen that happen. Is there a certain
series of bad applas we should all avoid?

I was just going to place these reeds into a design where I have to
artificially "press" the buttons of some gizmo and the currents are
microamps, so no big relays possible because of a lack of wetting current:

http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=9007-12-40&stock=1

Well, if they had gold contacts maybe a bigger one would work. But reeds
are nice because their contacts are naturally encapsulated.

... We've used
about 15,000 of the Fujitsu parts so far, and they are very reliable.
The latching versions have no measurable thermal EMFs, so are good for
thermocouple circuits and such.

The DPDT Fujitsu parts average about $1.50 each.

If gold-contacts, which kind?

 

[Joerg]

Joerg,

i don't remember if this part was mentined in this thread,
but here it goes anyway.: H11F1M

http://www.fairchildsemi.com/pf/H1/H11F1M.html
General Description

The H11FXM series consists of a Gallium-Aluminum-

Arsenide IRED emitting diode coupled to a symmetrical

bilateral silicon photo-detector. The detector is electrically

isolated from the input and performs like an ideal

isolated FET designed for distortion-free control of low

level AC and DC analog signals. The H11FXM series

devices are mounted in dual in-line packages.

I thought of those but dismissed them because they become really
non-linear past +/-30mV or so.

[...]

 

[Bill Sloman]

42, John Larkin wrote:
On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?
Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.
OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.
I haven't found reeds to be reliable.

Bell Labs invented them back in 1936 to get a more reliable relay for
use in the telephone system, and they worked pretty reliably for them
for many years.

Properly used, a dry reed switch is good for about 10^7 operations. A
mercury-wetted reed lasts about ten times as long.

I've seen data sheet claims of 1e9 cycles. In practise, they sometimes
stick closed, or make hi-z closures.

If you've abused the contacts. If the telephone system used them in
millions and found them reliable, it would seem that you are doing
something wrong.

The physics is bad: no wiping
action, possibility of residual magnetism latching them closed,
metallurgical bonding sticking contacts closed, magnetic interaction
with adjacent relays. I suspect that getting the reeds exactly
aligned, as molten glass cools and hardens, is a tricky production
challenge.

But one that was solved a long time ago. You can dream up explanations
of why your relays failed for reasons that don't involve you screwing
up, but don't expect anybody to take you seriously.

No, I've mostly used them in signal-level applications, not power
switching.

So you haven't been deliberately putting a lot of current through them
- that you know about - but if you haven't got your grounding right,
and you ground a subsystem whenever you close a relay you may have
been doing it accidentally and unintentionally. Sadly, the relay
contacts don't care about intentions.

They don't bounce, but they do twang. By "twang" I mean a sustained
mechanical ringdown that generates contact voltage in the magnetic
field of the coil. Lasts many milliseconds. It's hell in low-level
multiplexing.

They shouldn't. The film of liquid mercury between the two contacts
should damp any mechanical movement between them, and pretty rapidly
too. Mercury isn't all that viscous, but the film is pretty thin.

Uh, mercury is illegal now.

Releasing it into the environment is frowned on. Inside a hermetically
sealed reed it's fine.

Reeds really suck.

In unskilled hands.

A Fujitsu DPDT is smaller and cheaper and more reliable than a SPST
reed.

In your hands. Do you actually know what's inside the package that you
buy?

I recall that reeds have thermals even with zero coil power. Each end
of the reed has a thermocouple to the package leads or copper PCB
traces, and they are far apart. I recall measuring microvolts from
small local PCB thermal gradients. The reed metal seems to have a big
thermal potential against copper.

It certainly does. The nickel-iron alloy used make the reed generates
something like 50uV/C - IIRR - against copper, It pays to spread a lot
of copper ground plane around the area to minimise any thermal
gradients.

Reeds suck.

In unskilled hands.

We probably buy them a few hundred, sometimes 1000, at a time.

Got an exact part number? Or a DigiKey stock number? "Fujistu DPDT
relay" is a bit unspecific, but since you don't concoct your own parts
lists, you may not have noticed.

 

[Joerg]

On Thu, 30 Aug 2012 08:51:11 +0200, Jeroen Belleman

On 2012-08-29 21:42, John Larkin wrote:
On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?

Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.

OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.

Regards,
Jeroen Belleman
I haven't found reeds to be reliable. And they have nasty thermal EMFs
and bounces and electroacoustic twangs. And they are big.

Bounces? On reeds? I've never seen that happen. Is there a certain
series of bad applas we should all avoid?

Every non-mercury em relay that I've ever tested has bounced. Reeds
are especially bad; they may close in half a millisecond, then blip
open in random patterns for a couple of milliseconds until the
vibrations settle down.

Typically looks like this:

http://tinyurl.com/9pu8o2n

Thanks, interesting. I have never seen any relay behaving this badly.
Bounce, yes, but only 1-2 times. Never on reeds.

Have to measure some that I've got here. But right know half the world
seems to dump their switch mode converter designs on me, so no time.
Switchers are fun but it's like eating tomatoes all month

 

[Bill Sloman]

21,BillSlomanwrote:
On Aug 29, 6:30 pm, John Larkin
21:27 -0700, John Larkin
[snip]
I first invented a Manchester decoder (and encoder, and a supervisory
control system that sold megabucks) when I was an EE undergrad..
Deriving the receive clock is easy: a transition detector followed by
a one-shot of 0.75 times the bit period.
"Invented?" Bwahahahahaha! That's like Gore claiming he invented the
Internet. Your "invented" method can be found in ancient textbooks.
...Jim Thompson
I didn't say I invented it first, just that I invented it. Before the
internet, it was hard to access prior art. It's not shocking that lots
of people independently discovered the same circuits.
It's a bit depressing when an undergraduate with free access to a
university library declares that he couldn't access prior art.
Oh nonsense! Some things are so simple that reinventing them costs
less than looking them up. OK, so it's been done before. So what?
And some things look that simple, and aren't. When you look them up
you find out about the less obvious gotcha's.
More often, you find a circuit in RSI or somewhere and discover that
it's fundamantally unreliable: beta dependent, bad corner cases, stuff
like that. RSI is better read for amusement than guidance.

You are telling me this? I've now published five critical comments in
RSI, pointing out where their refereeing process had failed - in 1972,
1996, 1999, 2004 and 2009 - and they rejected similar comments I
submitted in 1980 and 1995.

Nobody is saying that every academic paper is to be relied on, but
rejecting the whole of the scientific literature because some of it is
flawed rather misses the point. We make progress because some of the
papers that get published reflect dramatic steps forward.

Larsen N T 1968 Rev. Sci. Instrum. 39 1 12

http://rsi.aip.org/resource/1/rsinak/v39/i1/p1_s1?isAuthorized=no

"Authorized=no"

I'd have to buy the silly thing to see if it's worth reading.

Or go to a university library near you to read it in the bound
journal.

Don't bother buying it. 1968 is a long time ago, and subsequent
authors - including me - have adapted the message for people who don't
work for one or other national bureau of standards. But it was ground-
breaking in its time.

I used to subscribe to RSI. It wasn't worth it. I downloaded maybe 20
interesting papers from the last 30 years or so, but never actually
used them for anything.

Because you didn't learn enough at university to be in a position to
fully understand or exploit them?

 

[Joerg]

John Larkin wrote:

[...]

FUJITSU FTR-B3GB4.5Z-B10 DPDT SMT 30V 1A 4.5V LATCH
OMRON G6KU-2F-Y-DC5

FUJITSU FTR-B3GA4.5Z DPDT SMT 30V 1A 4.5V
OMRON G6K-2G-Y-DC5

Thanks for the hint. Since I am doing a design with relays right now I
checked. Couldn't find the Fujitsu at Digikey but found the 12V
through-hole version (which I need) from Omron:

http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=G6K-2P-Y+DC12

Two bucks fifty, ouch. 2x of what the reeds cost. If you say it's worth
the extra money I'll plop these into the BOM instead of the reeds.

 

[Joerg]

On Aug 29, 11:37 pm, John Larkin<[email protected]>
wrote:
On Wed, 29 Aug 2012 13:36:15 -0700 (PDT),BillSloman









On 2012-08-29 19:21,BillSlomanwrote:

On Aug 29, 6:30 pm, John Larkin
On Wed, 29 Aug 2012 09:04:23 -0700, Jim Thompson

On Tue, 28 Aug 2012 07:21:27 -0700, John Larkin

[snip]

I first invented a Manchester decoder (and encoder, and a
supervisory
control system that sold megabucks) when I was an EE undergrad.
Deriving the receive clock is easy: a transition detector
followed by
a one-shot of 0.75 times the bit period.

"Invented?" Bwahahahahaha! That's like Gore claiming he
invented the
Internet. Your "invented" method can be found in ancient
textbooks.

...Jim Thompson

I didn't say I invented it first, just that I invented it.
Before the
internet, it was hard to access prior art. It's not shocking
that lots
of people independently discovered the same circuits.

It's a bit depressing when an undergraduate with free access to a
university library declares that he couldn't access prior art.

Oh nonsense! Some things are so simple that reinventing them costs
less than looking them up. OK, so it's been done before. So what?

And some things look that simple, and aren't. When you look them up
you find out about the less obvious gotcha's.

More often, you find a circuit in RSI or somewhere and discover that
it's fundamantally unreliable: beta dependent, bad corner cases, stuff
like that. RSI is better read for amusement than guidance.

You are telling me this? I've now published five critical comments in
RSI, pointing out where their refereeing process had failed - in 1972,
1996, 1999, 2004 and 2009 - and they rejected similar comments I
submitted in 1980 and 1995.

Nobody is saying that every academic paper is to be relied on, but
rejecting the whole of the scientific literature because some of it is
flawed rather misses the point. We make progress because some of the
papers that get published reflect dramatic steps forward.

Larsen N T 1968 Rev. Sci. Instrum. 39 1–12

http://rsi.aip.org/resource/1/rsinak/v39/i1/p1_s1?isAuthorized=no

"Authorized=no"

I'd have to buy the silly thing to see if it's worth reading.

is difficult to fault. If you'd read that more for amusement than for
guidance you'd have missed quite a bit

I used to subscribe to RSI. It wasn't worth it. I downloaded maybe 20
interesting papers from the last 30 years or so, but never actually
used them for anything.

The closest thing I've seen to a useful instruments-journal paper was
one by a Danish guy called Lars Bager, back in about 1990. He had this
idea to minimize mode hopping in diode lasers by dithering the bias
current, detecting the medium frequency noise power from the monitor
photodiode, putting that signal into a lock-in, and using an integrating
servo in the usual way to find the minimum.

I built that into one of my gizmos back in the day, but never wound up
using it because it wasn't reliable enough for ship in a product.

I've tried that with polarization hops. But same thing, it only worked
as long as there was no earthquake within 5000 miles and nobody sneezed.
And not when there was a full moon or a black cat crossed the street.

[...]

 

[Joerg]

John Larkin wrote:

[...]

FUJITSU FTR-B3GB4.5Z-B10 DPDT SMT 30V 1A 4.5V LATCH
OMRON G6KU-2F-Y-DC5

FUJITSU FTR-B3GA4.5Z DPDT SMT 30V 1A 4.5V
OMRON G6K-2G-Y-DC5

Thanks for the hint. Since I am doing a design with relays right now I
checked. Couldn't find the Fujitsu at Digikey but found the 12V
through-hole version (which I need) from Omron:

http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=G6K-2P-Y+DC12

Two bucks fifty, ouch. 2x of what the reeds cost. If you say it's worth
the extra money I'll plop these into the BOM instead of the reeds.

We're paying $1.38 average for the non-latch version, $1.80 for
latching. Both surface mount.

Well, since you blessed them the Omron relays are in the design now
instead of the reeds

To my surprise Eagle already had a model for the G6K-2P-Y.

 

[Bill Sloman]

13 pm, John Larkin
42, John Larkin wrote:
On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?
Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surface mount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction of an ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.
OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.
I haven't found reeds to be reliable.
Bell Labs invented them back in 1936 to get a more reliable relay for
use in the telephone system, and they worked pretty reliably for them
for many years.
Properly used, a dry reed switch is good for about 10^7 operations. A
mercury-wetted reed lasts about ten times as long.
I've seen data sheet claims of 1e9 cycles. In practise, they sometimes
stick closed, or make hi-z closures.

If you've abused the contacts. If the telephone system used them in
millions and found them reliable, it would seem that you are doing
something wrong.

But one that was solved a long time ago. You can dream up explanations
of why your relays failed for reasons that don't involve you screwing
up, but don't expect anybody to take you seriously.

So you haven't been deliberately putting a lot of current through them
- that you know about - but if you haven't got your grounding right,
and you ground a subsystem whenever you close a relay you may have
been doing it accidentally and unintentionally. Sadly, the relay
contacts don't care about intentions.

You are, as usual, being an obnoxious ass, going personal/insulting in
a technical discussion, and pretending to be an authority on things
you have no experience with.

I'm not kowtowing to your obvious brilliance? Produce some brilliance
and I'll kowtow to it.
Produce nonsense and I'll tell you it's nonsense.

If you find dry reed relays to be unreliable, you aren't being careful
enough to protect the contacts from excessive currents or voltages.
That you don't understand this is a bit odd. That you don't want to
understand this is rather less surprising.

I've done quite enough work with reed relays - admittedly a long time
ago - that I do happen to know what I'm talking about. You may have
mis-used more of them but you seem to have drawn the wrong conclusion
- reed relays are unreliable - from your occasional mistakes. The
right conclusion would have been that even John Larkin can get it
wrong, but you've got an aversion to recognising that.

I've used tens of thousands of reeds and ditto the little
Fujitsu-type telecom relays. The telecoms are better in every respect.
I won't use reed relays any more.

Your loss.

Next time you design a volume-production product, use reeds. Get back
to us on how they work.

The last ones I designed in were range-changing mercury wetted reeds
on the Cambridge Instruments EBMF 10.5. They worked exceptionally
well. The machine cost a million dollars or so, and Cambridge
Instruments might have sold 12 in a good year - they never did better
than a good six months when I was there.

Now you are making things up. Why would a thin film of mercury,
between closed contacts, damp a mechanical vibration if the reed
structure?

If the reeds are bending, the area in contact is going to be changed
by the differential movement. The mercury film between the contacts
has to move around to accommodate this and it's viscosity damps the
movement and turns the translational energy into heat.

http://en.wikipedia.org/wiki/Relay#Mercury-wetted_relay

http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=m...

But Pickering still lists them - series 88 and 89

http://www.pickeringrelay.com/pdfs/80.pdf

and even Newark will supply a Coto part - albeit it at an extravagant
price, and only in extravagant volume

http://www.newark.com/coto-technolo...F3361?Ntt=mercury+wetted+reed#similarProducts

I can't use a mercury relay in an ROHS product. They look to be
super-expensive anyhow.

They were never cheap, high-volume, parts but "super-expensive" is
wrong. Newark isn't the place to buy that kind of product.

I told you they were the FTR-B3G series. Two that we stock are

FUJITSU FTR-B3GB4.5Z-B10 DPDT SMT 30V 1A 4.5V LATCH
OMRON G6KU-2F-Y-DC5

FUJITSU FTR-B3GA4.5Z DPDT SMT 30V 1A 4.5V
OMRON G6K-2G-Y-DC5

Sorry about that. I should have noticed. I can't say I like the
warning about storage in a contaminated atmosphere degrading the
contacts. The nice thing about reed relays is that the contacts are
hermetically sealed behind a vacuum-tight glass-to-metal seal.

 

[Bill Sloman]

37 pm, John Larkin <[email protected]>
wrote:
21,BillSlomanwrote:
On Aug 29, 6:30 pm, John Larkin
21:27 -0700, John Larkin
[snip]
I first invented a Manchester decoder (and encoder, and a supervisory
control system that sold megabucks) when I was an EE undergrad.
Deriving the receive clock is easy: a transition detector followed by
a one-shot of 0.75 times the bit period.
"Invented?" Bwahahahahaha! That's like Gore claiming he invented the
Internet. Your "invented" method can be found in ancient textbooks.
...Jim Thompson
I didn't say I invented it first, just that I invented it. Before the
internet, it was hard to access prior art. It's not shocking that lots
of people independently discovered the same circuits.
It's a bit depressing when an undergraduate with free access to a
university library declares that he couldn't access prior art.
Oh nonsense! Some things are so simple that reinventing them costs
less than looking them up. OK, so it's been done before. So what?
And some things look that simple, and aren't. When you look them up
you find out about the less obvious gotcha's.
More often, you find a circuit in RSI or somewhere and discover that
it's fundamantally unreliable: beta dependent, bad corner cases, stuff
like that. RSI is better read for amusement than guidance.
You are telling me this? I've now published five critical comments in
RSI, pointing out where their refereeing process had failed - in 1972,
1996, 1999, 2004 and 2009 - and they rejected similar comments I
submitted in 1980 and 1995.
Nobody is saying that every academic paper is to be relied on, but
rejecting the whole of the scientific literature because some of it is
flawed rather misses the point. We make progress because some of the
papers that get published reflect dramatic steps forward.
Larsen N T 1968 Rev. Sci. Instrum. 39 1 12
http://rsi.aip.org/resource/1/rsinak/v39/i1/p1_s1?isAuthorized=no
"Authorized=no"
I'd have to buy the silly thing to see if it's worth reading.

Or go to a university library near you to read it in the bound
journal.

Don't bother buying it. 1968 is a long time ago, and subsequent
authors - including me - have adapted the message for people who don't
work for one or other national bureau of standards. But it was ground-
breaking in its time.

Because you didn't learn enough at university to be in a position to
fully understand or exploit them?

I always found the idea of an advanced degree, or an academic career,
to be repulsive.

Obviously. You didn't learn anything like as much as you should have
done to get your undergraduate degree, and you wouldn't have been able
to get away with that as a post-graduate student.

I got a BSEE because that's what engineers are
expected to have. I'm a circuit designer. They don't teach that
anyhow.

True. But they do teach a lot of stuff that circuit designers find
useful. You can learn it without going through the relevant
undergraduate courses - I did - but then you have to make your way
through the relevant text-books without the help of a lecturer or two
to keep your mind on the job.

The papers I loaded from RSI and some other journals had to do mostly
with tomographic atom probes and delay line detectors and microchannel
plates and FTMS. We actually did a fair amount of work in those areas,
with minimal commercial success. We were working with scientists who
spun out of universities and started companies. Those things tend to
go badly.

Sure. RSI show-cases a lot of what physicists do wrong when it comes
electronic design and development, which is basically that they don't
bother searching the literature and re-invent stuff without getting it
entirely right.

As a graduate student I did search the instrument literature, but this
was seen as odd behaviour.

And designing something that's only got to work for long enough to
generate a paper's worth of results in the hands of highly motivated
graduate student isn't a good background for designing stuff for
production - even small-scale production.

I can go over to the UC Medical research facility in Mission Bay, not
far away. I can access a mess of journals and print articles for 50
cents per sheet. But that's still a hassle. The food over there isn't
all that good either. BARTing to UC Berkeley is another possibility,
bigger hassle. It's easiest to hire some grad-school grunt, who has
student access, to download papers.

Sounds good. At Cambridge I had access to the University libraries by
virtue of joining the Cambridge Philosophical Society, which is why a
lot of people join the Cambridge Philosophical Society, but that was
before down-loading was entirely practical.

 

[Bill Sloman]

On Thu, 30 Aug 2012 00:48:42 -0700 (PDT),BillSloman

On Aug 29, 11:37 pm, John Larkin<[email protected]>
wrote:
21,BillSlomanwrote:
On Aug 29, 6:30 pm, John Larkin
21:27 -0700, John Larkin
[snip]
I first invented a Manchester decoder (and encoder, and a supervisory
control system that sold megabucks) when I was an EE undergrad.
Deriving the receive clock is easy: a transition detector followed by
a one-shot of 0.75 times the bit period.
"Invented?"  Bwahahahahaha!  That's like Gore claiming he invented the
Internet.  Your "invented" method can be found in ancient textbooks.
                                        ...Jim Thompson
I didn't say I invented it first, just that I invented it. Beforethe
internet, it was hard to access prior art. It's not shocking thatlots
of people independently discovered the same circuits.
It's a bit depressing when an undergraduate with free access to a
university library declares that he couldn't access prior art.
Oh nonsense! Some things are so simple that reinventing them costs
less than looking them up. OK, so it's been done before. So what?
And some things look that simple, and aren't. When you look them up
you find out about the less obvious gotcha's.
More often, you find a circuit in RSI or somewhere and discover that
it's fundamantally unreliable: beta dependent, bad corner cases, stuff
like that. RSI is better read for amusement than guidance.
You are telling me this? I've now published five critical comments in
RSI, pointing out where their refereeing process had failed - in 1972,
1996, 1999, 2004 and 2009 - and they rejected similar comments I
submitted in 1980 and 1995.
Nobody is saying that every academic paper is to be relied on, but
rejecting the whole of the scientific literature because some of it is
flawed rather misses the point. We make progress because some of the
papers that get published reflect dramatic steps forward.
Larsen N T 1968 Rev. Sci. Instrum. 39 1�12
http://rsi.aip.org/resource/1/rsinak/v39/i1/p1_s1?isAuthorized=no

"Authorized=no"

I'd have to buy the silly thing to see if it's worth reading.

is difficult to fault. If you'd read that more for amusement than for
guidance you'd have missed quite a bit

I used to subscribe to RSI. It wasn't worth it. I downloaded maybe 20
interesting papers from the last 30 years or so, but never actually
used them for anything.

The closest thing I've seen to a useful instruments-journal paper was
one by a Danish guy called Lars Bager, back in about 1990.  He had this
idea to minimize mode hopping in diode lasers by dithering the bias
current, detecting the medium frequency noise power from the monitor
photodiode, putting that signal into a lock-in, and using an integrating
servo in the usual way to find the minimum.

Sounds crazy. When we had trouble with mode hopping at Fisons Applied
Sensor Technology - because it changed the direction of the laser
beam, and thus the angle of incidence of the beam on the surface being
interrogated, which we were monitoring to about one second of arc - we
just went over to a single mode laser diode and stabilised it's
temperature well enough that it always stayed in the same mode.

I built that into one of my gizmos back in the day, but never wound up
using it because it wasn't reliable enough for ship in a product.

There are papers about instrument construction that are far, far above
RSI and JoP E, e.g. most of the collected works of R. V. Jones, but they
don't go in the instrument ghettos^H^H^H^H^H^H^H journals.

So where do they get published? The British Journal of Scientific
Instruments - now Measurement Science and Technology - got some of
R.V. Jones papers, and Blackett's as well. Today the journal isn't
quite so mind-numbingly academic as RSI, and the refereeing is
marginally better, but they still publish the usual proportion of
trivial rubbish.

http://en.wikipedia.org/wiki/Patrick_Blackett,_Baron_Blackett

 

[Bill Sloman]

00 pm, John Larkin
13 pm, John Larkin
42, John Larkin wrote:
On 2012-08-29 16:08, John Larkin wrote:
[...]
Well, 74-series TTL. Ironically, it's fairly recently that relays
started getting really good, small/fast/reliable/cheap enough to use
in signal circuits instead of CMOS multiplexers.
I must have been looking the other way. What relays are that?
Jeroen Belleman
We use a lot of the Fujitsu FTR-B3G type parts. DPDT, surfacemount,
available as latching or not. NEC and Omron have drop-in equivalents.
As an analog switch, they blow away any CMOS part on everything but
switching speed. Capacitances are sub-1-pF, Ron a fraction ofan ohm,
holdoff hundreds of volts, bandwidth GHz, isolated drive.
OK, thanks. I use a fair lot of Pickering Series 103 reed switches
myself, and sometimes the parasitics get in the way.
I haven't found reeds to be reliable.
Bell Labs invented them back in 1936 to get a more reliable relay for
use in the telephone system, and they worked pretty reliably for them
for many years.
Properly used, a dry reed switch is good for about 10^7 operations.. A
mercury-wetted reed lasts about ten times as long.
I've seen data sheet claims of 1e9 cycles. In practise, they sometimes
stick closed, or make hi-z closures.
If you've abused the contacts. If the telephone system used them in
millions and found them reliable, it would seem that you are doing
something wrong.
The physics is bad: no wiping
action, possibility of residual magnetism latching them closed,
metallurgical bonding sticking contacts closed, magnetic interaction
with adjacent relays. I suspect that getting the reeds exactly
aligned, as molten glass cools and hardens, is a tricky production
challenge.
But one that was solved a long time ago. You can dream up explanations
of why your relays failed for reasons that don't involve you screwing
up, but don't expect anybody to take you seriously.
If you haven't found them to be reliable, you probably haven't been
taking enough care to prevent arcing across the contacts
No, I've mostly used them in signal-level applications, not power
switching.
So you haven't been deliberately putting a lot of current through them
- that you know about - but if you haven't got your grounding right,
and you ground a subsystem whenever you close a relay you may have
been doing it accidentally and unintentionally. Sadly, the relay
contacts don't care about intentions.
You are, as usual, being an obnoxious ass, going personal/insulting in
a technical discussion, and pretending to be an authority on things
you have no experience with.

I'm not kowtowing to your obvious brilliance? Produce some brilliance
and I'll kowtow to it.
Produce nonsense and I'll tell you it's nonsense.

If you find dry reed relays to be unreliable, you aren't being careful
enough to protect the contacts from excessive currents or voltages.
That you don't understand this is a bit odd. That you don't want to
understand this is rather less surprising.

I suspect that reeds are more reliable *if* they work at higher
currents and voltages.

You need noble-metal plating on the contact area for low current -
"dry" - operation. Rhodium, rhenium or iridium, usually on top of a
thicker layer of gold. Power relays use very different contact
materials and need a minimum "wetting" current.

The troubles I've had with them were mostly
low-level analog switching where there were no significant currents or
voltages available at all. The symptoms are stuck contacts
(metallurgical bonding?) and high impedance closures.

If you didn't know enough to specify the right contact materials for
the current level you were working at, it isn't altogether surprising
that the relays that you bought didn't work reliably in your
application

Did you work with thousands of them? I have, and I won't again.

If at first you don't succeed, give up, rather than finding out what
you did wrong ...

Loss? The DPDT telecom relays are smaller, cheaper, surface mount,
multi-sourced, need less coil power, have lower thermal EMFs, and are
far more reliable.

Only in a "non-contaminating atmosphere" if the data sheet is anything
to go by.

So if those reeds had a 0.1% failure rate, you might not have seen it.

There were four per scan board, and two scan boards per machine (X and
Y) machine, and Cambridge Instruments might have sold a hundred
systems before they took over the Philips EBMF - which had been
designed from the ground up in the late 1980's rather than evolved
from a 1970's electron microscope ...

They'd have had an even chance of seeing a 0.1% failure rate, and a
100% chance of hearing about it if one failed. It was a million-dollar
machine, and the customers did recover their investment - the
acceptance test of the first one sold was to produce a couple of
wafer's worth of GaAs RF mosfets and somebody worked out that the
retail value of the parts produced to get acceptance was rather more
than the price of the machine.

Where is that warning? The Fujitsu, Omron, and NEC parts are all epoxy
sealed. We solder/reflow/wash without problems. If we had gigohm-level
leakage, we'd know it.

Try reading the data sheet.

http://components.omron.com/compone...2CFA49FCB185257201007DD585/$file/G6K_0911.pdf

It's half-way down page 8 of the pdf, page 60 on the page, under
"precautions".

Stuff diffuses through epoxy - admittedly hydrogen and helium do
diffuse through glass but at leas they don't screw up the contacts.

It's about contact resistance when the contacts are closed, not
leakage when the contacts are open.

 

[josephkk]

What ever happened to MEMS relays? They were hot stuff for a while.
All sorts of people promised me samples that never arrived.

Here's one: $210 each, stock 0.

http://www.mouser.com/ProductDetail/Omron/2SMES-01/?qs=MTB16yVWt2LlSQLSqPHk6Q==

It can switch half a mA at half a volt.

And that in a nutshell is exactly why they never took off. The physical
dimensions makes for very limited carrying current and very limited
contact operating voltage.

Possibly useful for switching some video and similar LV switching,
especially where high isolation is required.

?-)

 

[Jon Kirwan]

<snip>
Now you are making things up. Why would a thin film of mercury,
between closed contacts, damp a mechanical vibration if the reed
structure?


http://en.wikipedia.org/wiki/Relay#Mercury-wetted_relay
<snip>

Just noticed this quote from your link, John. It says that
mercury wetted relays are for "applications where the mercury
eliminates contact bounce." It's your own link, yet it
contrasts at least somewhat with your earlier comment just
above the link.

Jon

 

[Jeroen Belleman]

About Omron G6K relays.
<snip>

I've been perusing the Omron G6K relay series datasheet. I like
it, but I worry about RF characteristics. I worked out the open
contact capacitance to be about 160fF. It may be half that,
depending on how they got the HF isolation plot; They don't say.

I haven't found anything concerning coupling or capacitance
between the two poles in these relays. Would you have some
measured figures to share?

(I ordered a few to find out...)

Thanks,
Jeroen Belleman