Mains voltage zero crossing detector with optical isolated output

hrh1818

unread,

Feb 15, 2012, 2:27:20 PM2/15/12

to

I am looking for a zero crossing detector IC for use with 120/240,
50/60 Hz mains voltage and has an optical isolated output that is
compatible with micro controllers. I have seen zero crossing triac
drivers with an optical isolated input, such as the MOC3081. But
nothing with an optical isolated output. The nearest I have come so
far is a circuit made with discrete components Hence I will
appreciate it if somebody can post a part number or link to this type
of IC.

Howard

Joerg

unread,

Feb 15, 2012, 3:26:58 PM2/15/12

to

Maybe build it with this?

http://www.avagotech.com/docs/5953-0406E

--
Regards, Joerg

http://www.analogconsultants.com/

John Larkin

unread,

Feb 15, 2012, 3:48:05 PM2/15/12

to

On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
wrote:

I like this:

http://dl.dropbox.com/u/53724080/Circuits/Zero_Cross.JPG

It's a dual optoisolator, with the LEDs antiparallel and the
phototransistors totem-poled. It needs very little line-side current
since the phototransistor load will be a cmos port pin. Software
deglitch maybe, if it's necessary.

--

John Larkin, President
Highland Technology, Inc

jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation

Fred Bloggs

unread,

Feb 15, 2012, 4:23:52 PM2/15/12

to

On Feb 15, 3:48 pm, John Larkin <jlar...@highlandtechnology.com>
wrote:

> On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
> wrote:
>
> >I am looking for a zero crossing detector IC for use with 120/240,
> >50/60 Hz mains voltage and has an optical isolated output that is
> >compatible with micro controllers. I have seen zero crossing triac
> >drivers with an optical isolated input, such as the MOC3081. But
> >nothing with an optical isolated output. The nearest I have come so
> >far is a circuit made with discrete components Hence I will
> >appreciate it if somebody can post a part number or link to this type
> >of IC.
>
> >Howard
>
> I like this:
>
> http://dl.dropbox.com/u/53724080/Circuits/Zero_Cross.JPG
>
> It's a dual optoisolator, with the LEDs antiparallel and the
> phototransistors totem-poled. It needs very little line-side current
> since the phototransistor load will be a cmos port pin. Software
> deglitch maybe, if it's necessary.
>
> --
>
> John Larkin, President
> Highland Technology, Inc
>

> jlarkin at highlandtechnology dot comhttp://www.highlandtechnology.com

>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators
> Custom laser controllers
> Photonics and fiberoptic TTL data links
> VME thermocouple, LVDT, synchro acquisition and simulation

That circuit lacks precision and has too much dead time, also aging
and temperature issues with the LEDs..

http://www.edn.com/article/520185-Mains_driven_zero_crossing_detector_uses_only_a_few_high_voltage_parts.php
is slightly better. The zener defines the zero crossing more
precisely.

A plagiarism http://www.dextrel.net/diyzerocrosser.htm , but more
writeup.

Hmmm- your circuit http://www.edn-europe.com/cmsimages/0208/46f1.jpg
from http://www.edn-europe.com/improvedoptocouplercircuitsreducecurrentdrawresistledaging+article+2021+Europe.html

John Larkin

unread,

Feb 15, 2012, 4:41:07 PM2/15/12

to

How much dead time do you think it has? How much can the OP tolerate?

also aging
>and temperature issues with the LEDs..

Please elaborate.

>
>http://www.edn.com/article/520185-Mains_driven_zero_crossing_detector_uses_only_a_few_high_voltage_parts.php
>is slightly better. The zener defines the zero crossing more
>precisely.

9 parts instead of three. Probably glitch sensitive.

>
>A plagiarism http://www.dextrel.net/diyzerocrosser.htm , but more
>writeup.
>
>Hmmm- your circuit http://www.edn-europe.com/cmsimages/0208/46f1.jpg
>from

Not exactly the same... not totem pole outputs.

http://www.edn-europe.com/improvedoptocouplercircuitsreducecurrentdrawresistledaging+article+2021+Europe.html

Yikes, 19 parts!

Robert Macy

unread,

Feb 15, 2012, 5:04:56 PM2/15/12

to

You didn't say how accurately you want to do the zero crossing
detection. Allowed spiking line noie can significantly shift a simple
trigger point around.

*IF* you need crossover accuracy; you won't be able to get much better
than around 12uS accuracy [2uSrms]. The last time I measured period to
period time for 60Hz AC mains in California yielded around 10-4
accuracy, of course over 10+ cycles the utilities company always
adjusted down to better than 10-6. To avoid noise shifting your zero-
crossing detector around, the BEST is to PLL the 60Hz, generating your
own noiseless cycles, then tap off that with the appropriate phase
shift to 'exactly' sit ON the crossings. That way, you'll be
impervious to the test we used to apply - electric drill plugged into
the same outlet and drill with a lot of load on the bit. When we first
started that kind of testing, you'd be surprised how many SMPS's blew
up.

Jim Thompson

unread,

Feb 15, 2012, 6:03:42 PM2/15/12

to

On Wed, 15 Feb 2012 12:26:58 -0800, Joerg <inv...@invalid.invalid>
wrote:

Fred Bloggs

unread,

Feb 15, 2012, 6:34:14 PM2/15/12

to

On Feb 15, 4:41 pm, John Larkin <jlar...@highlandtechnology.com>
wrote:

Assuming this is 120VAC, then the dead time is at least from +/-1.2V
on the line, so that would be 2.4V/(377 x 170 V/s)= 37us, seems kinda
long for a possible metastable digital input.

>
> also aging

Maybe it's just me, but I go by the old HP standard of 25% efficiency
reduction over the lifetime.

>
> >and temperature issues with the LEDs..
>
> Please elaborate.
>
>
>

> >http://www.edn.com/article/520185-Mains_driven_zero_crossing_detector...

> >is slightly better. The zener defines the zero crossing more
> >precisely.
>
> 9 parts instead of three. Probably glitch sensitive.

It's just a proto, he should use a MOV at least and a bandlimiting
capacitor.

>
>
>
> >A plagiarismhttp://www.dextrel.net/diyzerocrosser.htm, but more
> >writeup.
>
> >Hmmm- your circuithttp://www.edn-europe.com/cmsimages/0208/46f1.jpg

> >from
>
> Not exactly the same... not totem pole outputs.
>

> http://www.edn-europe.com/improvedoptocouplercircuitsreducecurrentdra...
>
> Yikes, 19 parts!

There's a reason they don't like totem-pole.

DJ Delorie

unread,

Feb 15, 2012, 8:52:37 PM2/15/12

to

I put a zero-crossing detector in my laminator hack, using a standard
AC-input opto-isolator, a cap, and a resistor. It produces a signal
that's high for about 50% of the time, when the AC voltage magnitude is
highest. The trick here is, the MCU needs to measure the "off time" of
the signal it gets, and set a timer to trigger half that time from the
next pulse, for each pulse. That puts the timeout right at the zero
crossing. Many MCUs can do this kind of measurement in hardware.

http://www.delorie.com/electronics/laminator/
http://www.delorie.com/electronics/laminator/laminator.pdf

(detector is in the upper right corner)

(before anyone picks on my circuit - yes, I know it's a hack, but it's
MY hack and it works for me ;)

Jamie

unread,

Feb 15, 2012, 10:39:59 PM2/15/12

to

THen you have not looked hard enough.

Jamie

John Larkin

unread,

Feb 15, 2012, 11:23:31 PM2/15/12

to

It's not going to degrade much in your lifetime at, say, 0.25 mA
average LED current. And the totem pole output need an approximate CTR
(...scribbles mathematical equations furiously...) of zero.

>
>>
>> >and temperature issues with the LEDs..
>>
>> Please elaborate.
>>
>>
>>
>> >http://www.edn.com/article/520185-Mains_driven_zero_crossing_detector...
>> >is slightly better. The zener defines the zero crossing more
>> >precisely.
>>
>> 9 parts instead of three. Probably glitch sensitive.
>
>It's just a proto, he should use a MOV at least and a bandlimiting
>capacitor.
>
>>
>>
>>
>> >A plagiarismhttp://www.dextrel.net/diyzerocrosser.htm, but more
>> >writeup.
>>
>> >Hmmm- your circuithttp://www.edn-europe.com/cmsimages/0208/46f1.jpg
>> >from
>>
>> Not exactly the same... not totem pole outputs.
>>
>> http://www.edn-europe.com/improvedoptocouplercircuitsreducecurrentdra...
>>
>> Yikes, 19 parts!
>
>There's a reason they don't like totem-pole.

Maybe because they never thought of it?

--

John Larkin, President Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

hrh1818

unread,

Feb 16, 2012, 12:31:02 AM2/16/12

to

On Feb 15, 5:34 pm, Fred Bloggs <bloggs.fredbloggs.f...@gmail.com>

Thank you for your reply Fred.
The two EDN links you provided have some very good mains voltage,
optical isolated output, zero crossing detector circuits. I will be
using LTspice over the next few days to obtain a better understanding
of how the circuits work. The Dego circuit is the circuit that
initiated this hunt to see if it could be replaced by a single IC.
Evidently there isn't sufficient demand to justify manufacture an IC
for this application.

Howard

Fred Bloggs

unread,

Feb 16, 2012, 11:12:49 AM2/16/12

to

On Feb 15, 11:23 pm, John Larkin

<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Wed, 15 Feb 2012 15:34:14 -0800 (PST), Fred Bloggs
>
>
>
>
>

Nah- that circuit is just a noise amplifier, you will have ultra-long
storage times and a bunch of pulses when the off transistor finally
comes out of saturation, with a lot current shootthrough in between.
It looks good on paper but you will spend a lot of effort cleaning up
that messy output. Also, it would not be unusual for something like
that to take half an hour to finally settle down to steady state
behavior.

hrh1818

unread,

Feb 16, 2012, 12:18:44 PM2/16/12

to

Thank you for your reply Joerg
As far as I can tell the HCPL-3700 is a better fit for an application
that needs to know whether 207 Volts Ac or 240 volts AC is being
applied to a circuit than for use as zero crossing detector.

Howard

hrh1818

unread,

Feb 16, 2012, 12:24:20 PM2/16/12

to

On Feb 15, 2:48 pm, John Larkin <jlar...@highlandtechnology.com>
wrote:

> On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
> wrote:
>
> >I am looking for a zero crossing detector IC for use with 120/240,
> >50/60 Hz mains voltage and has an optical isolated output that is
> >compatible with micro controllers. I have seen zero crossing triac
> >drivers with an optical isolated input, such as the MOC3081. But
> >nothing with an optical isolated output. The nearest I have come so
> >far is a circuit made with discrete components Hence I will
> >appreciate it if somebody can post a part number or link to this type
> >of IC.
>
> >Howard
>
> I like this:
>
> http://dl.dropbox.com/u/53724080/Circuits/Zero_Cross.JPG
>
> It's a dual optoisolator, with the LEDs antiparallel and the
> phototransistors totem-poled. It needs very little line-side current
> since the phototransistor load will be a cmos port pin. Software
> deglitch maybe, if it's necessary.
>
> --
>
> John Larkin, President
> Highland Technology, Inc
>

> jlarkin at highlandtechnology dot comhttp://www.highlandtechnology.com

>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators
> Custom laser controllers
> Photonics and fiberoptic TTL data links
> VME thermocouple, LVDT, synchro acquisition and simulation

Thank you for your reply John.
Your suggestion looks like a nice low parts count approach. I will
simulate the circuit to see how it compares with the other
suggestions.

Howard

hrh1818

unread,

Feb 16, 2012, 12:42:30 PM2/16/12

to

Thank you for your comments on accuracy Robert.
However, the zero crossing detector will be used in application like
D. J,. Delorie's laminator, phase fired power controller, where 1 to
2% accuracy is adequate.

Howard

John Larkin

unread,

Feb 16, 2012, 1:03:29 PM2/16/12

to

How does an "off transistor come out of saturation" ? What does that
mean?

Where do "ultra-long storage times" come from? Makes no sense.

with a lot current shootthrough in between.
>It looks good on paper but you will spend a lot of effort cleaning up
>that messy output. Also, it would not be unusual for something like
>that to take half an hour to finally settle down to steady state
>behavior.

All insane. There is no shoot-through; one transistor turns off, then
the other turns on. Where is there a half hour of time constant? This
will start working nicely in 1/2 of a line cycle.

I mentioned that the uP could deglitch it in software if necessary. Or
if you don't want to do that, and are worried about glitches, add a
tiny-logic Schmitt trigger thing. That works it up to 4 parts total.

The antiparallel coupler/totem pole is far better than an optocoupler
with a pullup resistor, and much cooler besides.

Fred Bloggs

unread,

Feb 16, 2012, 1:27:54 PM2/16/12

to

On Feb 16, 1:03 pm, John Larkin <jlar...@highlandtechnology.com>
wrote:

> On Thu, 16 Feb 2012 08:12:49 -0800 (PST), Fred Bloggs
>
>
>
>
>

The transistor in saturation that your minimal opto is turning off.

>
> Where do "ultra-long storage times" come from? Makes no sense.

Take a look at some numbers with high impedance loading- storage times
in 100's us.

>
> with a lot current shootthrough in between.
>
> >It looks good on paper but you will spend a lot of effort cleaning up
> >that messy output. Also, it would not be unusual for something like
> >that to take half an hour to finally settle down to steady state
> >behavior.
>
> All insane. There is no shoot-through; one transistor turns off, then
> the other turns on. Where is there a half hour of time constant? This
> will start working nicely in 1/2 of a line cycle.

Nope- not even close.

>
> I mentioned that the uP could deglitch it in software if necessary. Or
> if you don't want to do that, and are worried about glitches, add a
> tiny-logic Schmitt trigger thing. That works it up to 4 parts total.
>
> The antiparallel coupler/totem pole is far better than an optocoupler
> with a pullup resistor, and much cooler besides.

Strictly speaking, it is not a totem pole as originally used in
reference to the TTL circuit topology which prevents simultaneous
conduction of the two transistors. Your 'stack' has no such feature.

Joerg

unread,

Feb 16, 2012, 2:13:04 PM2/16/12

to

They'll all be like that because, as you hinted in another post there is
not enough market for a isolated zero-crosser chip. The question is, how
precise do you need to detect the zero crossing? 1-2% as you have hinted
elsewhere should be feasible with an optocoupler device.

If you are willing to install a tiny transformer it gets even easier:
Run the other side into clipping diodes, via a resistor. Now place a
window comparator (can be bought in a chip) around zero and watch its
output.

Spehro Pefhany

unread,

Feb 16, 2012, 2:26:26 PM2/16/12

to

On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
wrote:

While it's not impossible to make such a device, I don't think they
exist. Consumer electronics (the main customer) is very cost
sensitive, and this sort of thing would not likely be cost-effective.

I suggest that you should design something using a standard
optoisolator that is safety-agency approved, along with a few
inexpensive appropriately rated discrete parts, just like everyone
else does.

John Larkin

unread,

Feb 16, 2012, 3:18:19 PM2/16/12

to

What the hell are you talking about? How does high impedance loading -
essentially no loading - slow down a phototransistor turning on? You
don't understand this stuff.

The period of the AC line is 16 milliseconds!

>
>>
>> with a lot current shootthrough in between.
>>
>> >It looks good on paper but you will spend a lot of effort cleaning up
>> >that messy output. Also, it would not be unusual for something like
>> >that to take half an hour to finally settle down to steady state
>> >behavior.
>>
>> All insane. There is no shoot-through; one transistor turns off, then
>> the other turns on. Where is there a half hour of time constant? This
>> will start working nicely in 1/2 of a line cycle.
>
>Nope- not even close.

That's crazy. When there's current in one of the LEDs, why would its
phototransistor wait a half hour to turn on?

>
>>
>> I mentioned that the uP could deglitch it in software if necessary. Or
>> if you don't want to do that, and are worried about glitches, add a
>> tiny-logic Schmitt trigger thing. That works it up to 4 parts total.
>>
>> The antiparallel coupler/totem pole is far better than an optocoupler
>> with a pullup resistor, and much cooler besides.
>
>Strictly speaking, it is not a totem pole as originally used in
>reference to the TTL circuit topology which prevents simultaneous
>conduction of the two transistors. Your 'stack' has no such feature.

It's a totem pole because two NPN transistors are stacked, one above
the other. They can't conduct simultaneously because only one LED can
be driven at a time. The drive frequency is 60 Hz, for Pete's sake.
I've used this totem pole opto circuit at 100 KHz, and it worked fine.

ftp://jjlarkin.lmi.net/Opto_Totem.JPG

I've also use it in linear mode, as the output stage of an amplifier
that swings 400 volts p-p.

ftp://jjlarkin.lmi.net/HVamp.JPG

All this stuff works.

Jim Thompson

unread,

Feb 16, 2012, 3:50:12 PM2/16/12

to

On Thu, 16 Feb 2012 11:13:04 -0800, Joerg <inv...@invalid.invalid>
wrote:

For accuracy, just use a capacitive dropper type power supply to power
a comparator that drives an opto-isolator. Then you can get as
accurate as you desire.

A straight opto-isolator, with no helper gain, is simply a
"straddle-the-zero-crossing" indicator :-)

Fred Bloggs

unread,

Feb 16, 2012, 3:51:36 PM2/16/12

to

On Feb 16, 3:18 pm, John Larkin <jlar...@highlandtechnology.com>
wrote:

I know how to read a datasheet and also understand transistor
saturated operation better than you do apparently. You have a
combination of two phenomena working to turn the detector transistor
OFF when it comes out of saturation: recombination of excess base
charge and collector current discharge. Since you have no collector
current to speak of, your circuit must rely on recombination alone. My
sampling of some popular Toshiba optos list storage times of 100us as
typical with even as much as 4.7K loading to 24V! This means the
transistor you will be trying to turn on will be loaded by the other
transistor which is still a fully conducting saturated switch.

>
> The period of the AC line is 16 milliseconds!
>
>
>
> >> with a lot current shootthrough in between.
>
> >> >It looks good on paper but you will spend a lot of effort cleaning up
> >> >that messy output. Also, it would not be unusual for something like
> >> >that to take half an hour to finally settle down to steady state
> >> >behavior.
>
> >> All insane. There is no shoot-through; one transistor turns off, then
> >> the other turns on. Where is there a half hour of time constant? This
> >> will start working nicely in 1/2 of a line cycle.
>
> >Nope- not even close.
>
> That's crazy. When there's current in one of the LEDs, why would its
> phototransistor wait a half hour to turn on?

Give us some part numbers and your input drive resistor values.

>
>
>
> >> I mentioned that the uP could deglitch it in software if necessary. Or
> >> if you don't want to do that, and are worried about glitches, add a
> >> tiny-logic Schmitt trigger thing. That works it up to 4 parts total.
>
> >> The antiparallel coupler/totem pole is far better than an optocoupler
> >> with a pullup resistor, and much cooler besides.
>
> >Strictly speaking, it is not a totem pole as originally used in
> >reference to the TTL circuit topology which prevents simultaneous
> >conduction of the two transistors. Your 'stack' has no such feature.
>
> It's a totem pole because two NPN transistors are stacked, one above
> the other. They can't conduct simultaneously because only one LED can
> be driven at a time. The drive frequency is 60 Hz, for Pete's sake.
> I've used this totem pole opto circuit at 100 KHz, and it worked fine.

This is a totem pole http://en.wikipedia.org/wiki/File:7400_Circuit.svg
, not what you have, which is just a stack. Your circuits were loaded
to work fine at 100kHz.

>
> ftp://jjlarkin.lmi.net/Opto_Totem.JPG
>
> I've also use it in linear mode, as the output stage of an amplifier
> that swings 400 volts p-p.
>
> ftp://jjlarkin.lmi.net/HVamp.JPG

Do you see that little branch there you have labeled 'FB'? Think that
would make any difference compared to your open loop drive for a zero
crossing app?

>

John S

unread,

Feb 16, 2012, 4:00:24 PM2/16/12

to

Here ya go, Fred...

Version 4
SHEET 1 880 680
WIRE -112 0 -400 0
WIRE 32 0 -112 0
WIRE 512 0 224 0
WIRE 304 64 224 64
WIRE 512 64 512 0
WIRE 32 96 -16 96
WIRE -16 128 -16 96
WIRE -400 144 -400 0
WIRE 304 160 304 64
WIRE 368 160 304 160
WIRE 512 176 512 144
WIRE 368 208 368 160
WIRE -400 240 -400 224
WIRE 48 256 -16 256
WIRE 304 256 304 160
WIRE 304 256 240 256
WIRE -16 288 -16 256
WIRE 368 304 368 288
WIRE 304 320 240 320
WIRE -112 352 -112 0
WIRE 48 352 -112 352
WIRE 304 400 304 320
FLAG 304 400 0
FLAG 512 176 0
FLAG 368 304 0
FLAG -400 240 0
FLAG -16 288 0
FLAG -16 128 0
SYMBOL Optos\\MOC205 128 64 R0
SYMATTR InstName U1
SYMBOL Optos\\MOC205 144 320 R0
SYMATTR InstName U2
SYMBOL current -400 144 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName I1
SYMATTR Value SINE(0 10m 60)
SYMBOL voltage 512 48 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 10
SYMBOL res 352 192 R0
SYMATTR InstName R1
SYMATTR Value 10k
TEXT -434 424 Left 2 !.tran .1

John Larkin

unread,

Feb 16, 2012, 4:01:13 PM2/16/12

to

You quit making sense long ago. All the stuff I posted is known to
work.

John Larkin

unread,

Feb 16, 2012, 4:18:34 PM2/16/12

to

On Thu, 16 Feb 2012 15:00:24 -0600, John S <Sop...@invalid.org>
wrote:

Cool, but if it's going to drive a uP port pin, it doesn't need the
10K. It's more symmetric and elegant without it.

It will behave slightly differently driven from AC line voltage
through a series resistor, 50K or some such, but not much.

Note that it starts up clean, immediately. But I didn't simulate it
for a half hour, so maybe it will break like Fred predicts.

Joerg

unread,

Feb 16, 2012, 4:55:25 PM2/16/12

to

> For accuracy, just use a capacitive dropper type power supply ...

And then some EPA dude makes a new law that outlaws your dropper because
it guzzles 50mW too much :-)

> ... to power

> a comparator that drives an opto-isolator. Then you can get as
> accurate as you desire.
>
> A straight opto-isolator, with no helper gain, is simply a
> "straddle-the-zero-crossing" indicator :-)
>

Very brazen folks use a (small!) Y-rated cap and use that to do an
el-cheapo transfer. Sans optocoupler. But one has to know what one is
doing or there'll be grief.

Fred Bloggs

unread,

Feb 16, 2012, 4:38:55 PM2/16/12

to

> TEXT -434 424 Left 2 !.tran .1- Hide quoted text -
>
> - Show quoted text -

That looks perfect.

Spehro Pefhany

unread,

Feb 16, 2012, 5:07:41 PM2/16/12

to

On Thu, 16 Feb 2012 13:55:25 -0800, Joerg <inv...@invalid.invalid>
wrote:

>
>Very brazen folks use a (small!) Y-rated cap and use that to do an
>el-cheapo transfer. Sans optocoupler. But one has to know what one is
>doing or there'll be grief.

One approach, where accuracy is important (such as phase control) is
to create a DC supply on the 'hot' side and pass the zero crossing as
a pulse straddling the zero crossing, using a high speed optocoupler
such as the venerable H11. You don't need much in the way of dropping
resistor since the LED drive current is only drawn for some tens of
microseconds at the zero crossing (while the input voltage is within a
few volts of zero). Takes a few more components though (on the hot
side).

The simple-simon cheap approach is to use an AC-input optocoupler and
a dropping resistor, and a pullup at the micro. But speed of the opto
vs. watts in the dropping resistor vs. pullup vs. accuracy becomes a
consideration.

Joerg

unread,

Feb 16, 2012, 5:10:23 PM2/16/12

to

Spehro Pefhany wrote:
> On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
> wrote:
>
>> I am looking for a zero crossing detector IC for use with 120/240,
>> 50/60 Hz mains voltage and has an optical isolated output that is
>> compatible with micro controllers. I have seen zero crossing triac
>> drivers with an optical isolated input, such as the MOC3081. But
>> nothing with an optical isolated output. The nearest I have come so
>> far is a circuit made with discrete components Hence I will
>> appreciate it if somebody can post a part number or link to this type
>> of IC.
>>
>> Howard
>
> While it's not impossible to make such a device, I don't think they
> exist. Consumer electronics (the main customer) is very cost
> sensitive, and this sort of thing would not likely be cost-effective.
>

In a consumer app I am sure they'd do it with a safety-rated Y-cap. For
countries with polarized plugs such as the US you only need one,
otherwise two. The rest is super cheap. The Y-cap is the most expensive
part, at around a dime in qties. Of course the circuit ground must be
tied to PE for this to work unless you do it differentially.

> I suggest that you should design something using a standard
> optoisolator that is safety-agency approved, along with a few
> inexpensive appropriately rated discrete parts, just like everyone
> else does.
>

For higher precision this would either require some gain on the line
side or some uC-smarts on the system side to find the middle of the window.

Fred Bloggs

unread,

Feb 16, 2012, 5:18:53 PM2/16/12

to

On Feb 16, 4:00 pm, John S <Soph...@invalid.org> wrote:

> TEXT -434 424 Left 2 !.tran .1- Hide quoted text -
>
> - Show quoted text -

Replace your load resistor with a 10M and parallel it with 7p in
shunt. Looks like it's delayed 300us from the zero crossing. And this
is going to be temperature and age dependent. Given the other
asymmetries, how do you intend to find the zero crossing with any
accuracy?

John Larkin

unread,

Feb 16, 2012, 5:43:20 PM2/16/12

to

John S has Spiced the totem-pole opto thing. With 100K to the AC line,
it outputs a square wave that lags the sine wave zero crossings by
about 400 us. Adding a simple RC phase lead thing on the input side
can nail the zero crossings.

Phil Hobbs

unread,

Feb 16, 2012, 5:49:51 PM2/16/12

to

Or the Tektronix heat shrink trick we talked about a few months ago.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Joerg

unread,

Feb 16, 2012, 6:24:34 PM2/16/12

to

Hmm ... <scratching head> ... can't remember. For LF signal transfer?

Spehro Pefhany

unread,

Feb 16, 2012, 6:40:25 PM2/16/12

to

Here's a refresh (/CAS before /RAS) for your memory:-

https://groups.google.com/group/sci.electronics.design/msg/fb2e2c6643cc3663?hl=en

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

Jim Thompson

unread,

Feb 16, 2012, 6:55:05 PM2/16/12

to

On Thu, 16 Feb 2012 18:40:25 -0500, Spehro Pefhany

aka "gimmick"

George Herold

unread,

Feb 16, 2012, 7:47:05 PM2/16/12

to

> Howard- Hide quoted text -

>
> - Show quoted text -

Can you use spice to understand?
Personally I liked JL's circuit. You'd want to add some serious
filtering to the AC input.

George H.

George Herold

unread,

Feb 16, 2012, 7:50:34 PM2/16/12

to

> http://www.analogconsultants.com/- Hide quoted text -

>
> - Show quoted text -

Yeah, why not a transformer?

George H.

George Herold

unread,

Feb 16, 2012, 8:02:29 PM2/16/12

to

On Feb 16, 6:24 pm, Joerg <inva...@invalid.invalid> wrote:
> Phil Hobbs wrote:
> > On 02/16/2012 04:55 PM, Joerg wrote:
> >> Jim Thompson wrote:

> >>> On Thu, 16 Feb 2012 11:13:04 -0800, Joerg<inva...@invalid.invalid>

> http://www.analogconsultants.com/- Hide quoted text -
>
> - Show quoted text -

Must be wire wrapped on input line, (a gimmick) Heat shrink for
'isolation'.

George H.

Joerg

unread,

Feb 16, 2012, 8:18:07 PM2/16/12

to

For "Is AC there or not" it's ok but I would be careful using that for
precise zero-crossers. On single-phase line-neutral type setups it'll
probably work, on 240VAC or some European systems maybe not so well.

John S

unread,

Feb 16, 2012, 10:10:47 PM2/16/12

to

On 2/16/2012 4:18 PM, Fred Bloggs wrote:

> Replace your load resistor with a 10M and parallel it with 7p in
> shunt. Looks like it's delayed 300us from the zero crossing. And this
> is going to be temperature and age dependent. Given the other
> asymmetries, how do you intend to find the zero crossing with any
> accuracy?

Okay. But let's be really nasty and replace the resistor with an open.
Let's also reduce Vcc to 5V to accommodate a uC.

Drawing on a comment by JL in this thread, I change the source to 170V
peak and add an R/C in series with the LEDs. Following is the modified
file. Is this closer to your needs?

Version 4
SHEET 1 880 680

WIRE -368 0 -400 0
WIRE -352 0 -368 0
WIRE -240 0 -288 0
WIRE -112 0 -160 0

WIRE 32 0 -112 0
WIRE 512 0 224 0
WIRE 304 64 224 64
WIRE 512 64 512 0
WIRE 32 96 -16 96
WIRE -16 128 -16 96
WIRE -400 144 -400 0
WIRE 304 160 304 64

WIRE 352 160 304 160
WIRE 368 160 352 160

WIRE 512 176 512 144

WIRE 368 192 368 160

WIRE -400 240 -400 224
WIRE 48 256 -16 256
WIRE 304 256 304 160
WIRE 304 256 240 256
WIRE -16 288 -16 256

WIRE 368 288 368 256

WIRE 304 320 240 320
WIRE -112 352 -112 0
WIRE 48 352 -112 352
WIRE 304 400 304 320
FLAG 304 400 0
FLAG 512 176 0

FLAG -400 240 0
FLAG -16 288 0
FLAG -16 128 0

FLAG 352 160 out
FLAG -368 0 line
FLAG 368 288 0

SYMBOL Optos\\MOC205 128 64 R0
SYMATTR InstName U1
SYMBOL Optos\\MOC205 144 320 R0
SYMATTR InstName U2

SYMBOL voltage 512 48 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1

SYMATTR Value 5
SYMBOL res -256 16 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 0 56 VBottom 2
SYMATTR InstName R1
SYMATTR Value 17k
SYMBOL cap -352 16 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C1
SYMATTR Value 2.5u
SYMBOL voltage -400 128 R0

WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2

SYMATTR InstName V2
SYMATTR Value SINE(0 170 60)
SYMBOL cap 352 192 R0
SYMATTR InstName C2
SYMATTR Value 7p
TEXT -432 424 Left 2 !.tran 0 1 .9
TEXT 472 288 Left 2 !.step temp list -40 25 70

John S

unread,

Feb 16, 2012, 10:14:47 PM2/16/12

to

On 2/16/2012 9:10 PM, John S wrote:
> On 2/16/2012 4:18 PM, Fred Bloggs wrote:
>
>> Replace your load resistor with a 10M and parallel it with 7p in
>> shunt. Looks like it's delayed 300us from the zero crossing. And this
>> is going to be temperature and age dependent. Given the other
>> asymmetries, how do you intend to find the zero crossing with any
>> accuracy?
>
> Okay. But let's be really nasty and replace the resistor with an open.
> Let's also reduce Vcc to 5V to accommodate a uC.
>
> Drawing on a comment by JL in this thread, I change the source to 170V
> peak and add an R/C in series with the LEDs. Following is the modified
> file. Is this closer to your needs?

Oh! And I also added a temperature step directive.

John Larkin

unread,

Feb 16, 2012, 11:00:01 PM2/16/12

to

On Thu, 16 Feb 2012 21:10:47 -0600, John S <Sop...@invalid.org>
wrote:

Nice. 50K and 0.47 uF also nails the zero crossings, for less power.

--

John Larkin, President Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

John S

unread,

Feb 16, 2012, 11:54:36 PM2/16/12

to

On 2/16/2012 10:00 PM, John Larkin wrote:

>>
>
> Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>
>

You are correct. I was thinking the higher current might be
advantageous, but I see that it offers no advantage. I like your values.

Got any optos to try it out? (I don't have any.)

Fred Bloggs

unread,

Feb 17, 2012, 9:30:37 AM2/17/12

to

The 100us tr/tf are not logic compatible inputs.

John Larkin

unread,

Feb 17, 2012, 10:18:06 AM2/17/12

to

More whining, no designing.

Jim Thompson

unread,

Feb 17, 2012, 10:28:48 AM2/17/12

to

On Fri, 17 Feb 2012 07:18:06 -0800, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Fri, 17 Feb 2012 06:30:37 -0800 (PST), Fred Bloggs
><bloggs.fred...@gmail.com> wrote:
>
>>On Feb 16, 11:54 pm, John S <Soph...@invalid.org> wrote:
>>> On 2/16/2012 10:00 PM, John Larkin wrote:
>>>
>>>
>>>
>>> > Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>>>
>>> You are correct. I was thinking the higher current might be
>>> advantageous, but I see that it offers no advantage. I like your values.
>>>
>>> Got any optos to try it out? (I don't have any.)
>>
>>The 100us tr/tf are not logic compatible inputs.
>
>More whining, no designing.

Quality engineering in the Ian Field/John Larkin tradition >:-}

Fred Bloggs

unread,

Feb 17, 2012, 10:35:15 AM2/17/12

to

On Feb 17, 10:18 am, John Larkin

<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Fri, 17 Feb 2012 06:30:37 -0800 (PST), Fred Bloggs
>

> <bloggs.fredbloggs.f...@gmail.com> wrote:
> >On Feb 16, 11:54 pm, John S <Soph...@invalid.org> wrote:
> >> On 2/16/2012 10:00 PM, John Larkin wrote:
>
> >> > Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>
> >> You are correct. I was thinking the higher current might be
> >> advantageous, but I see that it offers no advantage. I like your values.
>
> >> Got any optos to try it out? (I don't have any.)
>
> >The 100us tr/tf are not logic compatible inputs.
>
> More whining, no designing.
>
> --
>

> John Larkin, President Highland Technology Incwww.highlandtechnology.com jlarkin at highlandtechnology dot com

>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators
> Custom timing and laser controllers
> Photonics and fiberoptic TTL data links
> VME analog, thermocouple, LVDT, synchro, tachometer
> Multichannel arbitrary waveform generators

Seriously? A simple sine-to-square wave converter with some hand
waving is design?

John Larkin

unread,

Feb 17, 2012, 11:29:35 AM2/17/12

to

Yes, and it's fairly elegant, but I don't get your reference to hand
waving; you claimed it would take a half hour to settle, and you were
dead wrong. It works just like I said it would.

As I have noted, it could be software deglitched, or one could add a
Schmitt gate if risetime mattered. In many uP applications, the edge
rate of this opto thing would work fine; details depend on the
application.

So quit whining and design something better yourself. Post it. Show
us.

--

John Larkin, President Highland Technology Inc

www.highlandtechnology.com jlarkin at highlandtechnology dot com

John Larkin

unread,

Feb 17, 2012, 11:30:40 AM2/17/12

to

On Fri, 17 Feb 2012 08:28:48 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Fri, 17 Feb 2012 07:18:06 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>>On Fri, 17 Feb 2012 06:30:37 -0800 (PST), Fred Bloggs
>><bloggs.fred...@gmail.com> wrote:
>>
>>>On Feb 16, 11:54 pm, John S <Soph...@invalid.org> wrote:
>>>> On 2/16/2012 10:00 PM, John Larkin wrote:
>>>>
>>>>
>>>>
>>>> > Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>>>>
>>>> You are correct. I was thinking the higher current might be
>>>> advantageous, but I see that it offers no advantage. I like your values.
>>>>
>>>> Got any optos to try it out? (I don't have any.)
>>>
>>>The 100us tr/tf are not logic compatible inputs.
>>
>>More whining, no designing.
>
>Quality engineering in the Ian Field/John Larkin tradition >:-}

Stupid whiney old hen. If you can do better, show us.

Jim Thompson

unread,

Feb 17, 2012, 11:52:36 AM2/17/12

to

On Fri, 17 Feb 2012 08:30:40 -0800, John Larkin

<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Fri, 17 Feb 2012 08:28:48 -0700, Jim Thompson
><To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>
>>On Fri, 17 Feb 2012 07:18:06 -0800, John Larkin
>><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>
>>>On Fri, 17 Feb 2012 06:30:37 -0800 (PST), Fred Bloggs
>>><bloggs.fred...@gmail.com> wrote:
>>>
>>>>On Feb 16, 11:54 pm, John S <Soph...@invalid.org> wrote:
>>>>> On 2/16/2012 10:00 PM, John Larkin wrote:
>>>>>
>>>>>
>>>>>
>>>>> > Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>>>>>
>>>>> You are correct. I was thinking the higher current might be
>>>>> advantageous, but I see that it offers no advantage. I like your values.
>>>>>
>>>>> Got any optos to try it out? (I don't have any.)
>>>>
>>>>The 100us tr/tf are not logic compatible inputs.
>>>
>>>More whining, no designing.
>>
>>Quality engineering in the Ian Field/John Larkin tradition >:-}
>
>Stupid whiney old hen. If you can do better, show us.

After you defend 100us tr/tf as adequate.

Stupid whiney technologically inadequate asshole.

John Larkin

unread,

Feb 17, 2012, 12:03:13 PM2/17/12

to

On Fri, 17 Feb 2012 09:52:36 -0700, Jim Thompson

<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Fri, 17 Feb 2012 08:30:40 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>>On Fri, 17 Feb 2012 08:28:48 -0700, Jim Thompson
>><To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>
>>>On Fri, 17 Feb 2012 07:18:06 -0800, John Larkin
>>><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>
>>>>On Fri, 17 Feb 2012 06:30:37 -0800 (PST), Fred Bloggs
>>>><bloggs.fred...@gmail.com> wrote:
>>>>
>>>>>On Feb 16, 11:54 pm, John S <Soph...@invalid.org> wrote:
>>>>>> On 2/16/2012 10:00 PM, John Larkin wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>> > Nice. 50K and 0.47 uF also nails the zero crossings, for less power.
>>>>>>
>>>>>> You are correct. I was thinking the higher current might be
>>>>>> advantageous, but I see that it offers no advantage. I like your values.
>>>>>>
>>>>>> Got any optos to try it out? (I don't have any.)
>>>>>
>>>>>The 100us tr/tf are not logic compatible inputs.
>>>>
>>>>More whining, no designing.
>>>
>>>Quality engineering in the Ian Field/John Larkin tradition >:-}
>>
>>Stupid whiney old hen. If you can do better, show us.
>
>After you defend 100us tr/tf as adequate.

Idiot. I said, several times, that it can be software deglitched (but
you don't understand software) or Schmitt'ed if it matters. The opto
totem is a cute circuit, with potential, but you refuse to think about
it because I suggested it.

Did I mention lately that you're an idiot?

Design something better, you whiney old hag. Show us. Or tell us more
about your various aches and pains.

Phil Hobbs

unread,

Feb 17, 2012, 12:23:38 PM2/17/12

to

For detecting the presence of power. You heatshrink a piece of wire to
the power cord inside the case, and use (say) a two-section RC lowpass
to get rid of the junk and EMI. It's easy to control the phase shift,
and you don't need to bring the mains voltage onto your board.
Vladimir hated it, but I thought it was pretty.

John Larkin

unread,

Feb 17, 2012, 12:31:35 PM2/17/12

to

We power our benchtop delay generator from an external 24VDC wart, so
we don't get a line trigger. We considered all sorts of radical ways
to do that (phase lock to wart ripple, electrostatic antenna,
whatever). We would up requiring a second small AC wart to do line
trigger.

Fred Bloggs

unread,

Feb 17, 2012, 12:42:15 PM2/17/12

to

On Feb 17, 11:29 am, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>
> ... but I don't get your reference to hand

> waving; you claimed it would take a half hour to settle, and you were
> dead wrong. It works just like I said it would.

That's a bold assertion based upon a rather crude SPICE simulation.
You were already caught with at least one unanticipated performance
flaw with that ridiculous phase offset.

Phil Hobbs

unread,

Feb 17, 2012, 12:45:10 PM2/17/12

to

Modem technology to the rescue.

Well, so it has two warts then. ;) It's a pretty nice box otherwise
though--much better than my old SRS one.

John Larkin

unread,

Feb 17, 2012, 1:10:08 PM2/17/12

to

On Fri, 17 Feb 2012 09:42:15 -0800 (PST), Fred Bloggs
<bloggs.fred...@gmail.com> wrote:

>On Feb 17, 11:29 am, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>>
>> ... but I don't get your reference to hand
>> waving; you claimed it would take a half hour to settle, and you were
>> dead wrong. It works just like I said it would.
>
>That's a bold assertion based upon a rather crude SPICE simulation.

It's a very simple circuit. How could the sim be complex?

Are you still claiming it will take a half hour to settle? Check one
box:

YES \ \

NO \ \

>You were already caught with at least one unanticipated performance
>flaw with that ridiculous phase offset.

The unequalized phase lag is a few hundred us, somewhat less as you
burn more power in the line-side resistors. That's hardly
"unanticipated", it's obvious that this circuit has some time lag.
Whether that matters is application dependant. If this circuit doesn't
work for your application, then don't use it. As John S noted, adding
one cap greatly improves zero-crossing timing.

Quit being a whining old hag and design something yourself. Show us.

--

John Larkin, President
Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser controllers

Photonics and fiberoptic TTL data links

Jim Thompson

unread,

Feb 17, 2012, 2:02:00 PM2/17/12

to

On Wed, 15 Feb 2012 12:26:58 -0800, Joerg <inv...@invalid.invalid>

wrote:

>hrh1818 wrote:
>> I am looking for a zero crossing detector IC for use with 120/240,
>> 50/60 Hz mains voltage and has an optical isolated output that is
>> compatible with micro controllers. I have seen zero crossing triac
>> drivers with an optical isolated input, such as the MOC3081. But
>> nothing with an optical isolated output. The nearest I have come so
>> far is a circuit made with discrete components Hence I will
>> appreciate it if somebody can post a part number or link to this type
>> of IC.
>>
>> Howard
>
>
>Maybe build it with this?
>
>http://www.avagotech.com/docs/5953-0406E

That's mostly useful for slow events, such as telephone ring voltages.
I used that very part in my 800-number killer.

Joerg

unread,

Feb 17, 2012, 2:10:00 PM2/17/12

to

Jim Thompson wrote:
> On Wed, 15 Feb 2012 12:26:58 -0800, Joerg <inv...@invalid.invalid>
> wrote:
>
>> hrh1818 wrote:
>>> I am looking for a zero crossing detector IC for use with 120/240,
>>> 50/60 Hz mains voltage and has an optical isolated output that is
>>> compatible with micro controllers. I have seen zero crossing triac
>>> drivers with an optical isolated input, such as the MOC3081. But
>>> nothing with an optical isolated output. The nearest I have come so
>>> far is a circuit made with discrete components Hence I will
>>> appreciate it if somebody can post a part number or link to this type
>>> of IC.
>>>
>>> Howard
>>
>> Maybe build it with this?
>>
>> http://www.avagotech.com/docs/5953-0406E
>
> That's mostly useful for slow events, such as telephone ring voltages.
> I used that very part in my 800-number killer.
>

We've been left alone quite nicely since the Do-not-call registry came
into place. The occasional "poll" or something, but that's it.

Jim Thompson

unread,

Feb 17, 2012, 2:18:14 PM2/17/12

to

On Fri, 17 Feb 2012 11:10:00 -0800, Joerg <inv...@invalid.invalid>
wrote:

Here we've been plagued with "polls" using local numbers, so I'm
contemplating a white-list approach to call-blocking. If you're not
on the white-list you'll have to enter a roving code to even get to
voice-mail ;-)

Joerg

unread,

Feb 17, 2012, 3:02:06 PM2/17/12

to

Screening all calls? Yuck.

Next step would be that nobody can enter your house unless fingerprinted :-)

Jim Thompson

unread,

Feb 17, 2012, 3:10:11 PM2/17/12

to

On Fri, 17 Feb 2012 12:02:06 -0800, Joerg <inv...@invalid.invalid>

Why do you say "yuck"? I don't like to be bothered by all these
idiots who violate the do-not-call list. My present system doesn't
ring AT ALL if you're in the "killfile", yet reports CID to all phones
in the house... that's the problem I solved that commercial units
don't provide. I'm just going to turn it around... ring only if
you're whitelisted, voicemail opportunity otherwise... except if
you're any 800 variant... then black-hole ;-)

>
>Next step would be that nobody can enter your house unless fingerprinted :-)

Good grief! Though I do treat unexpected door-bell rings with
caution. You never know when some nutcase, like Larkin, will go
ballistic and show up at my door ;-)

John Fields

unread,

Feb 17, 2012, 3:15:41 PM2/17/12

to

On Fri, 17 Feb 2012 09:31:35 -0800, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>We power our benchtop delay generator from an external 24VDC wart, so
>we don't get a line trigger. We considered all sorts of radical ways
>to do that (phase lock to wart ripple, electrostatic antenna,
>whatever). We would up requiring a second small AC wart to do line
>trigger.

---
I'm surprised you didn't use a single AC out wart and do the AC -> DC
thing internal to the generator.

--
JF

Joerg

unread,

Feb 17, 2012, 3:38:47 PM2/17/12

to

So a new client or long-lost relative would automatically be directed
into voice mail? Ok, it's your phone.

[...]

Jim Thompson

unread,

Feb 17, 2012, 3:47:15 PM2/17/12

to

On Fri, 17 Feb 2012 12:38:47 -0800, Joerg <inv...@invalid.invalid>
wrote:

[snip]

>>
>
>So a new client or long-lost relative would automatically be directed
>into voice mail?

Yes

>Ok, it's your phone.
>

Yes, it is.

John Larkin

unread,

Feb 17, 2012, 3:48:47 PM2/17/12

to

The problem there is that it wouldn't be 90-240 volt universal. We
figured that a minority of users would want to line-trigger a
picosecond-resolution digital delay generator, so the extra wart isn't
a big deal.

--

John Larkin, President
Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser controllers

Photonics and fiberoptic TTL data links

Joerg

unread,

Feb 17, 2012, 4:18:14 PM2/17/12

to

Jim Thompson wrote:
> On Fri, 17 Feb 2012 12:38:47 -0800, Joerg <inv...@invalid.invalid>
> wrote:
>
> [snip]
>> So a new client or long-lost relative would automatically be directed
>> into voice mail?
>
> Yes
>

Yuck :-)

--
SCNR, Joerg

http://www.analogconsultants.com/

Michael A. Terrell

unread,

Feb 17, 2012, 4:33:18 PM2/17/12

to

The next thing you know, you'll be getting phone calls from people on
usenet. ;-)

--
You can't have a sense of humor, if you have no sense.

Jim Thompson

unread,

Feb 17, 2012, 4:44:52 PM2/17/12

to

Actually, I do. Regularly. Seeking circuit help without the pain and
agony of Larkin obfuscating a logical progression to a proper
solution.

k...@att.bizzzzzzzzzzzz

unread,

Feb 17, 2012, 7:58:25 PM2/17/12

to

That analogy doesn't float. Your argument is more like not shooting people
who enter your house uninvited. ;-)

Jim Thompson

unread,

Feb 17, 2012, 8:10:44 PM2/17/12

to

On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>

wrote:

>I am looking for a zero crossing detector IC for use with 120/240,
>50/60 Hz mains voltage and has an optical isolated output that is
>compatible with micro controllers. I have seen zero crossing triac
>drivers with an optical isolated input, such as the MOC3081. But
>nothing with an optical isolated output. The nearest I have come so
>far is a circuit made with discrete components Hence I will
>appreciate it if somebody can post a part number or link to this type
>of IC.
>
>Howard

It's downright trivial to produce an accurate edge transition at the
zero crossing, with enough oomph to whack an opto-coupler into
producing a sharp edge...

http://www.analog-innovations.com/SED/ZeroCross_SIM1.pdf

as I opined in a much earlier post: all it takes is a simple-minded
capacitive voltage dropper to power a comparator.

So the remaining question is what kind of pulse do you want to see on
the other side of the opto-coupler... positive going at each
zero-crossing with what width, or what?

Please define and I'll finish the exercise.

Michael A. Terrell

unread,

Feb 17, 2012, 8:51:26 PM2/17/12

to

Jim Thompson wrote:

>
> Michael A. Terrell wrote:
> >
> > The next thing you know, you'll be getting phone calls from people on
> >usenet. ;-)
>
> Actually, I do. Regularly. Seeking circuit help without the pain
> and agony of Larkin obfuscating a logical progression to a proper
> solution.

I was poking fun of calling you a year or so ago, jut to say hello.
:)

Jim Thompson

unread,

Feb 17, 2012, 9:21:45 PM2/17/12

to

On Fri, 17 Feb 2012 20:51:26 -0500, "Michael A. Terrell"
<mike.t...@earthlink.net> wrote:

>
>Jim Thompson wrote:
>>
>> Michael A. Terrell wrote:
>> >
>> > The next thing you know, you'll be getting phone calls from people on
>> >usenet. ;-)
>>
>> Actually, I do. Regularly. Seeking circuit help without the pain
>> and agony of Larkin obfuscating a logical progression to a proper
>> solution.
>
>
> I was poking fun of calling you a year or so ago, jut to say hello.
>:)

Ah, yes! You did call me ;-)

Jim Thompson

unread,

Feb 18, 2012, 11:41:13 AM2/18/12

to

On Fri, 17 Feb 2012 18:10:44 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Wed, 15 Feb 2012 11:27:20 -0800 (PST), hrh1818 <hr...@att.net>
>wrote:
>
>>I am looking for a zero crossing detector IC for use with 120/240,
>>50/60 Hz mains voltage and has an optical isolated output that is
>>compatible with micro controllers. I have seen zero crossing triac
>>drivers with an optical isolated input, such as the MOC3081. But
>>nothing with an optical isolated output. The nearest I have come so
>>far is a circuit made with discrete components Hence I will
>>appreciate it if somebody can post a part number or link to this type
>>of IC.
>>
>>Howard
>
>It's downright trivial to produce an accurate edge transition at the
>zero crossing, with enough oomph to whack an opto-coupler into
>producing a sharp edge...
>
> http://www.analog-innovations.com/SED/ZeroCross_SIM1.pdf
>
>as I opined in a much earlier post: all it takes is a simple-minded
>capacitive voltage dropper to power a comparator.
>
>So the remaining question is what kind of pulse do you want to see on
>the other side of the opto-coupler... positive going at each
>zero-crossing with what width, or what?
>
>Please define and I'll finish the exercise.
>
> ...Jim Thompson

Perhaps like this...

http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf

Choice of comparator is left as an exercise for the student... but
even a lowly LM339 will do if you can stand the 300ns of extra delay
;-)

Of course Mr. Snotnose, "company president", will say too many parts.

John Larkin

unread,

Feb 18, 2012, 11:53:52 AM2/18/12

to

On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson

Idiot. An LM339 will go nuts from the negative input swing. What did
you use to model it, an ideal comparator?

>
>Of course Mr. Snotnose, "company president", will say too many parts.
>
> ...Jim Thompson

Hey, 17 parts isn't bad, for an amateur. But you'll need to work on it
to get it past UL.

--

John Larkin, President Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

Jim Thompson

unread,

Feb 18, 2012, 12:08:35 PM2/18/12

to

Gotcha! But you're too dumb to know why ;-)

>
>>
>>Of course Mr. Snotnose, "company president", will say too many parts.
>>
>> ...Jim Thompson
>
>
>Hey, 17 parts isn't bad, for an amateur. But you'll need to work on it
>to get it past UL.

The 4N25 probably doesn't meet UL, but there's nothing else any worse
than your approach.

You're nothing more than a narcissistic ignorant SOB who constantly
needs to be on your soapbox so you can claim adequacy.

Your claim has been denied ;-)

Jim Thompson

unread,

Feb 18, 2012, 3:38:43 PM2/18/12

to

On Sat, 18 Feb 2012 10:08:35 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Sat, 18 Feb 2012 08:53:52 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>

[snip]

>>>
>>>Perhaps like this...
>>>
>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>
>>>Choice of comparator is left as an exercise for the student... but
>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>;-)
>>
>>Idiot. An LM339 will go nuts from the negative input swing. What did
>>you use to model it, an ideal comparator?
>
>Gotcha! But you're too dumb to know why ;-)

More on that later. Waiting to see how deeply Larkin will try to
immerse his foot in the cow patty ;-)

>
>>
>>>
>>>Of course Mr. Snotnose, "company president", will say too many parts.
>>>
>>> ...Jim Thompson
>>
>>
>>Hey, 17 parts isn't bad, for an amateur. But you'll need to work on it
>>to get it past UL.
>

A colleague reminds me that in 2001 we used a very similar method to
power a washing machine controller for Emerson Appliance Controls
Division. In fact most of the controller was on the line side of the
boundary. I assure you that Emerson got UL approval.

Jasen Betts

unread,

Feb 18, 2012, 3:45:40 PM2/18/12

to

On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>

>>Perhaps like this...
>>
>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>
>>Choice of comparator is left as an exercise for the student... but
>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>;-)
>
> Idiot. An LM339 will go nuts from the negative input swing. What did
> you use to model it, an ideal comparator?

LM339 behaves sensibly with inputs as low as -0.6V,
it'll need a pull-up on the output though,

--
⚂⚃ 100% natural

--- Posted via news://freenews.netfront.net/ - Complaints to ne...@netfront.net ---

Fred Bartoli

unread,

Feb 18, 2012, 4:09:35 PM2/18/12

to

John Larkin a �crit :

That one is pretty low count, low power, and need no clean up
circuit/soft,...

I don't think there's a mean to save one opto :-(

10n
||
.---||-------+----.
| || | |
| V -> |
| OPT1 - |
| | |
| |/ |
| .--| z
1K | 220K | |> A 24V
___ | ___ | | |
--|___|-+-|___|-+ +----+-----.
MAIN | | | | |
-. | | |< z |
| | '--| A 24V |
| | |\ | |
| | | | |
| | OPT2 V -> | |
| | 10n - | |
| | || | | |
| '---||-------+----' |
| || |
'------------------------------'

--
Thanks,
Fred.

amdx

unread,

Feb 18, 2012, 4:22:06 PM2/18/12

to

I bet you say that about all the kids!
Mikek

Jim Thompson

unread,

Feb 18, 2012, 5:28:37 PM2/18/12

to

On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:

>On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>>
>>>Perhaps like this...
>>>
>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>
>>>Choice of comparator is left as an exercise for the student... but
>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>;-)
>>
>> Idiot. An LM339 will go nuts from the negative input swing. What did
>> you use to model it, an ideal comparator?
>
>LM339 behaves sensibly with inputs as low as -0.6V,

Yes. Under most conditions. And only one input misbehaves even if
you suck current out of it.

>it'll need a pull-up on the output though,

Yes.

Jim Thompson

unread,

Feb 18, 2012, 5:32:47 PM2/18/12

to

No. I'm working with a company right now where _everyone_ is younger
than my children... with some even as young as my oldest grandchild.
And they are peaches to work with. Seems their professors at RIT used
some of my early circuit designs as teaching examples ;-)

And it's no piddling company... they just shipped their BILLIONETH
chip!

lang...@fonz.dk

unread,

Feb 18, 2012, 6:20:44 PM2/18/12

to

On 16 Feb., 23:43, John Larkin <jlar...@highlandtechnology.com> wrote:
> On Thu, 16 Feb 2012 17:07:41 -0500, Spehro Pefhany
>
>
>
>
>
>
>
>
>
> <speffS...@interlogDOTyou.knowwhat> wrote:
> >On Thu, 16 Feb 2012 13:55:25 -0800, Joerg <inva...@invalid.invalid>
> >wrote:
>
> >>Very brazen folks use a (small!) Y-rated cap and use that to do an
> >>el-cheapo transfer. Sans optocoupler. But one has to know what one is
> >>doing or there'll be grief.
>
> >One approach, where accuracy is important (such as phase control) is
> >to create a DC supply on the 'hot' side and pass the zero crossing as
> >a pulse straddling the zero crossing, using a high speed optocoupler
> >such as the venerable H11. You don't need much in the way of dropping
> >resistor since the LED drive current is only drawn for some tens of
> >microseconds at the zero crossing (while the input voltage is within a
> >few volts of zero). Takes a few more components though (on the hot
> >side).
>
> >The simple-simon cheap approach is to use an AC-input optocoupler and
> >a dropping resistor, and a pullup at the micro. But speed of the opto
> >vs. watts in the dropping resistor vs. pullup vs. accuracy becomes a
> >consideration.
>
> John S has Spiced the totem-pole opto thing. With 100K to the AC line,
> it outputs a square wave that lags the sine wave zero crossings by
> about 400 us. Adding a simple RC phase lead thing on the input side
> can nail the zero crossings.
>

if the signal goes to a micro, accurately detecting zero doesn't seem
so important, as long as the delay is repeatable it is easily
compensated
in software

the totem-pole is cute, but I think I'd wire the outputs as open
collector OR, that way everything is symmetric with regards to the
input
and using rising or falling edge can give you a leading or lagging
trigger

-Lasse

-Lasse

John Larkin

unread,

Feb 18, 2012, 8:01:26 PM2/18/12

to

On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:

>On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>>
>>>Perhaps like this...
>>>
>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>
>>>Choice of comparator is left as an exercise for the student... but
>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>;-)
>>
>> Idiot. An LM339 will go nuts from the negative input swing. What did
>> you use to model it, an ideal comparator?
>
>LM339 behaves sensibly with inputs as low as -0.6V,
>it'll need a pull-up on the output though,

Right, that makes two dumb mistakes. And using a pullup adds
complications.

On many LM339's, including the National original, pulling any input
more than 0.3 volts below ground will spray charge all over the chip
and mess up all the sections. Very weird things happen.

See the Natiional datasheet, footnote 3. I'm shocked that a
self-proclaimed Master Circuit Designer would screw up things this
badly.

Jim Thompson

unread,

Feb 18, 2012, 8:08:17 PM2/18/12

to

On Sat, 18 Feb 2012 17:01:26 -0800, John Larkin

<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:
>
>>On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>>>
>>>>Perhaps like this...
>>>>
>>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>>
>>>>Choice of comparator is left as an exercise for the student... but
>>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>>;-)
>>>
>>> Idiot. An LM339 will go nuts from the negative input swing. What did
>>> you use to model it, an ideal comparator?
>>
>>LM339 behaves sensibly with inputs as low as -0.6V,
>>it'll need a pull-up on the output though,
>
>Right, that makes two dumb mistakes. And using a pullup adds
>complications.
>
>On many LM339's, including the National original, pulling any input
>more than 0.3 volts below ground will spray charge all over the chip
>and mess up all the sections. Very weird things happen.

"Sprays charge"?? Are you really that ignorant?

>
>See the Natiional datasheet, footnote 3. I'm shocked that a
>self-proclaimed Master Circuit Designer would screw up things this
>badly.
>

"Note 3: This input current will only exist when the voltage at any of
the input leads is driven negative. It is due to the collector-base
junction of the input PNP
transistors becoming forward biased and thereby acting as input diode
clamps. In addition to this diode action, there is also lateral NPN
parasitic transistor action
on the IC chip. This transistor action can cause the output voltages
of the comparators to go to the V+ voltage level (or to ground for a
large overdrive) for the time
duration that an input is driven negative. This is not destructive and
normal output states will re-establish when the input voltage, which
was negative, again returns
to a value greater than -0.3 VDC (at 25°)C."

The 0.3V is a cover-your-ass specification.

Are you really that ignorant? Yes you are.

John Larkin

unread,

Feb 18, 2012, 8:38:34 PM2/18/12

to

You've posted pics of the interior. I wouldn't go in there if you paid
me.

k...@att.bizzzzzzzzzzzz

unread,

Feb 18, 2012, 8:40:36 PM2/18/12

to

You really can't count on doing that unless you have some means of
calibration. Withoug a lot of gain the uncertainly window is too large.

Jim Thompson

unread,

Feb 18, 2012, 8:41:51 PM2/18/12

to

Is that the best slur you can manage? You're seriously mentally ill
and should seek treatment.

John Larkin

unread,

Feb 18, 2012, 9:04:24 PM2/18/12

to

I've seen it happen, got bit by this *before* I noticed the note on
the datasheet. In the original National linear databook, the footnote
was microscopic type on ratty paper, practically invisible. A 1N4148
is absolutely not adequate to make your circuit safe. Are you willing
to design stuff that violates clear warnings like this, by about 2:1
or so? That's really scary. Hell, we even discussed this bug here
before. Sloppy, sloppy work.

But it's academic, since an LM339 won't work in your circuit at all.

What comparator did you use in the sim? Something "behavorial"
probably.

Fred Bloggs

unread,

Feb 18, 2012, 9:32:08 PM2/18/12

to

On Feb 18, 9:04 pm, John Larkin

<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 18 Feb 2012 18:08:17 -0700, Jim Thompson
>
>
>
>
>
> <To-Email-Use-The-Envelope-I...@On-My-Web-Site.com> wrote:
> >On Sat, 18 Feb 2012 17:01:26 -0800, John Larkin
> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>
> >>On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:
>

> probably.- Hide quoted text -
>
> - Show quoted text -

Just replace the diodes with Schottky's- that ought to be more than
safe.

John Larkin

unread,

Feb 18, 2012, 10:28:30 PM2/18/12

to

Maybe not. It takes a surprisingly small voltage and current to mess
up the National version. Why take the risk?

But a 339 won't work in that circuit anyhow.

John Larkin

unread,

Feb 18, 2012, 11:11:30 PM2/18/12

to

On Sat, 18 Feb 2012 15:28:37 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:
>
>>On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>>>
>>>>Perhaps like this...
>>>>
>>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>>
>>>>Choice of comparator is left as an exercise for the student... but
>>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>>;-)
>>>
>>> Idiot. An LM339 will go nuts from the negative input swing. What did
>>> you use to model it, an ideal comparator?
>>
>>LM339 behaves sensibly with inputs as low as -0.6V,
>
>Yes. Under most conditions. And only one input misbehaves even if
>you suck current out of it.
>
>>it'll need a pull-up on the output though,
>
>Yes.
>
> ...Jim Thompson

Hey, you're big on parts values. What value of pullup?

John Fields

unread,

Feb 19, 2012, 1:56:47 AM2/19/12

to

On Sat, 18 Feb 2012 18:04:24 -0800, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>I've seen it happen, got bit by this *before* I noticed the note on
>the datasheet. In the original National linear databook, the footnote
>was microscopic type on ratty paper, practically invisible.

---
So your inattention to detail is National's fault?

A poor workman blames his tools.

--
JF

John Fields

unread,

Feb 19, 2012, 2:04:06 AM2/19/12

to

---
Got anything quantitative instead of your usual pretense at sagacity?

--
JF

John Fields

unread,

Feb 19, 2012, 2:06:08 AM2/19/12

to

On Sat, 18 Feb 2012 20:11:30 -0800, John Larkin

<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Sat, 18 Feb 2012 15:28:37 -0700, Jim Thompson
><To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>
>>On 18 Feb 2012 20:45:40 GMT, Jasen Betts <ja...@xnet.co.nz> wrote:
>>
>>>On 2012-02-18, John Larkin <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>> On Sat, 18 Feb 2012 09:41:13 -0700, Jim Thompson
>>>>>
>>>>>Perhaps like this...
>>>>>
>>>>>http://www.analog-innovations.com/SED/ZeroCrossDiscussion_SED.pdf
>>>>>
>>>>>Choice of comparator is left as an exercise for the student... but
>>>>>even a lowly LM339 will do if you can stand the 300ns of extra delay
>>>>>;-)
>>>>
>>>> Idiot. An LM339 will go nuts from the negative input swing. What did
>>>> you use to model it, an ideal comparator?
>>>
>>>LM339 behaves sensibly with inputs as low as -0.6V,
>>
>>Yes. Under most conditions. And only one input misbehaves even if
>>you suck current out of it.
>>
>>>it'll need a pull-up on the output though,
>>
>>Yes.
>>
>> ...Jim Thompson
>
>
>Hey, you're big on parts values. What value of pullup?

---
Desperate, huh?

--
JF

Jan Panteltje

unread,

Feb 19, 2012, 3:35:37 AM2/19/12

to

On a sunny day (Sat, 18 Feb 2012 15:32:47 -0700) it happened Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote in
<7i90k7dpf4ltsmjvn...@4ax.com>:

>
>No. I'm working with a company right now where _everyone_ is younger
>than my children... with some even as young as my oldest grandchild.
>And they are peaches to work with. Seems their professors at RIT used
>some of my early circuit designs as teaching examples ;-)
>
>And it's no piddling company... they just shipped their BILLIONETH
>chip!

Googling "BILLIONTH chip shipped" gives:
Cypress.
Don't think Nvidia is your game.

Day before yesterday I watched Alien versus Predator 2,
and it reminded me of Thompson vsLarkin.
But I though, not a good analogy.
But then the gov nuked the city with the aliens in the end (great shot),
to prevent them spreading.
In fact the only good shot in the whole movie :-)
So then I was thinking of what Obanana is doing to you guys, and yes,
maybe the analogy holds ;-)
:-)

Do your customers read your quarrels?
You work for mil?
Now there is a ....
hehe
LOL

Jim Thompson

unread,

Feb 19, 2012, 10:28:46 AM2/19/12

to

Yes, John "Sick Puppy" Larkin certainly is.

John Larkin

unread,

Feb 19, 2012, 10:44:54 AM2/19/12

to

On Sun, 19 Feb 2012 08:28:46 -0700, Jim Thompson

Value?

John Larkin

unread,

Feb 19, 2012, 10:49:59 AM2/19/12

to

The data sheet says to not go below -0.3. That's not good enough for
you? Thompson is clamping it with a 1N4148, around twice the voltage
that National footnote 3 recommends.

Sound pretty risky to me. I only had this problem once, about 25 years
ago, and what I remember is that a silicon diode was not enough to
keep the comparator from getting goofy.

Hey, try it yourself and report back.

John Larkin

unread,

Feb 19, 2012, 10:51:18 AM2/19/12

to

On Sun, 19 Feb 2012 00:56:47 -0600, John Fields
<jfi...@austininstruments.com> wrote:

>On Sat, 18 Feb 2012 18:04:24 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>
>>I've seen it happen, got bit by this *before* I noticed the note on
>>the datasheet. In the original National linear databook, the footnote
>>was microscopic type on ratty paper, practically invisible.
>
>---
>So your inattention to detail is National's fault?
>

Thompson't inattention to detail - he screwed up twice here - is
certainly his own fault.