Maker Pro
Maker Pro

Strength of CD4000 substrate diodes

J

Joerg

Jan 1, 1970
0
Usually I am easy on them. This time I need to jam it pretty good and
have to drive a few CD40106 inputs below GND at 1mA-2mA. TI's datasheet
says 10mA is the abs max. Does anyone remember a manufacturer that might
spec less than 10mA?
 
J

Jim Thompson

Jan 1, 1970
0
Usually I am easy on them. This time I need to jam it pretty good and
have to drive a few CD40106 inputs below GND at 1mA-2mA. TI's datasheet
says 10mA is the abs max. Does anyone remember a manufacturer that might
spec less than 10mA?

Joerg! Joerg! Joerg! Joerg! ;-)

...Jim Thompson
 
J

Joerg

Jan 1, 1970
0
Jim said:
Joerg! Joerg! Joerg! Joerg! ;-)

I know, I know. The topper was a design (no, not by me) where a whole uC
plus periphery was supplied via one (!) port pin substrate diode. Guess
they needed to save that one cent for a rectifier diode. Now that was a
real white knuckle design.
 
J

Jim Thompson

Jan 1, 1970
0
I know, I know. The topper was a design (no, not by me) where a whole uC
plus periphery was supplied via one (!) port pin substrate diode. Guess
they needed to save that one cent for a rectifier diode. Now that was a
real white knuckle design.

What is your concern, finding a CD40106 that can't take 10mA, or are
you looking for a preventative measure?

Post your schematic ;-)

...Jim Thompson
 
J

John Larkin

Jan 1, 1970
0
Usually I am easy on them. This time I need to jam it pretty good and
have to drive a few CD40106 inputs below GND at 1mA-2mA. TI's datasheet
says 10mA is the abs max. Does anyone remember a manufacturer that might
spec less than 10mA?

The old A series parts would SCR latch if you looked at them wrong.
Pulling an input below ground would take out a power supply or some
wirebonds. A few mA should be OK for a B-series part.

Can you use an HC40106? They're a lot better.

John
 
Z

Zak

Jan 1, 1970
0
Joerg said:
I know, I know. The topper was a design (no, not by me) where a whole uC
plus periphery was supplied via one (!) port pin substrate diode. Guess
they needed to save that one cent for a rectifier diode. Now that was a
real white knuckle design.

I 'designed by acident' a sync amp. The scenario: monitor wants
composite sync at TTL level, computer delivers sync at video signal
level. Monitor input is at 0 volts, no pull up resistor.

Thus, I built a little amp with a CMOS inverter. This worked but the
extra power supply bothered me. Then, I switch off the power, but the
thing keeps working. Hmm, there was no voltage on teh input and teh sync
output is insufficient as well.

I decide to use just a decoupling cap, no power connection. Doesn't work.

I take the decoupling cap off and bing the thing works.

Must be the most ugly analog CMOS amp in existence. I suppose the
monitor outputs some pull-up voltage whenever it detects a signal on the
sync input. Never tried to figure it out; it worked and it was a one-off.


Thomas
 
G

Genome

Jan 1, 1970
0
Jim Thompson said:
What is your concern, finding a CD40106 that can't take 10mA, or are
you looking for a preventative measure?

Post your schematic ;-)

...Jim Thompson
--

I'm not sure I get this stuff......

Is that.......

Joerg! Joerg! Joerg! Joerg! ;-)

In a cheering sort of 'You Are The Man' sort of way?

Or is it a

Joerg! Joerg! Joerg! Joerg! ;-)

In a kind of 'You are such a child, let me explain it to you' sort of way?

DNA
 
J

Jim Thompson

Jan 1, 1970
0
I'm not sure I get this stuff......

Is that.......

Joerg! Joerg! Joerg! Joerg! ;-)

In a cheering sort of 'You Are The Man' sort of way?

Or is it a

Joerg! Joerg! Joerg! Joerg! ;-)

In a kind of 'You are such a child, let me explain it to you' sort of way?

DNA

The latter ;-)

...Jim Thompson
 
G

Genome

Jan 1, 1970
0
Jim Thompson said:
The latter ;-)

...Jim Thompson
--

Oh.... OK.......

There you go Jeorg, e-mail Jim, but don't be a bottom boy, Texans have a
problem with that stuff.

Between yourselves.

On a scale of 1 to 10 where 1 is not very cuddly and 10 is 'We are trying to
explain to the wives about the lesbien type stuff'....

Something else

DNA
 
J

Joerg

Jan 1, 1970
0
Jim said:
What is your concern, finding a CD40106 that can't take 10mA, or are
you looking for a preventative measure?

No, just wondering if there is a brand that specs much less. You know
how that goes, purchasing finds a cheaper source and then it's like
popcorn at final board test.

Post your schematic ;-)

Since I wrote the contract myself it doesn't specify the gun type with
which I'd be shot. But it kind of says that I would be :-(
 
J

Jim Thompson

Jan 1, 1970
0
No, just wondering if there is a brand that specs much less. You know
how that goes, purchasing finds a cheaper source and then it's like
popcorn at final board test.



Since I wrote the contract myself it doesn't specify the gun type with
which I'd be shot. But it kind of says that I would be :-(

Can't you show us the source impedance equivalent? Then we can solve
it for you ;-)

...Jim Thompson
 
J

Joerg

Jan 1, 1970
0
Zak said:
I 'designed by acident' a sync amp. The scenario: monitor wants
composite sync at TTL level, computer delivers sync at video signal
level. Monitor input is at 0 volts, no pull up resistor.

Thus, I built a little amp with a CMOS inverter. This worked but the
extra power supply bothered me. Then, I switch off the power, but the
thing keeps working. Hmm, there was no voltage on teh input and teh sync
output is insufficient as well.

I decide to use just a decoupling cap, no power connection. Doesn't work.

I take the decoupling cap off and bing the thing works.

Must be the most ugly analog CMOS amp in existence. I suppose the
monitor outputs some pull-up voltage whenever it detects a signal on the
sync input. Never tried to figure it out; it worked and it was a one-off.

The uC I mentioned was a lot scarier. AFAIR the supply came from a
higher voltage AC source via a resistor, could even have been 117V,
don't remember. Smack dab into the port pin. Anyhow, the "regulation" so
that it doesn't exceed abs max worked something like this: Bypass cap
was the minimum they could get away with but large enough so the boot
procedure could get a hold. Then the uC executed lots of dummy code just
to burn a fairly consistent amount of milliwatts. Supposedly the dummy
code was adjusted on the fly so that the MIPS levels kind of balanced
out slow workload phases. Woe to the firmware programmer who didn't know
this.

The way I found out was that I had to probe the oscillator. The probe's
capacitance stalled it for a brief time. KAPOOF! Big crater in the uC
and a puff of smoke wafting off.
 
J

Joerg

Jan 1, 1970
0
John said:
The old A series parts would SCR latch if you looked at them wrong.
Pulling an input below ground would take out a power supply or some
wirebonds. A few mA should be OK for a B-series part.

Yes, I remember the first generation chips. Those were the ones that
gave the 4000 series such a bad rap and scared a lot of people towards TTL.

I don't remember if the 74C series was as fickle. I never had one die on
me but others weren't very happy with them.

Can you use an HC40106? They're a lot better.

Unfortunately not here. Got to work with 12V.
 
M

Mike Monett

Jan 1, 1970
0
Joerg said:
The uC I mentioned was a lot scarier. AFAIR the supply came from a
higher voltage AC source via a resistor, could even have been
117V, don't remember. Smack dab into the port pin. Anyhow, the
"regulation" so that it doesn't exceed abs max worked something
like this: Bypass cap was the minimum they could get away with but
large enough so the boot procedure could get a hold. Then the uC
executed lots of dummy code just to burn a fairly consistent
amount of milliwatts. Supposedly the dummy code was adjusted on
the fly so that the MIPS levels kind of balanced out slow workload
phases. Woe to the firmware programmer who didn't know this.
The way I found out was that I had to probe the oscillator. The
probe's capacitance stalled it for a brief time. KAPOOF! Big
crater in the uC and a puff of smoke wafting off.
Regards, Joerg

Joerg, how could it blow a hole in the IC? That takes more than a
few watts.

If the uC only needed milliwatts to regulate the voltage, where did
the extra power come from? Especially since the bypass cap was the
minimum they could get away with.

There has to be some other explanation.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm
 
J

Joerg

Jan 1, 1970
0
Jim said:
Can't you show us the source impedance equivalent? Then we can solve
it for you ;-)

It's a 10uF cap on the input that is whipped around. During one phase it
is pulled to -12V via a 5K resistor. I could hang a largish resistor
between the cap and the CD input but I'd rather not.

It should be fine as long as the CD40106 all have the same specs.
 
J

Joerg

Jan 1, 1970
0
Mike said:
Joerg, how could it blow a hole in the IC? That takes more than a
few watts.

If the uC only needed milliwatts to regulate the voltage, where did
the extra power come from? Especially since the bypass cap was the
minimum they could get away with.

There has to be some other explanation.

No idea. All I know is that I had a little crater in there. The bypass
wasn't that small since there was a whole lot of other stuff that rode
on this VCC rail. It was an electrolytic and they didn't spring an extra
cent for a ceramic in parallel. Later I looked at the supply rail on a
working unit and the ripple on there was nauseating.

What possibly happened is that some PWM stopped, shedding the whole
load. Then the voltage crept up. This was a really old uC back from the
80's. The AC drop resistor was a big fella, maybe a watt or so. It got
quite toasty.
 
W

Winfield Hill

Jan 1, 1970
0
Joerg wrote...
I know, I know. The topper was a design (no, not by me) where a whole

I imagine all modern parts can take at least 20mA, or more, whatever
spec they might put out. It's easy to test representative production
parts to test reality... SCR supply breakdown is easy to observe.
A limiting resistor can protect your test sample for further analysis.
 
J

Joerg

Jan 1, 1970
0
Winfield said:
Joerg wrote...



I imagine all modern parts can take at least 20mA, or more, whatever
spec they might put out. It's easy to test representative production
parts to test reality... SCR supply breakdown is easy to observe.
A limiting resistor can protect your test sample for further analysis.
Yes, Win, I did such tests a while ago. They could stomach more than
20mA. But that always leaves the nagging uncertainty about the next
batch or chips from another manufacturer. Bottomline all designs here
need to comply with respect to the datasheet limits. Except, of course,
when doing scientific stuff like avalanching.

BTW, just read a laser article (Elsevier) from Rowland. Vollmer,
Fischer: "Frequency-domain displacement sensing with a fiber
ring-resonator containing a variable gap". Wow, interesting things that
you guys are doing over there.
 
A

Andy

Jan 1, 1970
0
Joerg said:
Yes, I remember the first generation chips. Those were the ones that
gave the 4000 series such a bad rap and scared a lot of people towards TTL.

I don't remember if the 74C series was as fickle. I never had one die on
me but others weren't very happy with them.



Unfortunately not here. Got to work with 12V.

The CD40106BC from National does not have this maximum current specified
in the datasheet at all. Ok, it's marked "discontinued", but it's still
in stock at some places.

MM74C14 from Fairchild does not have it specified neither.

-- Andy
 
T

Tony Williams

Jan 1, 1970
0
Joerg said:
It's a 10uF cap on the input that is whipped around. During one
phase it is pulled to -12V via a 5K resistor. I could hang a
largish resistor between the cap and the CD input but I'd rather
not.

Allowing for an external Schottky diode clamp on
the pcb layout would seem prudent.

App notes on 4000-series astables and monostables
may give some insight into the maximum recommended
negative input current whilst working.
 
Top