High impedance pins are susceptible to leakage?

Hello,

I was about to begin laying out a part and I read the layout considerations and I don't know what they are talking about, can anyone tell me what this means

"Due to the high impedances on the SD, VDD, and GATE pins, these pins are s usceptible to leakages to ground. For example, a leakage to ground on SD wi ll activate the shutdown state if greater than 1.6µA. Providing adequate spacing away from grounded traces and adding conformal coating on exposed p ins lowers the risk that leakage current will interrupt system operation."

What does it mean that a high impedance pin is susceptible to leakage? A h igh impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from groun ded traces?" I shouldn't route these pins close to ground?

this came from this datasheet, pg 16

formatting link

much thanks!

Reply to
panfilero
Loading thread data ...

ns and I don't know what they are talking about, can anyone tell me what th is means

susceptible to leakages to ground. For example, a leakage to ground on SD will activate the shutdown state if greater than 1.6µA. Providing adequat e spacing away from grounded traces and adding conformal coating on exposed pins lowers the risk that leakage current will interrupt system operation. "

 A high impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from  grounded traces?"  I shouldn't route these pins close to ground?

The pin is at some voltage but with a high source impedance. (So a voltage source with a soucre impedance of 100Meg ohm (or something)) If you have a ground trace near it, then you may get leakage currents between the pin and ground... becasue of the high source imepadance the pin voltage drops... and unwanted things happen.

(You should think not only about ground traces but any trace at a different voltage level... I guess ground is the most common.) (NPI)

George H.

Reply to
George Herold

I don't know what they are talking about, can anyone tell me what this means

susceptible to leakages to ground. For example, a leakage to ground on SD will activate the shutdown state if greater than 1.6µA. Providing adequate spacing away from grounded traces and adding conformal coating on exposed pins lowers the risk that leakage current will interrupt system operation."

impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from grounded traces?" I shouldn't route these pins close to ground?

Pretty weird (and expensive) part.

The transistor collector will likely leak more than the PC board.

A clean, dry board will leak picoamps, or femtoamps, between pads.

formatting link

That little PC board was deliberately crudded with fingerprints and rosin flux, and is pinning the meter on the 1e14 ohm range. The biggest leakage hazard is usually from inadequately washed water-soluble flux trapped under the part itself.

--
John Larkin         Highland Technology, Inc 

jlarkin at highlandtechnology dot com 
 Click to see the full signature
Reply to
John Larkin

and I don't know what they are talking about, can anyone tell me what this means

susceptible to leakages to ground. For example, a leakage to ground on SD will activate the shutdown state if greater than 1.6µA. Providing adequate spacing away from grounded traces and adding conformal coating on exposed pins lowers the risk that leakage current will interrupt system operation."

impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from grounded traces?" I shouldn't route these pins close to ground?

The crud seems to make dirty boards hydrophilic or something.. breath on them and you get will get a big decrease in resistance.

--sp

Reply to
Spehro Pefhany

I was involved in a product where we were trying to measure the total amount of moisture inside of a (supposedly) hermetic enclosure by measuring the humidity and then calculating the temperature at which water would condense.

It fell on its face, badly -- as soon as the temperature went up, we'd get an "excess water" alarm. We never dug into it because it was obviously a failed approach, but I have guesses. The system was about 25 pounds of circuit boards, optics, and wiring tightly packed into a ball. I'm almost certain that water was absorbing (or adsorbing) on all of the organic surfaces as the temperature went down and releasing it as it went up, much as a saturated salt will hold the atmosphere around it at a constant humidity.

So, yes -- I can believe in hydrophilic crud and changing conductances.

--
My liberal friends think I'm a conservative kook. 
My conservative friends think I'm a liberal kook. 
 Click to see the full signature
Reply to
Tim Wescott

Maybe it was just condensation onto the cold boards.

--
John Larkin         Highland Technology, Inc 

jlarkin at highlandtechnology dot com 
 Click to see the full signature
Reply to
John Larkin

e
d

ing

g

e to

n
25

ttdesign.com- Hide quoted text -

If you talk with ultra high vacuum people, they will tell you there is a mono-layer of water coating everything. (Hence the bake-out to

150C) I'm not sure if it's true or not... but it gives you a different perspective on the world.

George H.

Reply to
George Herold

I once encountered an interesting leakage-related problem... a "haunted" Macintosh II, which spontaneously switched itself on!

The Mac II has a "soft" power-on circuit, controlled by a signal wire in the ADB (Apple Desktop Bus) cable that goes out to the mouse and keyboard. There's a "power on" key on the keyboard. The line was powered from the on-board backup battery, through a very-high-value resistor... pulling it down to ground triggers a detection circuit (presumably CMOS) and sends an "on" pulse to the system's main power supply.

My Mac II switched itself on for no obvious reason a couple of times, without human intervention. The first time this happened I wasn't able to figure out why; I shut down the system and it stayed off.

The second time it happened I was present in the next room. I immediately shut the system down, and it started right back up again. I started unplugging things from the ADB cables, and when I disconnected the trackball, I was able to shut down the system and it stayed off.

I realized that both times this had happened, the Mac had been sitting in a cold room overnight (the house we were in had really lousy heating), and had started up on its own shortly after we had opened the door to that room and allowed warmer (and moister) air from the rest of the house to circulate. This had led to a tiny trace of condensation forming on the surfaces in the room, including those inside the Mac components. I inspected the trackball, and saw that the ADB jack pins were very close together on the PC board. I reconnected the trackball... Mac stayed off... touched the pins with my fingers... the Mac booted!

Apparently there was a bit of surface contamination on the trackball PC board... enough that when it was hit with a bit of condensation from warm moist air it created a conductive path between the ADB power-on and ground pins that was sufficient to "pull down" the high-impedance detector line and trigger the power-on detector.

I cleaned the PC board, dabbed the pins with something to seal them (bummed some clear nail polish from my wife, IIRC), closed everything up, and the problem was gone for good.

So, yes... a clean dry board's leakage is not likely to be a problem for anything other than specialized high-Z circuits. But, you can't always count on the board being clean and dry in the real world. Hence, the datasheet's recommendation to use conformal coating on the pins and traces involved... clean the board well, coat the high-Z points to keep moisture away, and you'll have less trouble.

--
Dave Platt                                    AE6EO 
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior 
 Click to see the full signature
Reply to
Dave Platt

Software

formatting link
Hide quoted text -

I did a bunch of library research on that stuff long ago for an air ionizer project. The equilibrium thickness of the water layer depends on the surface energy and the humidity. For metal, iirc you reach 1 full monolayer at about 40% RH, and it rises steeply above there, to infinity at 100% RH.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

yone

ns

ound

viding

ting

ge?

lose to

est

x

on

ut 25

l.

he

ent

.

scottdesign.com-Hide quoted text -

t -

Cool! Is it only metals or do dielectrics get a layer too? (I guess I was wondering what is the bonding mechanism?)

Molecular sieve certainly gobbles up water. (but that may be a different mechanism.)

George H.

Reply to
George Herold

Water chemically bonded with clay doesn't scoot off until well over

400°C. I guess the clay dessicant packets can't be recycled like the silica gel ones can, unless you open them and repackage the material.
Reply to
Spehro Pefhany

Hmm, As a grad student we used Linde molecular sieve (I can't recall the size maybe 13X?) for cryo-pumping. The prescription was to heat it to 250C to drive off the water. I made this little furnace from fire brick and an old toaster.

George H.

Reply to
George Herold

and I don't know what they are talking about, can anyone tell me what this means

susceptible to leakages to ground. For example, a leakage to ground on SD will activate the shutdown state if greater than 1.6µA. Providing adequate spacing away from grounded traces and adding conformal coating on exposed pins lowers the risk that leakage current will interrupt system operation."

impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from grounded traces?" I shouldn't route these pins close to ground?

For example...

It says in the data sheet that the SD pin is pulled up internally with a

1.6uA current source. If you connect this to an open collector to control it, you need to be sure that your transistor leakage when off isn't sufficient to pull it down.

I guess possible PCB leakage due to, for example, accidental surface contamination needs to be considered too.

It doesn't mean that the pins themselves leak.

Cheers

--
Syd
Reply to
Syd Rumpo

and I don't know what they are talking about, can anyone tell me what this means

susceptible to leakages to ground. For example, a leakage to ground on SD will activate the shutdown state if greater than 1.6µA. Providing adequate spacing away from grounded traces and adding conformal coating on exposed pins lowers the risk that leakage current will interrupt system operation."

high impedance pin... like the input to an op-amp, I would think would have very little leakage... and what do they mean, "adequate spacing from grounded traces?" I shouldn't route these pins close to ground?

Yep. I've designed in (at customer request) 100nA pull-ups :-( That's treacherous.

But then, one of my chips runs on 800nA total ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

one

s

und

iding

ing

e?

ose to

st

on

t 25

.
e

nt

cottdesign.com-Hide quoted text -

formatting link

I did an experiment based on their numbers as a final year chemistry undergraduate, and it stuck in my mind. BET stands for Stephen Brunauer, Paul Hugh Emmett, and Edward Teller, with the Edward Teller being the one who went on to work on the H-bomb.

The mono-layer of water atoms is stuck to surface by rather stronger bonds than those that bond subsequent layers of water atoms to underlying layers of water.

Bill Sloman, Sydney (but in New Zealand at the moment).

Reply to
Bill Sloman

Software

formatting link
quoted text -

I'm pretty sure that it depends on the surface energy--hydrophilic surfaces have a thicker equilibrium layer as long as it's below a monolayer or so. Above a couple of monolayers it's probably a universal curve. I looked, but I don't seem to have the papers in soft copy--this was way back when I was a postdoc, circa 1988.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

operation."

I had to clean, blow, clean, sometimes many times, to get rid of the water based flux. I forget the impedance of the circuit right now, but it was high. It was under opamps and caps. That was with at least 40 boards, and the problem of absorption of certain caps. Finally had to change to another cap.

Greg

Reply to
gregz

nyone

ins

round

oviding

ating

t

age?

close to

gest

ux

th on

d

out 25

ll.

the

went

s.

Neat, that's pretty cool Bill, I went on to this,

formatting link
(You can get lost for hours in wikipedia... )

As an aside, let me say I welcome your 'science' presence here on SED.

George H.

Reply to
George Herold

anyone

pins

ground

roviding

oating

pt

kage?

k
e

close to

d

ggest

lux

th on

l

'd

bout 25

all.

the

went

es..

s

ds

l
s

t -

Thanks Phil, No problem, I gather it's related to hydrogen bonding... dipole dipole interactions... all very confusing to me, since I've never studied it.

(In a metal you can imagine a dipole /induced dipole interaction.)

George H.

Reply to
George Herold

George Herold schrieb:

Hello,

somebody said: God made solids, but surfaces are a work of the devil!

Bye

Reply to
Uwe Hercksen

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.