CMOS short circuit protection question

Thread Starter

SilverFox33

Joined Sep 20, 2021
3
Designing a battery operated circuit that uses active shutter 3D glasses. So far, I'm able to drive them directly from a CMOS chip - a CD4060. They like 18v dc so there are some advantages to CD4000 series (vs 74HC series) circuits.

The glasses come with a previously wired 3.5mm plug.

I'm considering the situation where the plug gets pulled excessively, or gets frayed by being stepped on, chewed by the pet cat, or otherwise messed up.

I've seen this happen a lot with headphones - the plug and the wire going into it can get abused. When this happens, of course the headphones stop functioning - but when driven from the output of an audio headphone amplifier, headphone-out of a laptop, etc, usually the only failure is the headphones. A short in the wire or plug doesn't kill the amp or laptop (usually).

But what about a CMOS chip? Will a short to ground kill a CD4060? If it can, is there a standard method for output protection? A short-circuit tolerant buffer or inverter? Since I'm running off of a battery, anything that hogs current when it isn't protecting won't be a good idea. Spec sheets for the CD4000 series talk about short circuit output current, but they appear to max out at an equivalent R of about 160 ohms - that's for drive calculations, I think - not for true output short circuit protection.

Any suggestions?

Sorry for being long-winded here. To be succinct: Suggestion for low-current CD4000 series output short circuit protection?
 

dl324

Joined Mar 30, 2015
16,846
Welcome to AAC!

This is from an RCA COSMOS databook:
1632240264000.png
The 200mW maximum dissipation is for the whole package.

For an indefinite short, power dissipation can be a problem at higher operating voltages and you are operating outside of the recommended maximum for CD4060 (RCA):
1632240542041.png

Will the LCDs tolerate any series resistance?
 

crutschow

Joined Mar 14, 2008
34,285
You could likely put a resistor in series with the output to limit the short circuit current without significantly affecting the voltage to the load.
What is its output load?
 

Thread Starter

SilverFox33

Joined Sep 20, 2021
3
Welcome to AAC!

This is from an RCA COSMOS databook:
View attachment 248507
The 200mW maximum dissipation is for the whole package.

For an indefinite short, power dissipation can be a problem at higher operating voltages and you are operating outside of the recommended maximum for CD4060 (RCA):
View attachment 248512

Will the LCDs tolerate any series resistance?
Dennis -

I haven't seen any RCA COSMOS parts in years. These days, all I've been getting are Texas Instruments. I've seen some OnSemi and Nexperia but TI seems to be pretty common. The max supply V with TI is 20v - here it the pic from their pdf:TI4060databook.jpg
Measuring the LCD shutter glasses impedance (resistance only?) to see if it can tolerate a series R is a good idea. I have not measured that R yet but will do so soon. I think the way to do that would be to put an ac current meter in series with the glasses while I feed them a 50% duty cycle square wave.

However, with a series R there will be some power wasted in the resistor, which I am trying to avoid. I guess what I'm looking for is a circuit or a component that draws no power under normal operating conditions. Then if there is a short, it comes into action to prevent frying of the 4060.

A fuse would certainly work here, but they are huge in comparison to my other board components (I can get the CD4060 in a TSSOP package, and I've found my crystal in a GSX49-4 package, which is 4.3mm). And of course if the fuse blew, it would need to be replaced once the fault was cleared.

Maybe this discussion stimulates some other ideas or approaches?
 
Last edited by a moderator:

Ian0

Joined Aug 7, 2020
9,680
Texas CD4060s are rated for 20V maximum supply voltage, so 18V won't be a problem.
As the supply voltage is increased the effective output resistance decreases, so damage due to short circuit is much more likely at 18V. At 5V it will probably survive indefinitely.
If a series resistor won't cause problems with the circuit, it's by far the easiest and cheapest solution.
 

dl324

Joined Mar 30, 2015
16,846
I haven't seen any RCA COSMOS parts in years. These days, all I've been getting are Texas Instruments. I've seen some OnSemi and Nexperia but TI seems to be pretty common.
TI acquired RCA and many of their CD4xxx datasheets are just poor copies of RCA datasheets. The datasheet you referenced is from Harris, which TI also acquired.

If you operate at 18V, you'd better make certain you don't have any spikes on the power supply that can cause the devices to latch up. That will short the power supply until you remove power, or the IC fails opened. If it doesn't open before the battery runs down, you could have a fire on your hands.

Measuring the LCD shutter glasses impedance (resistance only?) to see if it can tolerate a series R is a good idea. I have not measured that R yet but will do so soon. I think the way to do that would be to put an ac current meter in series with the glasses while I feed them a 50% duty cycle square wave.

However, with a series R there will be some power wasted in the resistor, which I am trying to avoid. I guess what I'm looking for is a circuit or a component that draws no power under normal operating conditions.
It's been awhile since I worked with liquid crystals from a driving perspective and don't remember the particulars. I think LCDs will appear as a capacitive load, so power dissipation in the resistor won't be significant. Power dissipation will depend on the switching frequency.

EDIT: Use shorter sentences.

I wouldn't use a meter to measure current. I'd use a scope to measure the voltage across a resistor and calculate current.
 
Last edited:

sparky 1

Joined Nov 3, 2018
756
Harris founded 1895 today is L3 but back then the data book they released years of intelligence which reads very well but full history is disproportionate makes us wonder how and why some component's application notes were not as verbose.
This can be puzzling we can see the excellant technical writing such as the old data handbooks. So it feels like a cliff hanger when you study something like the 14 stage ripple counter with oscillator. It is what Harris/Intersil wanted to disclose.

I recall asking an L3 trouble shooter a question, he said "Uh... but there is a few problems and I gotta go"
We just respect that they are not in a position to discuss the details, such is the business.

Many years later there are gaps which can be accompanied by emotions, like a memorial to components that served honorable purpose at one time even though we understand semiconductors some application notes might not be available to the public even though there is nothing secret the military prowess of the writers does not allow them to disclose, this might be cold war related or some other headache they don't want to discuss.
 
Last edited:

DickCappels

Joined Aug 21, 2008
10,152
I will volunteer a not-to-creative potential solution. Put a low current two terminal current source in series with the Vdd supply. The current source should be set to a high enough current that is will not regulate (will be more like a voltage source) in the normal range of operating current of your circuit, but set low enough so that when an output of a device is shorted, the current source limits the current to a safe level.

Wendy has been exploring two terminal current sources lately and maybe you will get an idea from her thread.
https://forum.allaboutcircuits.com/threads/making-a-2-terminal-current-regulator.177561/
 

olphart

Joined Sep 22, 2012
114
I've thermally coupled a PTC to a MOSFET to open the gate on FET over current. Shortens the lag of the PTC just acting on over current when inline with power. Left as inline, FET went bad before PTC opened up.
 
Top