Author |
Message |
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Mon Dec 11, 2006 3:50 pm Post subject:
CMOS input protection Subject description: what value should the zener diodes have? |
|
|
hi all,
i built jürgen haibles wasp filter clone last week. and it sounds great. i think i will upload some sound samples next week.
i have one concern: the input signal goes straight into a 4069 which is powered from GND to 5V. i also have some oscillators with a hot output and i want to protect the CMOS input. i understand that i can achieve this using one zener diode from input to 5V and one from GND to input.
what value should they have? 5V or 2.5V? _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Mon Dec 11, 2006 11:37 pm Post subject:
|
|
|
and once again you where a light in the dark...
as you might recognize i am learning a lot these days.
i should use schottky BAT 85 diodes, i guess? _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue Dec 12, 2006 4:40 am Post subject:
|
|
|
fonik wrote: | i should use schottky BAT 85 diodes, i guess? |
Yes it should be schottky, and BAT 85 should be ok. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Mon Aug 20, 2018 10:29 am Post subject:
|
|
|
hello, do i understand it right, that the circuit shown in pic A protects for example a cd4040 powered on 5V against a 12V gate signal comming from another module?
edit: i created it on a breadboard using a 100k resistor and bat85 diodes and it seems it works (it is 5.2V on the out, does a cmos powered on 5V work with that or will this be enough to kill the chip?)
and should i use a diode before the resistor to keep negative voltage (like from an lfo) outside, too, or isn't that neccessary |
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Mon Aug 20, 2018 1:27 pm Post subject:
|
|
|
snother way of protection was to use a sturdy opamp that accepts a large differential voltage on its inputs that is way above the actual Vcc (i.e. 5V) it is powered from. _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Mon Aug 20, 2018 11:50 pm Post subject:
|
|
|
any hint on what opamp this would be? as i already fried some opamps in the past, i am a bit concerned
but if there is a less part counting alternative i am indeed interested |
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Tue Aug 21, 2018 2:14 am Post subject:
|
|
|
looks more complicated than it is. one 8pin IC and 4 resistors. _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Tue Aug 21, 2018 2:21 am Post subject:
|
|
|
thank you. from what i see there, it is much more place on the board and more parts. is there a remarkable benefit from this approach? since my test with the third diode in front kept negative voltages away, too |
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Tue Aug 21, 2018 2:46 am Post subject:
|
|
|
yes, it is one DIP8 and a resistor more. but as i said, you now can process anything, not just pulses. _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Tue Aug 21, 2018 2:59 am Post subject:
|
|
|
ok. thank you very much |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue Aug 21, 2018 6:22 am Post subject:
|
|
|
fonik wrote: | [...]you now can process anything, not just pulses. |
Could you explain that a bit please? as in .. why could the diode protection not process "anything" but just "pulses"? Not getting it ... _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Tue Aug 21, 2018 9:21 am Post subject:
|
|
|
from what i understand is that some (or all) circuits don't understand a rising voltage as a gate signal. at least my test i just did brings me to that conclusion. when i create a triangle wave form between 0 and 8 volt and feed this into the protection circuit it doesn't trigger the dinky taiko. however if i shorten the rise time there is some point at which it counts as a trigger signal again.
so, if the comparator circuit creates a nice trigger signal at one point, it would be the solution for a wider scaled input scenery. if i get all this correct |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue Aug 21, 2018 9:26 am Post subject:
|
|
|
Ah .. ok ... hysteresis added ... had assumed the CMOS input port would have done that. And was reasoning the other way around .. that the protection diodes would be better in passing "any signal" (as in that it will work for analog inputs too). _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
Posted: Tue Aug 21, 2018 10:04 am Post subject:
|
|
|
matthias, could you tell me where you found the lm311 symbol for LTspice (which i installed just today)?
all sources i found seem to be broken and I don't know the program well enough to create it by myself.
or, is there a general ressource site, for example to find bat85 and other parts? |
|
Back to top
|
|
|
ello
Joined: Aug 20, 2018 Posts: 10 Location: germany
|
|
Back to top
|
|
|
diophantine
Joined: May 10, 2011 Posts: 9 Location: San Diego, CA
|
Posted: Mon Jul 08, 2019 7:46 pm Post subject:
|
|
|
Sorry to bump, but it seems this thread has been bumped a few times!
What's the best way to handle such a circuit when you also need a pull-down (or pull-up) resistor on one of the CMOS inputs?
My first inclination was to do this:
But then I realized that the two resistors form a voltage divider, and when using 100k for both, 5V in will only give 2.5V to the CMOS pin (ignoring the effect of the diodes). Unfortunately this won't work with every chip I'm using...
Assuming I want to keep the input resistor at 100k (modular "standard"), should I change R11 to 1M or something to minimize the effect of the voltage divider?
Or should I add another resistor, like this?
Ideally I'd like to keep the part count as low as possible. For context: I'm basically gutting an old Ed-Lab 700 and replacing all the circuits with new equivalent circuits that won't be fried by modular signal levels. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue Jul 09, 2019 11:11 am Post subject:
|
|
|
You can add the pull up/down resistor on the other end .. teh K / P4 point _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
diophantine
Joined: May 10, 2011 Posts: 9 Location: San Diego, CA
|
Posted: Tue Jul 09, 2019 3:31 pm Post subject:
|
|
|
Blue Hell wrote: | You can add the pull up/down resistor on the other end .. teh K / P4 point |
Thanks, I gave that consideration at some point & then forgot about it... I think too much reading of datasheets & application notes that always put the pull-down closest to the pin.
But after further thought it should work fine as long as the module feeding it has an output resistor (usually always the case), and that value is usually pretty small so the voltage divider effect should be minimal.
Back to KiCad! Thanks! |
|
Back to top
|
|
|
|