[sdiy] Minimum parts count switch de-bouncer.

[wayfar]

If the cap is in a low-pass configuration - IE Integrator, you will 
already have an impedance (resistor) on the input voltage -  for 
adjusting the cutoff frequency,

http://en.wikipedia.org/wiki/RC_circuit



Harry Bissell Jr wrote:

>As I mentioned in the screwdriver saga...
>you might get 20A the first time but repeated
>discharges will be degrading the cap.
>
>I hear your backyard mechanic music... but you need
>to become a Harley Davidson engineer
>
>"At Harley-Davidson... if we need something
>stronger, we make it bigger. If that makes it ugly,
>we chrome it..."
>
>I'd scale the RC values... much bigger resistor and
>much smaller capacitor. That will cut the wear
>on the switch.  Even if you tossed 100 ohms in series
>with the switch, you would never lose sleep over
>a cap failure !
>
>(now the heat of the discharge is shared between
>the 100 ohms and the .01 ohms or so ESR)
>
>H^) harry
>
>--- Tim Daugard <daugard at sprintmail.com> wrote:
>
>  
>
>>From: "Harry Bissell Jr" <harrybissell at prodigy.net>
>>
>>    
>>
>>>One point NOT being made in many of these
>>>proposed circuits... is that the capacitor
>>>can draw unlimited current when the switch closes.
>>>      
>>>
>>In my circuit the problem is not the charge current,
>>but the discharge
>>current.
>>
>>    
>>
>>>Bigger the cap value, the higher the current.
>>>      
>>>
>>Aggreed
>>
>>    
>>
>>>Enough current to eventually degrade of destroy
>>>      
>>>
>>the
>>    
>>
>>>switch in many cases...
>>>      
>>>
>>Not sure which will go first, the cap or the switch.
>>
>>    
>>
>>>Kind of the the moral and legal equivalent of
>>>discharging the capacitor with a screwdriver
>>>(hi Bob ;^)
>>>      
>>>
>>Absolutely
>>
>>    
>>
>>>There should be a small resistor in series with
>>>      
>>>
>>the
>>    
>>
>>>switch, to limit that peak current. Just assume
>>>      
>>>
>>the
>>    
>>
>>>cap is a short, and Ohms' Law will tell you the
>>>      
>>>
>>peak
>>    
>>
>>>current.
>>>      
>>>
>>The caps I use are Xicon which I have to assume is
>>smoe cheap brand labeled
>>specifically for Mouser, a company search turns up a
>>distributor in Texas
>>and all links for data lead back to Mouser.
>>
>>Best I can figure from the Mouser data sheet is an
>>internal resistance as a
>>current source of about 6/10s ohms. With a 12 Volt
>>charging source, a fully
>>charged cap will deliever a peak current of about 20
>>Amps.
>>
>>    
>>
>>>No resistor will also rock the power supply rails,
>>>bigger the cap, bigger the problem.
>>>      
>>>
>>My cap was selected for the RC recharge time.
>>
>>    
>>
>>>You may want SOME substantial current to keep the
>>>switch contact clean, but not enough to wear it
>>>      
>>>
>>out.
>>
>>I wonder what the current rating is on the cheap
>>RatShak switch is.
>>
>>    
>>
>>>H^) harry
>>>      
>>>
>>Who has made me spend 30 minutes researching this to
>>come up with the
>>conclusions:
>>- I'm using the wrong cap
>>- I may be stressing the switch
>>- I can't find enough data to narrow down the
>>failure modes and effects
>>further
>>   --- However, what I did find says the cap will
>>fail open, meaning the
>>circuit still works, it just stops being debounced.
>>
>>And so my circuit continues to live as a backyard
>>mechanic's - "It works
>>but who knows for how long." When it fails, I'll
>>reengineer. Till then . .
>>.
>>
>>Tim Daugard
>>AG4GZ 30.4078N 86.6227W Alt: 12 feet above MSL
>>http://home.sprintmail.com/~daugard/synth.htm
>>
>>My version of the circuit in question:
>>http://home.earthlink.net/~synthfred/h_ssssg1.htm
>>
>>
>>
>>
>>    
>>
>
>
>  
>