[sdiy] Limiting current through 74-series 4051

[Mark Rivera]

Thanks for this. I had a 'magic smoke' accident last night that left
me a little confused about how my circuit actually operated. After
reading some more based on what you wrote, I discovered I had made a
poor selection as far as the op amp goes... Needed one that went
rail-to-rail on input and output, but the one I had was only
rail-to-rail on output.

On Sun, Nov 13, 2011 at 4:39 PM, Magnus Danielson
<magnus at rubidium.dyndns.org> wrote:
> On 11/13/2011 11:20 PM, Mark Rivera wrote:
>>
>> Hi all,
>>
>> I need to limit the current going through the switches of a 74-series
>> 4051 (0V Gnd/Vee, 5V Vcc). I am using the 4051 as a demux.. 8 DC
>> inputs to one output. The output is feeding into a single-supply,
>> rail-to-rail (0-5V) opamp with voltage-follower configuration.
>>
>> The input sources (pots as voltage dividers) can be anywhere between
>> Gnd and Vcc.
>>
>>
>> Is there any difference between these two setups:
>>
>> 1. voltage sources ->  8 resistors ->  inputs ->  output ->  opamp in+
>>
>> 2. voltage sources ->  inputs ->  output ->  1 resistor ->  opamp in+
>>
>>
>> Theoretically, I think there's not, but I wonder about having lots of
>> current available at the inputs versus having restricted current at
>> the inputs. Does it affect the lifetime or performance of the 4051?
>>
>> (Ultimately, opamp output goes into an ADC.)
>
> Wait... does your op-amp loads your circuit that much????
>
> Except for capacitor loading I don't see much needs of current flowing there
> at all.
>
> I am sure that your op-amp circuit would behave better if there was a little
> connection to ground all the time, in between the channels, you don't want
> it floating around too long at least.
>
> There is a benefit in locating the resistors on the source side if you also
> terminate the signals towards a virtual ground. That way the source
> impedance of the CMOS analog muxes has very low variance with the signal,
> especially if the resistors is 10-100 times the switch impedance... which is
> about 100-200 Ohms (don't recall numbers, but it does vary with the supply
> current and voltage of the terminals). Locating the switches close to 0 V
> during opeating time and making their internal impedance of less importance
> you have moved out quite a bit of non-linear problems.
>
> But then I think others have better experience than me in working these
> issues.
>
> Cheers,
> Magnus
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>