[sdiy] Voltage Feedback Resistors and Circuit Stability

[Ian Fritz]

This a convention that goes way back to the early days.  Of course, having a high enough input impedance to avoid excessive loading has always been a consideration. With 1k output resistances, two outputs can usually be mixed directly, which is quite handy sometimes. So if you want no more than 1% loading, 100k input impedance is reasonable.  But if noise from 100k resistors bothers you, then you should definitely go lower. Myself, I don’t notice it, and would prefer having the extra power savings.

> On Nov 13, 2020, at 1:56 PM, Mattias Rickardsson <mr at analogue.org> wrote:
> 
> Out of curiosity, what is your reasoning behind 100k being a good compromise? 
> I thought that the widespread use of 100k was an unfortunate consequence of having 100k input impedance as standard in eurorack. :-)
> 
> A bog-standard TL07x draws 1.4 mA per op-amp and has a noise around 18 nV/rtHz.
> 
> - Any resistor above 22k is noisier, so let's not go higher than 10k.

Sure, if noise is your only criterion.

> - Assuming a 5 V signal through 10k, the current is only 0.5 mA. Quite small compared to what the op-amp eats constantly.

0.5 here, 0.5 there can add up, though.

> I'd say anything between 4k7 and 10k is a good compromise between noise and power consumption for TL07x circuits. And perhaps 10k-22k for TL06x. For low-noise op-amps 2k2 is often a good choice, since higher values increase noise levels.
> 
> /mr
> 
> Den fre 13 nov. 2020 20:13Ian Fritz <ijfritz at comcast.net> skrev:
>> R values around 100k or so give a good compromise between noise and power consumption.
>> 
>> Ian
>> 
>>> On Nov 13, 2020, at 9:59 AM, tpuefke via Synth-diy <synth-diy at synth-diy.org> wrote:
>>> 
>>> I was hoping someone more experienced could help shine some light on this issue. This is still quite puzzling to me as a non-engineer.
>>> 
>>> Looking through my collection of schematics from various places like MFOS (bless his soul), I see a lot of the standard 100k resistor in negative feedback op-amp configuration, for inverters, mixers, attenuators... For a long time i have been adapting these as my go-to values without giving it much thought and usually it works pretty well. The odd schematic here and there uses lower values.
>>> 
>>> Considering that high feedback resistance implies a higher gain potential, shouldn't lower values (e.g. 10k, 1k even) usually be a better choice in terms of circuit stability? 
>>> Or is this a non-issue for fractional gain setups? 
>>> 
>>> I've been wondering about this for a while now and recently stumbled across this post discussing Rf values and op-amp input capacitance:
>>> 
>>> https://www.analog.com/en/analog-dialogue/raqs/raq-issue-122.html
>>> 
>>> 
>>> What are the pros and cons in your experience for using high resistances like these, in attenuator / unity inverter setups especially? 
>>> 
>>> Higher resistances simply to reduce current consumption in bigger circuits? What about resistor noise?
>>> 
>>> Just trying to wrap my head around this to be able to make more informed decisions. 
>>> Any feedback is HIGHly appreciated. :)
>>> 
>>> 
>>> thanks,
>>> Tom
>>> 
>>> 
>>> _______________________________________________
>>> Synth-diy mailing list
>>> Synth-diy at synth-diy.org
>>> http://synth-diy.org/mailman/listinfo/synth-diy
>>> Selling or trading? Use marketplace at synth-diy.org
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
>> Selling or trading? Use marketplace at synth-diy.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20201113/e8b4951f/attachment.htm>