[sdiy] triangle core VCO comparators

[rsdio at audiobanshee.com]

On Feb 26, 2016, at 11:09 PM, mskala at ansuz.sooke.bc.ca wrote:
> On Thu, 25 Feb 2016, Richie Burnett wrote:
>> Some op-amps have anti-parallel protection diodes between the two
>> inputs. Not a problem in typical op-amp circuits where negative feedback
>> keeps both inputs at almost the same potential during normal operation.
>> But it causes problems in "comparator" service where there could be
>> considerable voltage difference forced between the two inputs!
> 
> How common is this *really* in the ordinary kind of op amp today?  The
> OP27 seems to have it.  Any others?

The Linear Technologies LT1677 has ±1.4V limit, as I found out in an integrator circuit. So, it's not just comparator circuits that are sensitive to this, but integrators as well. That op-amp is super low noise, so we had to redesign the biasing to ensure both + and - were within 1.4V of each other.

While searching for a replacements with similar low noise specifications, I found:

LT1028/LT1128 with ±1.8V differential limit
LT1678/LT1679 with ±1.4V differential limit
LT1007/LT1037 with ±0.7V differential limit

I wouldn't call these common or ordinary. You'd think that low noise would mean they'd have to remove the protection diodes, but I'm no op-amp design expert. As I mentioned above, we ended up redesigning the circuit to work within the differential limits because we needed the ultra low noise specifications and couldn't find a substitute that didn't have the same or similar protection diodes without getting into much higher noise specifications.

Brian Willoughby
Sound Consulting