On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:>On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >> >>> It may be just a scheme to improve _net_ >>> loop-gain-bandwidth-product. >>> >>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>> when I find it. >> >> See Analog Devices App Note AN-107 Active Feedback Improves Amplifier >> Phase Accuracy - James Wong (1987) . > >That's the trick where they used a dual, and put the input of the other >op amp in series with the feedback? It was in a PMI app note back >around then. I think it also wound up in the OP227 datasheet, or maybe >it was the OP270. > >Cute idea that I've never actually needed to use. > >Cheers > >Phil HobbsIf you want less error, why not buy a faster opamp, or cascade two amps, each with gain of sqrt(11)? -- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
Odd Circuit, OP Amp over OP Amp
Started by ●October 9, 2013
Reply by ●October 10, 20132013-10-10
Reply by ●October 10, 20132013-10-10
On Thu, 10 Oct 2013 09:38:39 -0700, John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:>On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs ><pcdhSpamMeSenseless@electrooptical.net> wrote: > >>On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >>> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >>> >>>> It may be just a scheme to improve _net_ >>>> loop-gain-bandwidth-product. >>>> >>>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>>> when I find it. >>> >>> See Analog Devices App Note AN-107 Active Feedback Improves Amplifier >>> Phase Accuracy - James Wong (1987) . >> >>That's the trick where they used a dual, and put the input of the other >>op amp in series with the feedback? It was in a PMI app note back >>around then. I think it also wound up in the OP227 datasheet, or maybe >>it was the OP270. >> >>Cute idea that I've never actually needed to use. >> >>Cheers >> >>Phil Hobbs > >If you want less error, why not buy a faster opamp, or cascade two amps, each >with gain of sqrt(11)?The scheme dates back to when such OpAmps did not exist, or at least not at a reasonable price. So the need for greater GBW was attained by using two slower OpAmps. Did you not note the date, 1987, in Blogg's post? IIRC, the paper I will post, when I find it, shows configurations using up to four OpAmps. ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I love to cook with wine. Sometimes I even put it in the food.
Reply by ●October 10, 20132013-10-10
Seems to me this would also improve linearity by "mirroring" whatever non-linearity exists. I wonder if such a scheme might be useful for fanatic level, goldenears audiophiles.... BTW, I ain't one. The people who claim to hear the difference in 0.007 % and 0.005 % THD are more likely psychic than have that much aural accuity. However, if there is money to be made I'm all ears. (pun intended)
Reply by ●October 10, 20132013-10-10
On 10/10/2013 1:22 PM, jurb6006@gmail.com wrote:> Seems to me this would also improve linearity by "mirroring" whatever non-linearity exists. > > I wonder if such a scheme might be useful for fanatic level, goldenears audiophiles.... > > BTW, I ain't one. The people who claim to hear the difference in 0.007 % and 0.005 % THD are more likely psychic than have that much aural accuity. However, if there is money to be made I'm all ears. (pun intended) >See the section on "Error take-off": http://www.tubecad.com/2004/blog0020.htm
Reply by ●October 10, 20132013-10-10
On 10/10/2013 12:38 PM, John Larkin wrote:> On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >>> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >>> >>>> It may be just a scheme to improve _net_ >>>> loop-gain-bandwidth-product. >>>> >>>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>>> when I find it. >>> >>> See Analog Devices App Note AN-107 Active Feedback Improves Amplifier >>> Phase Accuracy - James Wong (1987) . >> >> That's the trick where they used a dual, and put the input of the other >> op amp in series with the feedback? It was in a PMI app note back >> around then. I think it also wound up in the OP227 datasheet, or maybe >> it was the OP270. >> >> Cute idea that I've never actually needed to use. >> >> Cheers >> >> Phil Hobbs > > If you want less error, why not buy a faster opamp, or cascade two amps, each > with gain of sqrt(11)? > >It makes the phase error second order, which is a bigger win than either of those, if low phase error is what you care about. I haven't read it in some years, but iirc the phase error was in the millidegrees out to 1/10 of the 3 dB cutoff. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Reply by ●October 10, 20132013-10-10
On Thu, 10 Oct 2013 10:22:20 -0700 (PDT), jurb6006@gmail.com wrote:>Seems to me this would also improve linearity by "mirroring" whatever non-linearity exists.It does violate the first rule of low-distortion opamp design, "always invert." -- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
Reply by ●October 10, 20132013-10-10
On Thu, 10 Oct 2013 13:32:04 -0400, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:>On 10/10/2013 12:38 PM, John Larkin wrote: >> On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs >> <pcdhSpamMeSenseless@electrooptical.net> wrote: >> >>> On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >>>> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >>>> >>>>> It may be just a scheme to improve _net_ >>>>> loop-gain-bandwidth-product. >>>>> >>>>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>>>> when I find it. >>>> >>>> See Analog Devices App Note AN-107 Active Feedback Improves Amplifier >>>> Phase Accuracy - James Wong (1987) . >>> >>> That's the trick where they used a dual, and put the input of the other >>> op amp in series with the feedback? It was in a PMI app note back >>> around then. I think it also wound up in the OP227 datasheet, or maybe >>> it was the OP270. >>> >>> Cute idea that I've never actually needed to use. >>> >>> Cheers >>> >>> Phil Hobbs >> >> If you want less error, why not buy a faster opamp, or cascade two amps, each >> with gain of sqrt(11)? >> >> > >It makes the phase error second order, which is a bigger win than either >of those, if low phase error is what you care about. I haven't read it >in some years, but iirc the phase error was in the millidegrees out to >1/10 of the 3 dB cutoff. > >Cheers > >Phil HobbsOK, but the ADI paper shows a roughly 6 dB gain peaking as the cost. And the opamp GBWs have to be matched to 1-2%. I don't think they mention distortion or resistor tolerances. If you cascaded two inverting stages, each with lower gain, there would be two opportunities for gain peaking, sort of like their 3rd order version. -- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
Reply by ●October 10, 20132013-10-10
On 10/10/2013 01:49 PM, John Larkin wrote:> On Thu, 10 Oct 2013 13:32:04 -0400, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 10/10/2013 12:38 PM, John Larkin wrote: >>> On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs >>> <pcdhSpamMeSenseless@electrooptical.net> wrote: >>> >>>> On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >>>>> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >>>>> >>>>>> It may be just a scheme to improve _net_ >>>>>> loop-gain-bandwidth-product. >>>>>> >>>>>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>>>>> when I find it. >>>>> >>>>> See Analog Devices App Note AN-107 Active Feedback Improves Amplifier >>>>> Phase Accuracy - James Wong (1987) . >>>> >>>> That's the trick where they used a dual, and put the input of the other >>>> op amp in series with the feedback? It was in a PMI app note back >>>> around then. I think it also wound up in the OP227 datasheet, or maybe >>>> it was the OP270. >>>> >>>> Cute idea that I've never actually needed to use. >>>> >>>> Cheers >>>> >>>> Phil Hobbs >>> >>> If you want less error, why not buy a faster opamp, or cascade two amps, each >>> with gain of sqrt(11)? >>> >>> >> >> It makes the phase error second order, which is a bigger win than either >> of those, if low phase error is what you care about. I haven't read it >> in some years, but iirc the phase error was in the millidegrees out to >> 1/10 of the 3 dB cutoff. >> >> Cheers >> >> Phil Hobbs > > OK, but the ADI paper shows a roughly 6 dB gain peaking as the cost. > And the opamp GBWs have to be matched to 1-2%. I don't think they > mention distortion or resistor tolerances. > > If you cascaded two inverting stages, each with lower gain, there > would be two opportunities for gain peaking, sort of like their 3rd > order version.I've never used the circuit myself, but there are situations (e.g. your constant-phase bandpass filter) where low phase error is really what you want. I'd be sort of surprised if the GBWs of monolithic duals weren't matched to a percent or two. Cheers Phil -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Reply by ●October 10, 20132013-10-10
On Wednesday, October 9, 2013 10:44:09 PM UTC-4, bloggs.fred...@gmail.com wrote:> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: > > > > See Analog Devices App Note AN-107 Active Feedback Improves Amplifier Phase Accuracy - James Wong (1987) .Interesting, thanks. (another old trick that I didn't know.) Say would this improve (reduce) the Q-enhancement one sees in State-variable filters? (Or other integrating filter topologies) George H.> > > It may be just a scheme to improve _net_ loop-gain-bandwidth-product. > > > > I have a paper somewhere on such multi-OpAmp schemes. I'll post when > > > > I find it. > > > ...Jim Thompson > > >> > | James E.Thompson | mens | > > > > > > | Analog Innovations | et | > > > > > > | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | > > > > > > | San Tan Valley, AZ 85142 Skype: Contacts Only | | > > > > > > | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | > > > > > > | E-mail Icon at http://www.analog-innovations.com | 1962 | > > > > > > > > > > > > I love to cook with wine. Sometimes I even put it in the food.
Reply by ●October 10, 20132013-10-10
Le Thu, 10 Oct 2013 13:52:59 -0400, Phil Hobbs a écrit:> On 10/10/2013 01:49 PM, John Larkin wrote: >> On Thu, 10 Oct 2013 13:32:04 -0400, Phil Hobbs >> <pcdhSpamMeSenseless@electrooptical.net> wrote: >> >>> On 10/10/2013 12:38 PM, John Larkin wrote: >>>> On Thu, 10 Oct 2013 10:00:40 -0400, Phil Hobbs >>>> <pcdhSpamMeSenseless@electrooptical.net> wrote: >>>> >>>>> On 10/09/2013 10:44 PM, bloggs.fredbloggs.fred@gmail.com wrote: >>>>>> On Wednesday, October 9, 2013 7:43:39 PM UTC-4, Jim Thompson wrote: >>>>>> >>>>>>> It may be just a scheme to improve _net_ >>>>>>> loop-gain-bandwidth-product. >>>>>>> >>>>>>> I have a paper somewhere on such multi-OpAmp schemes. I'll post >>>>>>> when I find it. >>>>>> >>>>>> See Analog Devices App Note AN-107 Active Feedback Improves >>>>>> Amplifier Phase Accuracy - James Wong (1987) . >>>>> >>>>> That's the trick where they used a dual, and put the input of the >>>>> other op amp in series with the feedback? It was in a PMI app note >>>>> back around then. I think it also wound up in the OP227 datasheet, >>>>> or maybe it was the OP270. >>>>> >>>>> Cute idea that I've never actually needed to use. >>>>> >>>>> Cheers >>>>> >>>>> Phil Hobbs >>>> >>>> If you want less error, why not buy a faster opamp, or cascade two >>>> amps, each with gain of sqrt(11)? >>>> >>>> >>>> >>> It makes the phase error second order, which is a bigger win than >>> either of those, if low phase error is what you care about. I haven't >>> read it in some years, but iirc the phase error was in the >>> millidegrees out to 1/10 of the 3 dB cutoff. >>> >>> Cheers >>> >>> Phil Hobbs >> >> OK, but the ADI paper shows a roughly 6 dB gain peaking as the cost. >> And the opamp GBWs have to be matched to 1-2%. I don't think they >> mention distortion or resistor tolerances. >> >> If you cascaded two inverting stages, each with lower gain, there would >> be two opportunities for gain peaking, sort of like their 3rd order >> version. > > I've never used the circuit myself, but there are situations (e.g. your > constant-phase bandpass filter) where low phase error is really what you > want. I'd be sort of surprised if the GBWs of monolithic duals weren't > matched to a percent or two. >Exactly. I used that exact circuit coupled with modern fast opamps for a *precision* 100kHz IF amplifier in a high accuracy VNA IF and detector board. Works beautifully. The ppm level sync detector was fun too... Note that in this case, the HF peaking wasn't important, being out of band, hence filtered... -- Thanks, Fred.
