Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > sci.electronics.design > #518146
| From | Jeroen Belleman <jeroen@nospam.please> |
|---|---|
| Newsgroups | sci.electronics.design |
| Subject | Re: Interesting LT/Analog op amp |
| Date | 2018-10-02 21:39 +0200 |
| Organization | Aioe.org NNTP Server |
| Message-ID | <pp0hic$1rsc$1@gioia.aioe.org> (permalink) |
| References | <uXMsD.182542$RN2.174536@fx40.iad> <pp0ae2$1eic$1@gioia.aioe.org> |
On 02/10/18 19:37, 698839253X6D445TD@nospam.org wrote: > bitrex wrote >> The LT1364/1365 dual/quad, 70MHz GBW and 1000V/uS slew rate: >> >> <http://www.analog.com/media/en/technical-documentation/data-sheets/13645fa.pdf> >> >> The simplified internal schematic is quite different than most >> traditional three-gain-stage V -> I -> V amps, looks like a cascade of >> "diamond"-configuration differential current amplifiers. > > Dunno what you mean by 'diamond shape' > input is complementary emitter followers, nothing differential there, via R1 to next stage. > > The opposite phases are generated by driving the complementary transistor emitters with R1. > Some current sources follow.. > IN+ drives in fact .. complementary emitter followers... sets the working point. > Output is complementary emitter followers, > > Temp tracking should be really good, else it distorts like .. > The term comes from the OPA660 'diamond transistor' which, I believe, was the first to use this architecture in an IC. The complementary emitter followers combined with the current mirrors behave much like a single transistor, with a high impedance 'base' at +IN, a high impedance 'collector' at the C-Rc node and a low impedance 'emitter' at the right side of R1, with the added twist that currents can be flowing either way. It's a nice architecture really, although the OPA660 never got very popular. The LT1364 adds another emitter follower at -IN, turning it into something resembling a long-tailed pair, but owing to the symmetry, the 'tail' of the pair is no longer needed. A pity R1 is so high, because that sets the transconductance of the central stage to a very modest 2mS at most. Jeroen Belleman
Back to sci.electronics.design | Previous | Next — Previous in thread | Find similar | Unroll thread
Interesting LT/Analog op amp bitrex <user@example.net> - 2018-10-02 12:22 -0400
Re: Interesting LT/Analog op amp <698839253X6D445TD@nospam.org> - 2018-10-02 17:37 +0000
Re: Interesting LT/Analog op amp Jeroen Belleman <jeroen@nospam.please> - 2018-10-02 21:39 +0200
csiph-web