I've been doing some simple test sources, basically just one or other carefully selected LED driven by fast CMOS. Regular hex inverter packages (both 74AC04 and 74AC14) look great with resistive loads--clean 250 ps -- 300 ps edges, as measured with an 11801C sampling scope with SD-26 head. <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> (There's a bit of residual inductive rise after the fast edges on the resistive load tests, but we can fix that.) They're built dead-bug fashion with SO14 packages wired up like this: 10nF +5 C C | *---C C-------* | C C | GND | A *--|>o--* in | | 432R 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler | | | *--|>o--* GND | | *--|>o--* | GND With 82 pF to ground from node A, they fall completely apart--the edges show an initial fast rise/fall of a volt or so, followed by a ringy mess for the next 10 ns. <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> <https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> In real life the LEDs won't be faster than 2 ns, and will be driven via a resistor, of course. However, I might well want to use a speedup cap of the same order, so it's very likely to be worthwhile spending an extra buck or so and using LVC1G04 gates in SC-70, with individual bypass caps. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Fast edges from cheap logic
Started by ●March 10, 2023
Reply by ●March 10, 20232023-03-10
On Fri, 10 Mar 2023 12:56:01 -0500, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:> >I've been doing some simple test sources, basically just one or other >carefully selected LED driven by fast CMOS. Regular hex inverter >packages (both 74AC04 and 74AC14) look great with resistive loads--clean >250 ps -- 300 ps edges, as measured with an 11801C sampling scope with >SD-26 head. > ><https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> > ><https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> > >(There's a bit of residual inductive rise after the fast edges on the >resistive load tests, but we can fix that.) > >They're built dead-bug fashion with SO14 packages wired up like this: > > 10nF +5 > C C | > *---C C-------* > | C C | > GND | A > *--|>o--* >in | | 432R >0---|>o---|>o---*--|>o--*---RRRR---0) To sampler > | | | > *--|>o--* GND > | | > *--|>o--* > | > GND > >With 82 pF to ground from node A, they fall completely apart--the edges >show an initial fast rise/fall of a volt or so, followed by a ringy mess >for the next 10 ns. > ><https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> > ><https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> > >In real life the LEDs won't be faster than 2 ns, and will be driven via >a resistor, of course. However, I might well want to use a speedup cap >of the same order, so it's very likely to be worthwhile spending an >extra buck or so and using LVC1G04 gates in SC-70, with individual >bypass caps. > >Cheers > >Phil HobbsFor cheap klunky old parts, the AC's are fast. This is the fastest edge I've seen from CMOS: https://www.dropbox.com/s/7gajbmt923oesli/NC7SV74_2.JPG?raw=1 No, I can't explain the asymmetry. You might consider driving an SAV541 or two from Tiny Logic gates. Rds-on is about 2 ohms with 0.7 on the gate. Heck, you could drive it from ECL, like an EP89. If you drive the SAV gates really hard, like a volt or so, they keep enhancing. Abs Max is for sissies.
Reply by ●March 10, 20232023-03-10
On Friday, March 10, 2023 at 12:56:13 PM UTC-5, Phil Hobbs wrote:> I've been doing some simple test sources, basically just one or other > carefully selected LED driven by fast CMOS. Regular hex inverter > packages (both 74AC04 and 74AC14) look great with resistive loads--clean > 250 ps -- 300 ps edges, as measured with an 11801C sampling scope with > SD-26 head. > > <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> > > <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> > > (There's a bit of residual inductive rise after the fast edges on the > resistive load tests, but we can fix that.) > > They're built dead-bug fashion with SO14 packages wired up like this: > > 10nF +5 > C C | > *---C C-------* > | C C | > GND | A > *--|>o--* > in | | 432R > 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler > | | | > *--|>o--* GND > | | > *--|>o--* > | > GND > > With 82 pF to ground from node A, they fall completely apart--the edges > show an initial fast rise/fall of a volt or so, followed by a ringy mess > for the next 10 ns. > > <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> > > <https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> > > In real life the LEDs won't be faster than 2 ns, and will be driven via > a resistor, of course. However, I might well want to use a speedup cap > of the same order, so it's very likely to be worthwhile spending an > extra buck or so and using LVC1G04 gates in SC-70, with individual > bypass caps.Slightly overcompensated resistive attenuator is what they used in the old days.> > Cheers > > Phil Hobbs > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC / Hobbs ElectroOptics > Optics, Electro-optics, Photonics, Analog Electronics > Briarcliff Manor NY 10510 > > http://electrooptical.net > http://hobbs-eo.com
Reply by ●March 10, 20232023-03-10
On 2023-03-10 13:29, John Larkin wrote:> On Fri, 10 Mar 2023 12:56:01 -0500, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> >> I've been doing some simple test sources, basically just one or other >> carefully selected LED driven by fast CMOS. Regular hex inverter >> packages (both 74AC04 and 74AC14) look great with resistive loads--clean >> 250 ps -- 300 ps edges, as measured with an 11801C sampling scope with >> SD-26 head. >> >> <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> >> >> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> >> >> (There's a bit of residual inductive rise after the fast edges on the >> resistive load tests, but we can fix that.) >> >> They're built dead-bug fashion with SO14 packages wired up like this: >> >> 10nF +5 >> C C | >> *---C C-------* >> | C C | >> GND | A >> *--|>o--* >> in | | 432R >> 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler >> | | | >> *--|>o--* GND >> | | >> *--|>o--* >> | >> GND >> >> With 82 pF to ground from node A, they fall completely apart--the edges >> show an initial fast rise/fall of a volt or so, followed by a ringy mess >> for the next 10 ns. >> >> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> >> >> <https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> >> >> In real life the LEDs won't be faster than 2 ns, and will be driven via >> a resistor, of course. However, I might well want to use a speedup cap >> of the same order, so it's very likely to be worthwhile spending an >> extra buck or so and using LVC1G04 gates in SC-70, with individual >> bypass caps. >> >> Cheers >> >> Phil Hobbs > > For cheap klunky old parts, the AC's are fast. > > This is the fastest edge I've seen from CMOS: > > https://www.dropbox.com/s/7gajbmt923oesli/NC7SV74_2.JPG?raw=1 > > No, I can't explain the asymmetry. > > You might consider driving an SAV541 or two from Tiny Logic gates. > Rds-on is about 2 ohms with 0.7 on the gate. Heck, you could drive it > from ECL, like an EP89. > > If you drive the SAV gates really hard, like a volt or so, they keep > enhancing. Abs Max is for sissies.That would be a win for a laser, which is pretty fast--you can put the quiescent bias just below threshold, and then bang on it. The 'simmer' makes its dynamic impedance much lower, and there's not a lot of delta-V on its capacitance, so it can turn off pretty well. LEDs produce light more or less linearly with current, but it might be worth thinking about whether it would be better behaved if I didn't drop it all the way to zero current between pulses. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●March 10, 20232023-03-10
On Friday, March 10, 2023 at 11:04:50 AM UTC-8, Phil Hobbs wrote:> > On Fri, 10 Mar 2023 12:56:01 -0500, Phil Hobbs > > <pcdhSpamM...@electrooptical.net> wrote:> >> I've been doing some simple test sources, basically just one or other > >> carefully selected LED driven by fast CMOS.> LEDs produce light more or less linearly with current, but it might be > worth thinking about whether it would be better behaved if I didn't drop > it all the way to zero current between pulses.If you want to look at detector fall times, pulling current OUT on the drop will be beneficial; the alternative, letting the recombination time determine the light output, doesn't show the fast fall time of the driver, but rather of the (slower?) LED light-emission process that gradually uses up the minority carriers. For an oscillator-driven LED, that just means a parallel R-C element in series with the current limit R, assuming single-supply drive with cathode on the negative rail.
Reply by ●March 10, 20232023-03-10
On 2023-03-10 15:34, whit3rd wrote:> On Friday, March 10, 2023 at 11:04:50 AM UTC-8, Phil Hobbs wrote: > >>> On Fri, 10 Mar 2023 12:56:01 -0500, Phil Hobbs >>> <pcdhSpamM...@electrooptical.net> wrote: > >>>> I've been doing some simple test sources, basically just one or other >>>> carefully selected LED driven by fast CMOS. > >> LEDs produce light more or less linearly with current, but it might be >> worth thinking about whether it would be better behaved if I didn't drop >> it all the way to zero current between pulses. > > If you want to look at detector fall times, pulling current OUT on the drop > will be beneficial;Ya'd think so, but it generally does nothing useful. Upper state lifetime in direct bandgap semiconductors is short, and the hole mobility is very low, so you can't actually make them drift out of the junction very fast.> the alternative, letting the recombination time > determine the light output, doesn't show the fast fall time of > the driver, but rather of the (slower?) LED light-emission process that > gradually uses up the minority carriers. > > For an oscillator-driven LED, that just means a parallel R-C element in series > with the current limit R, assuming single-supply drive with cathode on the negative rail.Right, that's the speed-up cap I was talking about upthread. Helps a lot with BJT switching, generally not much with LEDs. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●March 10, 20232023-03-10
Am 10.03.23 um 18:56 schrieb Phil Hobbs:> > I've been doing some simple test sources, basically just one or other > carefully selected LED driven by fast CMOS. Regular hex inverter > packages (both 74AC04 and 74AC14) look great with resistive loads--clean > 250 ps -- 300 ps edges, as measured with an 11801C sampling scope with > SD-26 head. > > <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> > > <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> > > (There's a bit of residual inductive rise after the fast edges on the > resistive load tests, but we can fix that.) > > They're built dead-bug fashion with SO14 packages wired up like this: > > 10nF +5 > C C | > *---C C-------* > | C C | > GND | A > *--|>o--* > in | | 432R > 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler > | | | > *--|>o--* GND > | | > *--|>o--* > | > GND > > With 82 pF to ground from node A, they fall completely apart--the edges > show an initial fast rise/fall of a volt or so, followed by a ringy mess > for the next 10 ns. > > <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> > > <https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> > > In real life the LEDs won't be faster than 2 ns, and will be driven via > a resistor, of course. However, I might well want to use a speedup cap > of the same order, so it's very likely to be worthwhile spending an > extra buck or so and using LVC1G04 gates in SC-70, with individual > bypass caps.> CheersI got rise = 481 ps trise and 323 ps tfall from just 2 74LVC1G04DCKR, 100 Ohms each, both driving the 50 Ohm input of an Agilent 54846B scope, 2.25 GHz BW, 156 ps trise. No need to risk the 2V samplers on 5V logic that does not deserve it. That was just a byproduct of a 1pps signal from my GPS-synced 10 MHz source. More intended to be clean than to be fast. < http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf > page 13 +- 74AC was OK when there was the old Fairchild around. But now there is 5V tolerant 74LVC on a contemporary process, 30 years later. Gerhard
Reply by ●March 10, 20232023-03-10
On 2023-03-10 17:14, Gerhard Hoffmann wrote:> Am 10.03.23 um 18:56 schrieb Phil Hobbs: >> >> I've been doing some simple test sources, basically just one or other >> carefully selected LED driven by fast CMOS. Regular hex inverter >> packages (both 74AC04 and 74AC14) look great with resistive >> loads--clean 250 ps -- 300 ps edges, as measured with an 11801C >> sampling scope with SD-26 head. >> >> <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> >> >> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> >> >> (There's a bit of residual inductive rise after the fast edges on the >> resistive load tests, but we can fix that.) >> >> They're built dead-bug fashion with SO14 packages wired up like this: >> >> 10nF +5 >> C C | >> *---C C-------* >> | C C | >> GND | A >> *--|>o--* >> in | | 432R >> 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler >> | | | >> *--|>o--* GND >> | | >> *--|>o--* >> | >> GND >> >> With 82 pF to ground from node A, they fall completely apart--the >> edges show an initial fast rise/fall of a volt or so, followed by a >> ringy mess for the next 10 ns. >> >> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms82pF.png> >> >> >> <https://electrooptical.net/www/sed/MC74AC04_14_RisingEdge500ohms82pF.png> >> >> >> In real life the LEDs won't be faster than 2 ns, and will be driven >> via a resistor, of course. However, I might well want to use a >> speedup cap of the same order, so it's very likely to be worthwhile >> spending an extra buck or so and using LVC1G04 gates in SC-70, with >> individual bypass caps. > >> Cheers > > I got rise = 481 ps trise and 323 ps tfall from just 2 > 74LVC1G04DCKR, 100 Ohms each, both driving the 50 Ohm input of > an Agilent 54846B scope, 2.25 GHz BW, 156 ps trise. > > No need to risk the 2V samplers on 5V logic that does not > deserve it. > > That was just a byproduct of a 1pps signal from my GPS-synced > 10 MHz source. More intended to be clean than to be fast. > > < http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf > page > 13 +- > > 74AC was OK when there was the old Fairchild around. But now > there is 5V tolerant 74LVC on a contemporary process, 30 years > later. > > Gerhard >Seems slower though! Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●March 10, 20232023-03-10
Am 10.03.23 um 23:20 schrieb Phil Hobbs:> On 2023-03-10 17:14, Gerhard Hoffmann wrote: >> Am 10.03.23 um 18:56 schrieb Phil Hobbs: >>> >>> I've been doing some simple test sources, basically just one or other >>> carefully selected LED driven by fast CMOS. Regular hex inverter >>> packages (both 74AC04 and 74AC14) look great with resistive >>> loads--clean 250 ps -- 300 ps edges, as measured with an 11801C >>> sampling scope with SD-26 head. >>> >>> <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> >>> >>> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> >>> >>> (There's a bit of residual inductive rise after the fast edges on the >>> resistive load tests, but we can fix that.) >>> >>> They're built dead-bug fashion with SO14 packages wired up like this: >>> >>> 10nF +5 >>> C C | >>> *---C C-------* >>> | C C | >>> GND | A >>> *--|>o--* >>> in | | 432R >>> 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler >>> | | | >>> *--|>o--* GND >>> | | >>> *--|>o--* >>> | >>> GND >>>>> >> I got rise = 481 ps trise and 323 ps tfall from just 2 >> 74LVC1G04DCKR, 100 Ohms each, both driving the 50 Ohm input of >> an Agilent 54846B scope, 2.25 GHz BW, 156 ps trise. >> >> No need to risk the 2V samplers on 5V logic that does not >> deserve it. >> >> That was just a byproduct of a 1pps signal from my GPS-synced >> 10 MHz source. More intended to be clean than to be fast. >> >> < http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf > >> page 13 +- >> >> 74AC was OK when there was the old Fairchild around. But now >> there is 5V tolerant 74LVC on a contemporary process, 30 years >> later. >> >> Gerhard >> > > Seems slower though!Yes, with 4 drivers into 500 Ohms, that's easy living. Not so easy if you need a presentable 1pps... But just take a look at the quality of these stones... (Live of Brian) < https://www.youtube.com/watch?v=OVUwHv43HqM >> CheersGerhard
Reply by ●March 10, 20232023-03-10
On 2023-03-10 17:49, Gerhard Hoffmann wrote:> Am 10.03.23 um 23:20 schrieb Phil Hobbs: >> On 2023-03-10 17:14, Gerhard Hoffmann wrote: >>> Am 10.03.23 um 18:56 schrieb Phil Hobbs: >>>> >>>> I've been doing some simple test sources, basically just one or >>>> other carefully selected LED driven by fast CMOS. Regular hex >>>> inverter packages (both 74AC04 and 74AC14) look great with resistive >>>> loads--clean 250 ps -- 300 ps edges, as measured with an 11801C >>>> sampling scope with SD-26 head. >>>> >>>> <https://electrooptical.net/www/sed/MC74AC04-14RisingEdges500ohmLoad.png> >>>> >>>> >>>> <https://electrooptical.net/www/sed/MC74AC04_14_FallingEdge500ohms.png> >>>> >>>> (There's a bit of residual inductive rise after the fast edges on >>>> the resistive load tests, but we can fix that.) >>>> >>>> They're built dead-bug fashion with SO14 packages wired up like this: >>>> >>>> 10nF +5 >>>> C C | >>>> *---C C-------* >>>> | C C | >>>> GND | A >>>> *--|>o--* >>>> in | | 432R >>>> 0---|>o---|>o---*--|>o--*---RRRR---0) To sampler >>>> | | | >>>> *--|>o--* GND >>>> | | >>>> *--|>o--* >>>> | >>>> GND >>>> > >>> >>> I got rise = 481 ps trise and 323 ps tfall from just 2 >>> 74LVC1G04DCKR, 100 Ohms each, both driving the 50 Ohm input of >>> an Agilent 54846B scope, 2.25 GHz BW, 156 ps trise. >>> >>> No need to risk the 2V samplers on 5V logic that does not >>> deserve it. >>> >>> That was just a byproduct of a 1pps signal from my GPS-synced >>> 10 MHz source. More intended to be clean than to be fast. >>> >>> < http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf > >>> page 13 +- >>> >>> 74AC was OK when there was the old Fairchild around. But now >>> there is 5V tolerant 74LVC on a contemporary process, 30 years >>> later. >>> >>> Gerhard >>> >> >> Seems slower though! > > Yes, with 4 drivers into 500 Ohms, that's easy living. > Not so easy if you need a presentable 1pps... > > > But just take a look at the quality of these stones... > (Live of Brian) > < https://www.youtube.com/watch?v=OVUwHv43HqM >Monty Python dubbed in German is a bit surreal. Yessir, a very nice writeup. Out of curiosity, why did you choose crystal notch filters, as opposed to an LC tuned amp? Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
