https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1

Wonder how that will work. I'll have one of my people try it.

It would certainly work with a fast comparator with hysteresis.

--

I yam what I yam - Popeye

On 2/2/2022 11:53 AM, jlarkin@highlandsniptechnology.com wrote:
> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>
> Wonder how that will work. I'll have one of my people try it.
>
> It would certainly work with a fast comparator with hysteresis.
>
>

ECL seems well-suited to generating sines also, as it kinda has a
soft-limiting "tanh-shaper" built into it to begin with

Old timey NOR-gate:

<https://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/ECL.svg/350px-ECL.svg.png>

bias it into class C with a negative voltage below Vee on the inputs and
connect an LC back to the input from not-Q and I think it should sing.
For 1+ GHz maybe some trace inductance and an inter-digital capacitor
would be enough.

<https://www.mouser.com/ProductDetail/onsemi/MC100LVEL01DR2G?qs=zorda86t5M9xivhmK851Lw%3D%3D>

On 2/2/2022 1:28 PM, bitrex wrote:
> On 2/2/2022 11:53 AM, jlarkin@highlandsniptechnology.com wrote:
>> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>>
>> Wonder how that will work. I'll have one of my people try it.
>>
>> It would certainly work with a fast comparator with hysteresis.
>>
>>
>
> ECL seems well-suited to generating sines also, as it kinda has a
> soft-limiting "tanh-shaper" built into it to begin with
>
> Old timey NOR-gate:
>
> <https://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/ECL.svg/350px-ECL.svg.png>
>
>
> bias it into class C with a negative voltage below Vee on the inputs and
> connect an LC back to the input

Series LC to be specific if that wasn't clear

On Wed, 2 Feb 2022 13:28:51 -0500, bitrex <user@example.net> wrote:

>On 2/2/2022 11:53 AM, jlarkin@highlandsniptechnology.com wrote:
>> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>>
>> Wonder how that will work. I'll have one of my people try it.
>>
>> It would certainly work with a fast comparator with hysteresis.
>>
>>
>
>ECL seems well-suited to generating sines also, as it kinda has a
>soft-limiting "tanh-shaper" built into it to begin with
>
>Old timey NOR-gate:
>
><https://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/ECL.svg/350px-ECL.svg.png>
>
>bias it into class C with a negative voltage below Vee on the inputs and
>connect an LC back to the input from not-Q and I think it should sing.
>For 1+ GHz maybe some trace inductance and an inter-digital capacitor
>would be enough.
>
><https://www.mouser.com/ProductDetail/onsemi/MC100LVEL01DR2G?qs=zorda86t5M9xivhmK851Lw%3D%3D>
>
>

The old 10K parts had low gain and could be used as linear amplifiers.
The EL and EP gates have a lot more gain, so have harder edges.

--

John Larkin Highland Technology, Inc trk

The cork popped merrily, and Lord Peter rose to his feet.
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"

onsdag den 2. februar 2022 kl. 23.07.13 UTC+1 skrev John Larkin:
> On Wed, 2 Feb 2022 13:28:51 -0500, bitrex <us...@example.net> wrote:
>
> >On 2/2/2022 11:53 AM, jla...@highlandsniptechnology.com wrote:
> >> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
> >>
> >> Wonder how that will work. I'll have one of my people try it.
> >>
> >> It would certainly work with a fast comparator with hysteresis.
> >>
> >>
> >
> >ECL seems well-suited to generating sines also, as it kinda has a
> >soft-limiting "tanh-shaper" built into it to begin with
> >
> >Old timey NOR-gate:
> >
> ><https://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/ECL.svg/350px-ECL.svg.png>
> >
> >bias it into class C with a negative voltage below Vee on the inputs and
> >connect an LC back to the input from not-Q and I think it should sing.
> >For 1+ GHz maybe some trace inductance and an inter-digital capacitor
> >would be enough.
> >
> ><https://www.mouser.com/ProductDetail/onsemi/MC100LVEL01DR2G?qs=zorda86t5M9xivhmK851Lw%3D%3D>
> >
> >
> The old 10K parts had low gain and could be used as linear amplifiers.
> The EL and EP gates have a lot more gain, so have harder edges.

https://shrubbery.net/~heas/willem/PDF/NSC/AN/AN-88.pdf

On Wed, 02 Feb 2022 08:53:27 -0800, jlarkin@highlandsniptechnology.com
wrote:

>https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>
>Wonder how that will work. I'll have one of my people try it.

Oscillates at 1.5 GHz, reasonable square wave, bad tempco.

Can be tuned down maybe 20% with C2.

There are tons of "clock generator" chips. Some are actually just
buffers or dividers. Some are very complex synthesizers. Most are
unavailable.

--

John Larkin Highland Technology, Inc trk

The cork popped merrily, and Lord Peter rose to his feet.
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"

On Wed, 02 Feb 2022 08:53:27 -0800, jlarkin@highlandsniptechnology.com
wrote:

>https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>
>Wonder how that will work. I'll have one of my people try it.
>
>It would certainly work with a fast comparator with hysteresis.

This should work.

https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1

--

I yam what I yam - Popeye

jlarkin@highlandsniptechnology.com wrote:
> On Wed, 02 Feb 2022 08:53:27 -0800, jlarkin@highlandsniptechnology.com
> wrote:
>
>> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
>>
>> Wonder how that will work. I'll have one of my people try it.
>>
>> It would certainly work with a fast comparator with hysteresis.
>
> This should work.
>
> https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1
>
>
>
>

That looks similar to the late Jim Thompson's MC1648.

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

On Monday, February 7, 2022 at 6:56:15 PM UTC-8, jla...@highlandsniptechnology.com wrote:

> This should work.
>
> https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1

Substituting a tapped inductor (drive the tap, sense/resonate the endpoint) would improve
the slew rate. Tapped inductor is a gain boost, and the chip has input clamps.

On Tuesday, February 8, 2022 at 4:38:58 PM UTC+11, Phil Hobbs wrote:
> jla...@highlandsniptechnology.com wrote:
> > On Wed, 02 Feb 2022 08:53:27 -0800, jla...@highlandsniptechnology.com
> > wrote:
> >
> >> https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
> >>
> >> Wonder how that will work. I'll have one of my people try it.
> >>
> >> It would certainly work with a fast comparator with hysteresis.
> >
> > This should work.
> >
> > https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1
> >
> That looks similar to the late Jim Thompson's MC1648.

The MC1648 is remarkably old. The Motorala - ON Semiconductor - ECLinPS parts go back to the 1990's,, and the MC1648 shows up in the 1973

<https://doc.lagout.org/science/0_Computer%20Science/0_Computer%20History/old-hardware/motorola/_dataBooks/Motorola_Phase-Locked_Loop_Systems_Data_Book_2ed_Aug73.pdf>

The topology may be similar, but the transistors are going to be a lot smaller that the ones Jim designed in.

--
Bill Sloman, Sydney

On Mon, 7 Feb 2022 23:07:06 -0800 (PST), whit3rd <whit3rd@gmail.com>
wrote:

>On Monday, February 7, 2022 at 6:56:15 PM UTC-8, jla...@highlandsniptechnology.com wrote:
>
>> This should work.
>>
>> https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1
>
>Substituting a tapped inductor (drive the tap, sense/resonate the endpoint) would improve
>the slew rate. Tapped inductor is a gain boost, and the chip has input clamps.

That chip has lots of gain, and we have lots of low-nH 0603 inductors
in stock.

--

I yam what I yam - Popeye

On Wednesday, February 9, 2022 at 12:25:02 AM UTC+11, jla...@highlandsniptechnology.com wrote:
> On Mon, 7 Feb 2022 23:07:06 -0800 (PST), whit3rd <whi...@gmail.com>
> wrote:
> >On Monday, February 7, 2022 at 6:56:15 PM UTC-8, jla...@highlandsniptechnology.com wrote:
> >
> >> This should work.
> >>
> >> https://www.dropbox.com/s/chgyy2kvj6ye5gq/Osc_EP11_LC1.jpg?raw=1
> >
> >Substituting a tapped inductor (drive the tap, sense/resonate the endpoint) would improve
> >the slew rate. Tapped inductor is a gain boost, and the chip has input clamps.
>.
> That chip has lots of gain, and we have lots of low-nH 0603 inductors
> in stock.

Not all that much gain at 1.5 GHz.

And you don't seem to know enough about how your SMD inductors work. Most of the ones I used printed a straight conducting track onto a lump of ferrite.

Getting much magnetic coupling between them might be difficult. Ferrite beads don't look as neat, but at least you know exactly the route the current follows.

--
Bill Sloman, Sydney

On Saturday, February 5, 2022 at 12:38:04 AM UTC+11, John Larkin wrote:
> On Wed, 02 Feb 2022 08:53:27 -0800, jla...@highlandsniptechnology.com
> wrote:
> >https://www.dropbox.com/s/627ifxuw3rsexj2/EP11_Oscillator.jpg?raw=1
> >
> >Wonder how that will work. I'll have one of my people try it.
>
> Oscillates at 1.5 GHz, reasonable square wave, bad tempco.
>
> Can be tuned down maybe 20% with C2.

This is a pretty silly post. The MC10EP11 has a typical propagation delay of 220psec at room temperature - rising from 200psec at -40C to 240psec at 85C.

If what John is setting up is a delay line oscillator which relies on getting a total round trip delay of 333psec - 180 degrees of phase shift at 1.5 GHz - his "delay line" is contributing 113psec of that delay. At -40C, the EP11 propagation delay would deliver 1.6GHz if the external delay remained the same, and at 85C it would deliver 1.42GHz.

113psec is good for about a 3.3cm propagation distance in free space, maybe 2.4 cm through printed circuit substrate.

The schematic he has posted doesn't seem to have been drawn with that in mind. What his mental model of his oscillator actually might be is something of a mystery - perhaps more of a mystery to him than the rest of us.

--
Bill Sloman, Sydney