102,400 one-shots...

[Phil Hobbs]

On 2023-04-09 10:38, John Larkin wrote:

On Sat, 8 Apr 2023 19:09:50 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2023-04-08 17:55, Klaus Vestergaard Kragelund wrote:
On 07-04-2023 19:59, Phil Hobbs wrote:
So we\'re just finishing up a lidar detector chip in collaboration with
the chip design house (who does the actual circuitry, floor plan, and
polygon-pushing) and the airplane folks.

We did the original proof of concept using a dead-bug prototype with
pHEMTs and CFAs and sampling diodes and stuff, which was enough to get
the program going.  At this point we\'re subcontractors to the camera
folks, basically helping with the design, doing the demo camera, and
coaching everybody.  (We aren\'t head coach--maybe offensive
coordinator.)

It\'s been going for a couple of years, and we\'re at the point of
taping out the first of the second-generation chips.

It contains 102,400 one-shots and 106,496 Class A amplifiers, unless
I\'ve miscounted. (We tried to get them to use a sane number of
monostables, such as 0, but couldn\'t make that one stick.  They take
up less space than registers, we\'re told.)

The power budget is going to be entertaining.


Interesting.

So how does it actually work? I have heard some uses phased arrays so no
need for a spinning detector, other uses simple timing detection, but
need spinning detector.

Short bursts of very high frame rate. Details under NDA, unfortunately.

Do the one-shots store the hits until you can read them all out?

See above.

(Not meaning to be coy--it\'s for a lidar project that\'s under NDA.)

I thought that putting that many one-shots on a chip was fairly amusing,
is all. (If the chip were bigger, we could set it to music and go on
tour--\"Five hundred twenty-five thousand, six hundred one shots....\" (It
even scans.)

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

 

[Joe Gwinn]

EPC2007COn Fri, 07 Apr 2023 18:08:02 -0700, John Larkin
<jlarkin@highlandSNIPMEtechnology.com> wrote:

On Fri, 7 Apr 2023 19:24:52 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2023-04-07 16:23, John Larkin wrote:
On Fri, 7 Apr 2023 13:59:33 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

So we\'re just finishing up a lidar detector chip in collaboration
with the chip design house (who does the actual circuitry, floor
plan, and polygon-pushing) and the airplane folks.

Just between us, what did that cost?

It was surprisingly reasonable--even a dedicated engineering run (20ish
wafers) was about $50k, I think, and the multiproject wafer approach (a
la MOSIS) is considerably cheaper than that. This was on an XFAB 180 nm
process with APDs and SPADs.

We\'ve been kicking around doing the guts of a laser noise canceller on a
transistor array chip--onsemi will do that for just a few $k.

We did the original proof of concept using a dead-bug prototype
with pHEMTs and CFAs and sampling diodes and stuff, which was
enough to get the program going. At this point we\'re
subcontractors to the camera folks, basically helping with the
design, doing the demo camera, and coaching everybody. (We aren\'t
head coach--maybe offensive coordinator.)

It\'s been going for a couple of years, and we\'re at the point of
taping out the first of the second-generation chips.

It contains 102,400 one-shots and 106,496 Class A amplifiers,
unless I\'ve miscounted. (We tried to get them to use a sane number
of monostables, such as 0, but couldn\'t make that one stick. They
take up less space than registers, we\'re told.)

I\'d like to do some custom mixed-signal chips. Apparently it\'s not a
totally crazy idea any more.


The power budget is going to be entertaining.

A mutual friend wants me to do another driver for his electro-optical
gadget. Fast driver. How does one get rid of 60 watts on maybe half a
square inch of PCB? And not block the light?

A loop of copper tube with water running through it, or maybe a vapor
chamber heat spreader with a finned heat sink someplace nearby. Vapor
chambers are about 10x better than copper, iirc. Digikey sells them.

Charging and discharging capacitors, in principle, takes no power.

The canonical method for charging and discharging caps fast and
efficiently is to use an inductor and diode to force the charge waveform
to be half a cycle of a sine wave. That\'s how a lot of lidar laser
drivers work.

Cheers

Phil Hobbs

In theory, one can run up a parallel LC circuit and, once pumped, get
a zillion cap charge and discharge cycles for free after that.

Then, at any zero crossing, short it out for as long as you like, and
turn it loose any time after that to resume the oscillation.

When the inductor is shorted, its current will decay at the L/R rate,
which may be far slower than the oscillation period of the LC tank
when L is not shorted.

EPC2007C was discussed some time ago. It has a very low on
resistance. I imagine it\'s best to short the inductor only when
current is flowing in the correct direction.


>Nice example of a differential equation with initial conditions.

Yes. We normally think of this as starting with a charged capacitor,
and forget the dual situation. But a \"charged\" inductor will do
nicely.

Joe Gwinn

 

[John Larkin]

On Mon, 17 Apr 2023 14:57:10 -0400, Joe Gwinn <joegwinn@comcast.net>
wrote:

EPC2007COn Fri, 07 Apr 2023 18:08:02 -0700, John Larkin
jlarkin@highlandSNIPMEtechnology.com> wrote:

On Fri, 7 Apr 2023 19:24:52 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2023-04-07 16:23, John Larkin wrote:
On Fri, 7 Apr 2023 13:59:33 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

So we\'re just finishing up a lidar detector chip in collaboration
with the chip design house (who does the actual circuitry, floor
plan, and polygon-pushing) and the airplane folks.

Just between us, what did that cost?

It was surprisingly reasonable--even a dedicated engineering run (20ish
wafers) was about $50k, I think, and the multiproject wafer approach (a
la MOSIS) is considerably cheaper than that. This was on an XFAB 180 nm
process with APDs and SPADs.

We\'ve been kicking around doing the guts of a laser noise canceller on a
transistor array chip--onsemi will do that for just a few $k.

We did the original proof of concept using a dead-bug prototype
with pHEMTs and CFAs and sampling diodes and stuff, which was
enough to get the program going. At this point we\'re
subcontractors to the camera folks, basically helping with the
design, doing the demo camera, and coaching everybody. (We aren\'t
head coach--maybe offensive coordinator.)

It\'s been going for a couple of years, and we\'re at the point of
taping out the first of the second-generation chips.

It contains 102,400 one-shots and 106,496 Class A amplifiers,
unless I\'ve miscounted. (We tried to get them to use a sane number
of monostables, such as 0, but couldn\'t make that one stick. They
take up less space than registers, we\'re told.)

I\'d like to do some custom mixed-signal chips. Apparently it\'s not a
totally crazy idea any more.


The power budget is going to be entertaining.

A mutual friend wants me to do another driver for his electro-optical
gadget. Fast driver. How does one get rid of 60 watts on maybe half a
square inch of PCB? And not block the light?

A loop of copper tube with water running through it, or maybe a vapor
chamber heat spreader with a finned heat sink someplace nearby. Vapor
chambers are about 10x better than copper, iirc. Digikey sells them.

Charging and discharging capacitors, in principle, takes no power.

The canonical method for charging and discharging caps fast and
efficiently is to use an inductor and diode to force the charge waveform
to be half a cycle of a sine wave. That\'s how a lot of lidar laser
drivers work.

Cheers

Phil Hobbs

In theory, one can run up a parallel LC circuit and, once pumped, get
a zillion cap charge and discharge cycles for free after that.

Then, at any zero crossing, short it out for as long as you like, and
turn it loose any time after that to resume the oscillation.

When the inductor is shorted, its current will decay at the L/R rate,
which may be far slower than the oscillation period of the LC tank
when L is not shorted.

EPC2007C was discussed some time ago. It has a very low on
resistance. I imagine it\'s best to short the inductor only when
current is flowing in the correct direction.


Nice example of a differential equation with initial conditions.

Yes. We normally think of this as starting with a charged capacitor,
and forget the dual situation. But a \"charged\" inductor will do
nicely.

Joe Gwinn

The \"in theory\" stipulation allows me to use ideal parts.

 

[Klaus Vestergaard Kragelund]

On 09-04-2023 18:38, Fred Bloggs wrote:

On Sunday, April 9, 2023 at 10:37:34 AM UTC-4, John Larkin wrote:
On Sun, 9 Apr 2023 03:03:25 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Saturday, April 8, 2023 at 10:39:18?AM UTC-4, John Larkin wrote:

... an elegant CD ignition circuit that worked something
like that. It was crazy simple ...

Umm- right out of the GE Bipolar Power Handbook from 60s-70s.
I have a lot of cool old data books, but not that one. I\'ll hunt for
one.

My mistake, it was RCA. See Automotive Applications starting page 243. Be warned- it is from a time when engineers did math:

http://www.rsp-italy.it/Electronics/Databooks/RCA/_contents/RCA%20Power%20Transistor%20Applications%20PTA400%201983.pdf

Not sure if it\'s the exact one I had in mind, which I\'m sure pre-dated 1983.

Thanks for the link.

My first databook was RCA CMOS 4000 book, back when I was a kid. I had
only one electronics book, so I spend countless hours looking through
all the datasheets. In the end of the databook, was a couple of chapters
on how to use inverter stages in different ways, for example the trick
of biasing it for a simple amplifier configuration.

 

[Anthony William Sloman]

On Tuesday, April 18, 2023 at 5:42:47 AM UTC+10, John Larkin wrote:

On Mon, 17 Apr 2023 14:57:10 -0400, Joe Gwinn <joeg...@comcast.net> wrote:
EPC2007COn Fri, 07 Apr 2023 18:08:02 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Fri, 7 Apr 2023 19:24:52 -0400, Phil Hobbs <pcdhSpamM...@electrooptical.net> wrote:
On 2023-04-07 16:23, John Larkin wrote:
On Fri, 7 Apr 2023 13:59:33 -0400, Phil Hobbs <pcdhSpamM...@electrooptical.net> wrote:

<snip>

> The \"in theory\" stipulation allows me to use ideal parts.

Good luck with finding even one. Selling LTSpice simulations doesn\'t seem to be a practical business model.

--
Bill Sloman, Sydney