Re: Fast sampler

On 12/03/2025 5:12 am, Phil Hobbs wrote:

On 2025-03-09 19:22, Tom Del Rosso wrote:

Phil Hobbs wrote:

Hi, All,
>
Late last year we did a fast sampler/TDR with nice clean 60 ps edges.
>
We're gearing up to actually sell them, so I did a short technical
writeup on the design, which may be of interest.
>
<https://electrooptical.net/News/a-high-performance-time-domain-reflectometer>
>
Cheers
>
Phil Hobbs

>
Why is a LASER ruler that can measure distances in air with 2mm accuracy
$16,
>
and an OTDR for measuring fiber $600?
>

 Well, the OTDR needs a fiber-coupled laser, for one thing, and a decent TIA.  Also as John says, it's a time-domain instrument.  A laser ruler can just be a diode laser with a monitor photodiode, a collimating lens, a simple TIA and a micro with built-in ADC.
 You put a small (1-3 mA) current ramp on the laser, and look at the beat signal coming out of the monitor photodiode.  The frequency gives you the round-trip delay.  This sort of laser feedback measurement can be pretty good if the diode stays reasonably single-mode.
 I have no idea how the $16 ones do it, but if I were building one, that's the first thing I'd try.

Keeping a laser diode single mode seems to mean that you keeps its junctions temperature stable to better than a degree (and they do self-heat in operation.
My 1996 millidegree thermostat was designed to stabilised the temperature of the laser's target, but the machine worked better if the laser was operating single mode, and stayed in that single mode, so my thermostat got duplicated to stabilise the temperature of the laser diode as well. The stable temperature was set up during final test to give a temperature half-way between that particular laser's single-mode switching temperatures, but that just meant writing a single word into memory.
--
Bill Sloman, Sydney