negative voltage regulator

People do, as in John L.'s photos, but I'll be darned if I know why. I just solder everything directly to some copper PCB stock. I use a couple mm of #36 wire to connect SMD resistors, caps, and other mechanically-fragile parts.

-- john, KE5FX

Well, not every application is so stringent, and there are nice cheap pulse-rated resistors out there. Also alpo caps don't seem to dry out like wet Al, and they come in a fairly wide range of ESR values.

Cheers

Phil Hobbs

Pansy. ;)

Cheers

Phil "the only good bug is a dead bug" Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510

Only 200 MHz, It's noisier than the TEK1001, so I don't use it much/

Huh, why not? I've been doing all this boring stuff and playing with a single photon thing is fun. I'll need to measure efficiency too. I'm hoping I can use the linear APD response to cross over to the counting regime.

Oh, BTW, I may still have some DC offset issues with the AD8001, (maybe I need a different amp) waving my hands around the circuit* I was moving the offset around, a few mV. (it was +12 mV for this new opamp, which is still too big.)

George H.

You can't reasonably solder down TSSOP or US8 or suchlike tiny stuff without adapters. Or at least I can't.

Got any pictures?

Phil, did you look at my 'scope shot upstream. (I know a faster 'scope would be nice) but I'm seeing these 'fat' pulses, which look like multiple events, I always thought after-pulsing, would be smaller than the original breakdown. (But I think my previous experience with spads were ones with one or a few break down channels... mostly independent of each other.)

With a bunch of interacting channels, Well a weak (higher resistance) channel could set off another strong one nearby?...

On Monday, I'll try to blow this one up and then hook up the good spad. :^)

George H.

The tools needed for this are a good stereo microscope ('good' meaning a 7x-45x zoom Amscope with a 0.5x Barlow lens and LED ring light, less than a kilobuck on eBay) and some UV-cure glue or 5-minute epoxy when you need the result to survive being knocked around on the workbench for a while.

Some random examples:

Abstinence from espresso beforehand is helpful but not as much as you'd think. Somehow, the microscope closes a feedback loop around the whole process.

-- john, KE5FX

Is this one of the 2**14-element ones?

Could be optical. You get hot-carrier emission from APDs, so in geiger mode you usually have to quench all of them, I believe.

Fun. Last summer I got to blow up several of the lasers on an Uber lidar board on purpose, and get paid for it. ;)

Cheers

Phil Hobbs

What do you use to keep parts off the ground plane?

Why such flimsy wires? So easy to break.

I used an old American Optical Stereo zoom microscope for many years. It has a narrow field of view, short working distance, requires manual focusing, and started giving me neck pains from the fixed eyepieces. I invariably burn my fingers when I'm trying to find out where the soldering iron tip is.

I made a steel mounting for a Canon VIXIA Camcorder. It has full zoom capability, auto focus, and displays on a standard HDMI monitor. The working distance is around 10 inches, so I no longer burn my fingers when I'm trying to locate the soldering iron tip.

A nice thing about the Canon is you can get many different adapters on ebay. I'm currently using a Canon 52mm closeup lense P/N 2500. I got some very expensive lens and some very cheap ones. I found the cheap ones work almost as good as the expensive ones. You can get all kinds of adapters for different lens. These are also real cheap on ebay.

I find the zoom capability is more than adequate. I can fill the entire HDMI screen with the shaft of a 6-32 screw. I couldn't do this with the stereo zoom microscope.

Another nice thing about the Canon is you can get all kinds of different filters, such as UV, visible light, polarizing, etc. You can't put these on a stereo zoom.

Another nice thing about the Canon is you can capture still or video images to a PC. This makes it easy to document something to save or upload to the web.

I don't know anything about camera or optics, but ebay is so cheap it is easy to experiment to find out what works best.

I put in provision to adjust the camera height, but I found this is seldom needed.

I don't miss the stereo image at all. I rarely have to examine the side of something. If I do, the large working distance allows me to simply tilt the object to get the view I need.

Now that I have a working model, I plan on sending the design to a machine shop to have it made professionally.

Scrap my comment. I forget what triggered it.

Your job is fascinating. I can never get my mind around single photon work.

Waving your hand around is agood test for parasitic oscillations. You need a fast scope to see them. Unfortunately, touching the circuit with a probe can often kill them. You may need a sensitive microwave spectrum analyzer so you can put the probe an inch or so from the circuit to check for oscillations. HP 22GHz spectrum analyzers are cheap on ebay. They weigh a ton:)

Fuddy Duddy. Can't see a good idea when you see one:)

The AmScope has a working distance of something like 8", so it's rarely the cause of burned fingers. There is almost always plenty of room to maneuver.

Yesterday I was juggling three soldering irons -- two Metcals with different tip sizes and a big honking 80-watt Weller -- as well as a hot-air tool and preheater, stuffing a real PCB with BGAs and all the trimmings. At some point I thought I needed a different nozzle on the hot-air tool, so I changed it with a pair of pliers as usual. Then I realized the old nozzle was better, so I picked it back up without bothering with the pliers. The (only) good part is that now I can commit crimes without leaving fingerprints for a while.

Solder. Just about every part that doesn't have a grounded pin needs to be soldered to something else that does.

For everything else, there's Bondic or epoxy, or maybe a Teflon standoff.

-- john, KE5FX

Some beta blockers or a little bit of ethanol can also help. The dose of the ethanol must be tightly controlled.

I have chosen the way in the middle. I have a collection of standard circuits like ECL gates, ring mixers, inv. op27, noninv op27 etc that has grown over the years. All are small, in the size of a stamp. That can be easily combined:

The pinout of the 3V3 PECL gates is fairly standard, so things like deblocking and input terminations are on the stamps. I/O is on 100 mil centers to fit it to raster board.

This is an LTC2500-32 digitizer in statu nascendi with a BeagleBoneBlack as just another macro. Everything from network access to controlling/fetching the results of the ADC.

Using the Altium Designer snippet feature they can be recycled for a larger board, local layout inclusively.

Without the test points, that results in quite tight boards; here it was not worth to remove the test points since there is a large crystal oven on the back side that sets the board size. A Xilinx Coolrunner for $1 collects what used to be SSI / MSI.

cheers, Gerhard

Hey, Gerhard, how can one reach you by email? A while ago I tried via the address in your header but no answer. It was about an RF job I wanted to hand off.

Every DIP package comes with at least 8 stout standoffs. If you don't bend them back on themselves as in live bug, you can really reef on them while bending Rs and Cs into place.

Cheers

Phil "Dead bugs still rule" Hobbs

Insect cruelty!

But dip packages? Not many data sheets still show dips.

I have about 2 million surface-mount parts 12 feet below my office. And a few sheets of the Bellin adapters. And a couple square feet of gold-plated FR4! And a Dremel!

But my drawers have thousands of them. ;)

It's a bird! It's a plane!

Cheers

Phil Hobbs

No a single 500 um/side diode.

Huh, I never thought about optical. I should really try digging into the literature. A day in the library (or online) can save a week in the lab. :^)

AFAICT the current limit is working fine... I need to measure the photon flux from some light source. (just check the numbers.) A light bulb is sorta tricky, maybe my Rb lamp, that's mostly a single wavelength.

George H.