Blob-board for high-freq prototyping

Don't let Joerg near that. One of his "phfffft"'s could cost you hours of re-work.

Grins, James Arthur

Am 02.04.2017 um 07:03 schrieb John Larkin:

I use unetched FR4 that is usually chemically tin-plated and stamp sized carriers that are either snippets of bought proto cards or macro functions that repeat often and were found important enough to lay out.

The LT1677 follower last week was not important.

Standard blocks are inverting & non-inverting op27-behave-alikes,

100EP gates or 100EP flipflops together with optional Thevenin- terminations and decoupling or chips that are hard to handle like LT3042 with its thermal pad.

The stamps are all single sided, bottom is completely GND and lays on the global ground plane. Some tin blobs make positive GND continuity.

Over the years, I have collected quite a library. The pre-layouted blocks also exist in the Altium library without the test points on the 100 mil grid. Condensing a hand soldered board to a real layout is then nothing but shoving pre-routed macros across the screen and doing the high-level interconnects. Maybe there are some ferrite beads that are not really necessary, I don't care.

Example of a pulse generator with 2 semi-rigid delay lines:

< >

The ADCMP580 picture 2 or 3 places to the left has been done from hardware in this style.

regards, Gerhard.

Oh, the camera of this Sony cell phone is soo cruel... It looks much better to the nekkid eye!

I occasionally order a couple of square feet of gold-plated double-side FR4, which I shear up as needed. It makes beautiful litte prototypes. They don't tarnish like bare copper or solder-coat.

That's cool. We planned a product like that called Timing Tiles, which would be a lot of similar small boards: amplifiers, buffers, comparators, picosecond one-shots, fanouts, laser drivers, things like that. The idea was that people could use them to assemble custom timing systems. But we then figured that the support requirements would wipe out any profits, so we didn't do it. Pity, it would have been fun.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

So why not use those blocks internally for the same idea? Have a web page where people can tick off the functions they want, then the software spits out which tiles to solder to the ground plane and how to interconnect them. Extra points if you can program that into the pick and place machine. Alternatively, the software figures out the few interconnections between the function blocks and spits out a custom PC board layout. Add a custom front panel and you can be producing one of a kind instrumentation at nearly mass production prices.

Or not...

Selling one-offs to grad students is not a good way to make money. It's better to have an OEM that demonstrates how serious they are by ordering a bunch of first articles for serious money.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Well, that's why the "Or not" was there. I guess it all depends on the size of the untapped market, which in this case would be admittedly small.

There are lots of people who want instrumentation but don't understand electronics and often don't understand their own problem. Given free consulting, they'll take all you have.

The tiles concept could make someone a good living, but they would have to include, say, one hours' consulting per application, and charge for more.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

That's the beauty of CAMAC. All you need is a sufficiently large budget. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net