measuring low-value device capacitances

Measuring the capacitance of devices that have very low values is difficult, unless you have the right measuring instrument, and the right fixtures. In the case of all types of SMT components, what's really needed is custom-crafted PCBs. Nine years ago I set my PCB engineer, Chuck Fisk, to create such a board, really a set of boards, with common resistor pads. In general confusion of the time, I lost track of the file, but yesterday found it again. It looked good, but I wanted to add more kinds of items, transistors, op-amps, diodes, etc. Also, he had come up with some unusual pad types, so I deleted many of them.

It has a few motherboards to connect instruments like an HP 4280A or 4192A, measuring to 0.001pF. There are 85 daughterboards for test components.

formatting link

I have five empty locations left - I'm wondering if any of you have some other good candidates?

--
 Thanks, 
    - Win
Reply to
Winfield Hill
Loading thread data ...

A signal generator and an oscilloscope can measure very small capacitances. Set a function generator for, say, a 1 MHz sine wave at max output, 10 volts peak maybe. Trigger the scope from the generator and maybe set up some signal averaging.

If the scope is 15 pF in, 15 fF coupling makes 20 mV p-p, easily resolved.

A fet probe might have a few pF input capacitance, even better. A homemade fet follower could be 1:1 with a pF or two loading. Resolving

1 fF shouldn't be difficult.

Signal averaging on a scope is like the synchronous detector in a classic c-meter.

I occasionally do a 4-layer test board, for circuits that are difficult to breadboard any other way. I ask the kids if they want anything, so we toss on various other circuits and shear the board up into bits. We accumulate a wish list for the next proto board.

Some parts, like US8's, are impossible to handle, so we make adapters for them. Little connectorized coplanar waveguides, connector adapters, filters, splitters, all sorts of things can be handy.

formatting link

formatting link

formatting link

formatting link

We assign these boards drawing and part numbers, the "Z" series, and archive the board files and any interesting test results. A few have wound up being sellable products.

How about a little fet follower board, for making those small capacitance measurements? We could scribble brainstorm that here.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Indeed. However, an issue is still mounting the part and getting a measurement with its PCB pads. That's the reason for the 85 small boards.

That's a good idea, modify one of the motherboards.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

Surely the answer is yes. I can test that by mangling a few daughterboards. It'll also be interesting to see the reduction from a ground plane or fat ground wires between the the pads.

But grounding methods aren't always safe to use. High-value resistors may have different portions of their current flow diverted along their length.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

It's too bad that my Realistic Minimus 0.3 speaker isn't SMD and won't fit on a daughterboard. :( But its inductance is still needed for an accurate spice model.

Here's a picture of my own, large component, DUT holder:

formatting link

And here's a picture of my DUT holder next to the speaker.

formatting link

My plan for the time being is to use two solid bare wires of the shortest length possible to connect the speaker to the DUT holder for VNA measurement of inductance.

Thank you,

--
Don Kuenz KB7RPU
Reply to
Don Kuenz

You're fixture mounts the part on a PCB, that's good. It's convenient to use a SMD test adapter

formatting link

But measuring this way lacks the accuracy of a PCB-mounting fixture, for small capacitances. I hold the fixture contacts apart during zero setting, but still lack pad capacitance info.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

I'm wondering if any of you have some other good candidates? Anybody?

--
 Thanks, 
    - Win
Reply to
Winfield Hill

Do you have the tiny TSOP package?

Cheers

Reply to
Martin Riddle

Those super small minimold packages that the late lamented NE3509 came in. There are decent diodes in that package too.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

I've been avoiding that package size whenever I can -- it's a trace too small for easy hand soldering. But sometimes there's no getting around it, so OK it's a good idea. I'll make room and add a few.

Meanwhile I've updated the files on DropBox, if anybody wants to take a look.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

That one isn't too awful to solder by hand. I just wish people would standardize on which pin is the big one, and how they orient things.

I have found a couple of subs for the 3508/3509 phemts, but they need to be rotated or otherwise fudged.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

In my boards like yours (mostly add-ons to other circuits, not as ambitious as yours) I sometimes add bypass caps to the obvious corner pins of some of the IC footprints, so I can use the resulting adapters for fast stuff without breadboard wire inductance trashing things so much.

formatting link

I should have added a couple more ground pads to each one, for less grounding inductance. It is a 4-layer board.

Well, next time.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Make a board up with a copy of the TI FDC1004 EVM module for when you want to quickly measure something without a big HP box?

Steve

Reply to
sroberts6328

Steve, that's a good idea, but doesn't the FDC1004 only work with grounded sensors?

--
 Thanks, 
    - Win
Reply to
Winfield Hill

We have a P6701 around here, X1 fet probe with 3 pF loading. That would get us below 1 fF resolution working with a 20 v p-p function generator putput.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

The optical converter probe?

--
 Thanks, 
    - Win
Reply to
Winfield Hill

I must have too much optics on my brain lately. That should have been P6201.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

I have a bunch of those. A small pHEMT is way under a pF C_iss--that nanoamp cascoded pHEMT front end I did for Samsung's research division some years ago had a total input capacitance of just under 1 pF.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Given a bit of gain, it shouldn't be difficult to resolve, say, 10 aF.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.