TO can package parasitics

Nov 29, 2021 Last reply: 4 years ago 16 Replies

Hi, all,



I need to build a fast front end for a fairly huge photodiode--2 mm square, 12 pF, TO-8 can package.



The TO-8 has a pin spacing of 0.2 inch, so based on the usual



20-nH-per-inch rule of thumb, I expect its package inductance to come in at 5 nH or thereabouts.

I looked around on the Web for some table of parasitics for varioius packages, but didn't find anything that included old-timey packages such as metal cans.



Does anybody have such a reference handy?



Thanks



Phil Hobbs


Have you checked bitsavers.org or archive.org? Al Kossow has sent archive.org a lot of his scanned parts catalogues and reference manuals. Easier to find on bitsavers though...

John :-#)#

You sure you mean TO-8 ?

TO-8 is a package that was used almost exclusively for power germanium parts.

RL

You might compare it to some laser diode packages for Mech/Elec strays.

formatting link
RL

Yup.

And a whole lot of larger-size optical detectors that aren't passivated well enough for non-hermetic packages.

Cheers

Phil Hobbs

I could if I had one handy, but ATM I'm dealing with a requirement to make a front end with ~ 350 MHz BW from a TO-8.

12 pF and 5 nH resonate at 650 MHz, so you see the issue.

I'm trying to decide whether to have a go at it or insist on the goal posts being moved.

Cheers

Phil Hobbs

formatting link
I could TDR that one, but it's a TO5.

Any idea why gold-plated kovar would reduce inductance by factor of

3:1 ?

RL

Maybe they got the decimal point in the wrong place. Otherwise it doesn't make sense. John

Probably just closer to the baseplate--reduces loop area. Or it could be wrong. ;)

Cheers

Phil Hobbs

I think kovar is magnetic, perhaps that would make a difference, at least at some frequencies.

I've seen butterfly package lasers with a long wirebond from the pin to the laser chip, and some with a ceramic transmission line ramp sloping down to the chip.

formatting link

TL or just and SMD link?

formatting link
Doesn't help if the canned part is prescribed.

RL

. . . any increase in permeability in the neighborhood could be expected to make things worse, not better.

RL

Depends on the frequency. Up at UHF the skin depth in the kovar will be zilch, and the permeability will be pretty small anyway--it goes to 1.0 around 500 MHz for most iron/nickel alloys AFAICT.

A quick search didn't turn up anything more recent than this:

formatting link
(open access).

Cheers

Phil Hobbs

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required