piclist 2002\08\22\204237a >
Thread: RF transistor colouring
face BY : Jim email (remove spam text)

Noise figure is something that can be assured
over a wafer a little easier than things like
Idss, Vp and g_subm (gm), and these last two
parameters are something that can be probed on-wafer
whereas NF (on a discrete part such as this)
cannot owing to the fact that there are no pads/RF
structures to bring in the 'stripline RF probes'
and perform this high-frequency measurement at
frequency (like 1 GHz).

To assure a wafer is at a particular NF value testing
is done on specialized transisitor test-structures
that possess the necessary RF pads in test reticles
that allow just a few easy, quick RF measurements
(RF measurements take more time per part, require
specialized RF probes versus the simpler "probe
cards" used for DC tests and also requires a Noise
Figure Measuring Test Set, plus DC supplies to bias
the part during testing).

Since the gain of the device can be tracked via so-called
"DC on-wafer tests" (that don't require specialized
conditions like on-wafer RF testing does) the
gm (transconductance), Vp (gate pinch-off voltage) and
Idss (drain current with zero gate voltage) parameters
can be measured and logged -

- tests which also assure device functionality - I'd say
that those marks were for gain-binning.

But, I could be wrong. I don't know what this manufacturer
was doing - and I was just extrapolating based on what our
foundry did (our product was GaAs devices, including
discretes and MMIC's (Microwave Monolithic Integrated
Circuits - almost RF SOIC's in some cases!).

RF Jim

{Original Message removed}
<002b01c24a3d$ea8461c0$0100a8c0@piii500a> 7bit

See also: www.piclist.com/techref/io/serials.htm?key=rf
Reply You must be a member of the piclist mailing list (not only a www.piclist.com member) to post to the piclist. This form requires JavaScript and a browser/email client that can handle form mailto: posts.
Subject (change) RF transistor colouring

month overview.

new search...