The four bridge diodes were identical in my instrument. (Serial Number 002993.)
Since the front and back pairs don't need to be matched as closely as the diodes in a pair, it would not have been unreasonable to make up two pairs instead of a quad.
Tek-made Schottky diodes were a troublesome, low-volume product. D11, the 1S1 sampling bridge, changed several times during its production life.
Less forward drop -> better efficiency and loop gain
050-0422-02 mod kit
Here are the characteristics of 152-0433-00 from "Tektronix Schottky Diodes Memo From 1979 (PDF)" at TekWiki.
Reverse breakdown >= 8V at 1uA
Forward resistance <= 25 ohms at 10mA
Capacitance <= 0.6pF at 0V and 1MHz
Leakage <= 30nA at 3V and 55C
Forward voltage matching, each pair <= 50mV 1mA to 10mA
Forward voltage matching, pair to pair <= 200mV 1mA to 10mA
The 1S1 had quite a few changes overall, not just D11. Q324 was a sore spot. The Mod Summary is good reading.
I'm currently using four HSMS-8202, with two diodes in series replacing one original diode, but they are delicate.
If I were to do it over again I'd try HSMS-2822/HSMS-2825/BAT68-04, again putting two diodes in series, this time to reduce capacitance not increase breakdown voltage.
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of snapdiode via groups.io <firstname.lastname@example.org>
Sent: Friday, December 11, 2020 5:58 PM
Subject: Re: [TekScopes] 1S1 sampling bridge GaAs diodes: What alternatives?
One thing that springs to mind is the four diodes are not identical in the bridge. The two facing the input are physically a bit smaller and have a yellow dot painted on, and diodes facing the memory circuit are bigger and have an orange dot.
Who knows what that means...