Tektronix AN-USM-488/2235 High +8.6 V Ripple

[jdragoset]

I get roughly 50 mv p-p, seems like it should be less than 5.
Checked a few W jumpers with a milli-ohm meter and W-960 (main +8.6 V from PS)  measured 1.2 milli-ohms.
Gives an mV drop indicating a current of 0.4 amps.
Seems like all of the wire jumpers = 1 to 1.5 milli-ohms (depending on length), meaning they can serve as shunts for measuring millivolt drop and calculating current (current in amps = millivolt drop/milli-ohm shunt.
I don't know if the Tek folks intended this, but it does give a good rough estimate without de-soldering.
Back to the problem,
I'm suspicious some or all of the electrolytic caps have gone open circuit and not much filtering going on.
Given the 0.4 amp total current, I'm not really suspecting shorted caps or other components.
Scope behavior is normal other than high 40 KHz on  2/5 mv/div either channel. 

[coromonadalix]

Add a capacitor  around 100uf 220uf 470uf at 16v on the 8.6vdc line and see what results it gives ??   maybe add an 0.1uf in parallel too ...

[floobydust]

Just replace all the electrolytic capacitors in the power supply section. It's less time and money than troubleshooting each one.
The blue Sangamo (in 2235) caps have epoxy seals that crack with age and they dry out. The Sprague 672D is also very old, time to change them.

For the 2235A Tek upsized all the +/-8.6V, +5.2V rail capacitors from 840uF to 1,000uF. Total six. I use UCC EKYB250ELL102MJ20S, Digikey 565-3892-ND, 1,000uF 25V to fit and has lower ESR, under $1 ea.

edit: fixed typo in 2235,2235A PSU capacitor list spreadsheet, now Ver2 Ver3

[jdragoset]

I have twelve 1000 uf I will change all six tomorrow.
Also have some 2.2 and 22's on the way for several other loations
Thanks,

[jdragoset]

Bingo!
Changed +,- 8.6 V, + 5.2 V, + 30 V & + 100 V PS caps and problem is solved.
Checking what was removed with the LCR meter, it's easy to see that the rest of the electrolytic caps also should be replaced.
Thanks,