No Vacuum on Probes

Mihir Tendulkar mihir.tendulkar at
Fri Jul 8 15:32:13 PDT 2011

Debugging the manipulator vacuum, for me, usually went in several

1) Check that the probe station surface is clean; wipe off with IPA
2) Clean the rubber seals underneath each manipulator (IPA wipe) 
3) Blow out the yellow vacuum tubes with a gas canister

More often than not, these first three steps would solve the problem and
restore vacuum to a working capacity. If not, the next step is to open
up the Cascade and clean the vacuum manifold. I believe there is one for
each side of the Cascade (left/right). 

-----Original Message-----
From: John Shott [mailto:shott at] 
Sent: Friday, July 08, 2011 3:28 PM
To: J. Jason Lin
Cc: cascade at
Subject: Re: No Vacuum on Probes


Have you received any response on this?  Although I don't have any 
specific experience with the Cascade system and probes, I've got a bit 
of experience with vacuum mounted probes in general and have a couple of

things that you might look at.

In general, the vacuum distribution to probes tends to be low 
conductance .... that means skinny vacuum lines.  As a result, it 
doesn't take too much to block or partially block the vacuum path.  
Equally, it doesn't take too much of a leak somewhere near the probes to

really degrade the vacuum near that leak.

If you have good vacuum to one probe and bad vacuum to the other four, 
is there anything about the routing and geometry of the vacuum lines 
that might help you isolate the problem.  For example, often there is a 
left side and a right side of the vacuum distribution.  Is the good 
vacuum on one side and the bad vacuum on another?  Is there a kinked or 
collapsed line that might result in little vacuum on the far end of the 
kink/collapse?  If not all probes are connected, is there a valve or 
something similar that should be closed that is leaking?

I will try to wander in and take a look ... Are the probes numbers so 
that you can tell me which has good vacuum and which have bad?



> Hello Cascade users,
> Can the superusers take a look at this problem, or is there something 
> we can do ourselves to fix this?  Right now there seems to be only one

> probe with vacuum.  The other 4 move quite a bit when manipulating the

> probes, making measurements very difficult and probably risking the 
> probes accidentally hitting each other.
> Thanks,
> Jason

More information about the cascade mailing list