No clean flux has been a bane of the in-circuit test world ever since its introduction. There are several approaches to help, but no cure the problem:
1) Reduce usage of flux to the bare minimum. This usually causes a conflict between the solder process engineer who finds that the more flux he/she uses the better the quality of the solder joints.
The test engineer on the other hand has to live with the residual flux and its effect on test probes. There are a number of solder/flux vendors who have formulated "no residue" fluxes that can help.
They may cost more than conventional "low residue" flux but can save the customer in the long term from the hidden cost of re-test and rework. Contact vendors such as Kester, Indium, Nihon, etc.
2) Change probe styles, materials and spring forces.
a. There are a number of probe head styles that have been created to deal with high flux environments. Many of the head styles have more aggressive/"self-cleaning" designs that can help.
b. The traditional plating on the probe tip is gold. But many customers find that switching to a harder material such as steel can provide a more reliable contact over extended use.
c. The probes are available with higher spring forces that can help punch through flux residue and oxidation such as found on OSP coated boards.
d. Contact vendors such as Ingun, IDI, QATech and Everett Charles Technologies for more information.
3) Use a test fixture system that supports re-cycle on fail. Most commercial automated test system vendors can provide the ability to automatically cycle the PCBA in the fixture upon a failure. Customer-developed systems can do the same if properly programmed. Some systems can cycle the fixture prior to test as an effort to punch thru flux residue.
4) Use "dry well" fixtures. Pneumatic-style fixtures or vacuum box fixtures that don’t expose the probes to vacuum-related air flow can be more reliable since the air movement past the probes can exacerbate the effect residual flux by attracting dust onto the probes.
Some customers with vacuum fixture have also found that having the tester too close to the soldering system can result in flux-laden vapors being sucked into the fixture!
Hope these ideas are useful.