Can I trust my Valve Tester

[gingertube]

I bought a huge box of new and used radio and television valves recently and have been putting them through my Taylor 45D Valve Tester.

As expected I got ano more than 10% recovery rate on the used tubes.

This was until I started to test the 26 6CM5 tubes. I tested cathode leakage, gm and anode current.
Of these I got:
6 tested 100% or better
6 tested 95%
6 tested 90%
1 tested 85%
2 tested 80%
2 mechanically damaged
2 with gm <80%
1 with marginal cathode leakage

That is 19 of the 26 tested what I would deem acceptable. 18 of them tested as good as the 4 NOS items I had on the shelf.

The problem with the Taylor 45D and possibly other Valve Testers is that on bigger power tubes they invariably test these at low voltages - in this case 100V Anode and Screen.

These tubes were all pulls from TV Horizontal Deflection Amplifier.
Should I assume:
1) There is something wrong with them that my Valve Tester is'nt revealling - should I retest cathode leakage at 250v Anode and Screen for instance.
2)They came from a particular TV repair man who replaced 6CM5s as a matter of course without bothering to check them.

Cheers,
Ian

[EC8010]

I'm not familiar with the Taylor tester but I treat most valve testers with suspicion - including my AVO VCM163. Nevertheless, I'd say all of those valves look fine. It's time to reject a valve when heater/cathode insulation falls below 1M, and even then, it may not matter in some circuits. For example, poor heater/cathode insulation matters in the input valve of an RIAA stage but may not matter in a grid-biased output valve. I doubt if retesting with a higher anode voltage will make the slightest bit of difference.