Telefunken AC701 Primer

The first chapter, below, is condensed from Oliver Archut's historical knowledge of the AC701. What follows is a summary of my experience with this tube.
"In the beginning, Telefunken had two different AC701 versions, both low noise but the version with the 'K' designation was a frame grid construction vs. a round grid for the none-K. 'K' stands for 'klingarm' (low microphonics) for the frame grid version. The AC701 without the 'k' was made for the audio section in AEG-Telefunken handheld two way radios for the West German army, as well as some other military equipment.
"Neumann only used AC701k tubes that were pre selected at Telefunken's factory in Ulm (Southern Germany), and then Neumann selected them again for random noise and microphonics; those tubes that passed, got the white paper wrapper with a Neumann ser.# around it. The "k" versions that didn't make it at Telefunken became standard AC701.
"Since 1965 Telefunken only made frame grid AC701, so there was no need to specially mark them. But careful: Telefunken also sold some East German RFT Neuhaus-made ones under the "Telefunken" name that can be quite noisy and more microphonic than the Tele-made ones. Those ones had a white vs. a red indicator dot for identification of the anode/plate-filament wire.
"The AC701 works nearly with any gain and mostly for impedance conversion so distortion test is not an issue here. I never had any problems with high distortion AC701 tubes, more often with noise, rather than microphonics. 

"You can sometimes save a noisy AC701 via ionization discharge over a 50 Meg resistor, but you have to come up first with a 1KV power supply and then the chance is about 30% that it will work properly over a longer time. That process is also very dangerous: you may get shocked!
"You can decipher the tube's date of manufacture:
 If you have 'U' (for Ulm) and two letters, sometimes with two numbers, it is made before 1965. If there is a 'U' and seven numbers, it was made after 1965.

Adding to Oliver's primer, here is some empirical data, from my 30+ years experience with the tube:

1. On average, two out of three non-Neumann selected AC701 (mostly non-k) are not usable in microphones: over minutes, hours, sometimes days (see #4 below), the tube will develop steady white noise or irregular discharges beyond an acceptable level for critical recording situations.

2. One out of three or four non-Neumann-select k are unusable, for the same reason.

3. About one out of six Neumann-selected AC701 (in Neumann-box, with Neumann serial number on paper band around tube, or Neumann number hand written in white ink) is unusable.

4. Never judge an AC701's health before at least a 48hr. burn-in. That means, the tube is installed in the mic for which it is intended, and voltage-adjusted to the proper operating voltages; then the mic is left on for at least two full days.
Tubes which were quiet for the first hour of testing will sometimes develop discharges later on, especially once the tube has reached its thermal stability (thermal stability = operating temperature of the tube, once the tube's heat dissipation through heat sinking around it has stabilized, and the tube's temperature and its surroundings won't get any hotter).

5. Measuring noise, whether steady-state or sporadic, in tubes to be used in super-high input impedance circuits, is not possible with conventional testers. Those that are designed to work in that application are extremely rare, and cost a fortune, even used. And even then, certain types of noise are not captured, which accounted for a high percentage of returns from Neumann back to Telefunken during the period AC 701s were manufactured. (As far as I know, only two testers were in existence during the manufacture of the tube- one owned by Schoeps, one by Telefunken.)

That's why I never bothered to invest in a regular tube tester, but still use the most reliable and cheapest method to test a microphone tube: ear detection. Wire up the tube, plug in, set the correct voltages, let run for seven days (I use seven, not two!), and I know that 90% of all tubes tested this way usually will remain quiet for at least a year or longer.

6. Unlike tubes which you plug in, you cannot easily test an AC701 for noise or other functions outside of the mic, for example with a tube tester. This tube needs to be soldered into the mic for testing. That is why:

7. Unfortunately, you rarely will be able to return AC701 tubes to the seller once you have installed them, because the seller cannot easily sell an AC701 to the next customer as new: its filament wires will have already been cut and twisted for installation. This is especially true for Neumann KM-series mic installations.

7. Please note that the AC 701 is very sensitive to oversupply of heater voltage and can easily self-destruct under those circumstances. Therefore many dealers are reluctant to sell a tube worth several hundred dollars with a money-back warranty unless they have control over the installation, and are satisfied that the installer knows how to properly heat-sink the tube during installation, and that the power supply was correctly adjusted before and right after installation, to minimize stress on the tube.

8. I adjust my AC701 buying prices using the rule of thumb about the AC701's failure rate explained above: I discount the price per tube so that the percentage of failure is reflected in the price. That way I don't need to bother with money-back warranties which few sellers will give anyway: If the price per AC701k tube is, say, $700*.- with no money back warranty, I pay $400.*- to $500*.-

* prices used for demonstration only