[ricard]

Yes, you are right of course. I'll put it down to bad memory.

I wonder why they have separate windings for the various heaters? In the schematics I found at rm.org they are all grounded on one side anyway with no hum canceling pot. Is it to minimize some form of interaction between the valves (the EF804 and ECC81 have separate windings, and a third one for the rest).

It's interesting to look at early recorders like this one, how practices changed over the years. The TK819 I recall has one unique ground point on the chassis where a whole bunch of black wires converge. In principle this is good practice, yet later machines don't have an obvious ground point like this. I supposed designers figured out a way to do something that was good enough without running ground wires all over the place.

Another issue is the record treble lift which is accomplished with a series resonant circuit in the cathode path of the first ECC81 triode. While a resonant circuit certainly will cause quite a marked peak, later machines seem to use simpler RC circuits to accomplish this - I'm thinking in order to minimize the severe phase shift at the resonant frequency that a resonance circuit incurs. And also I'm thinking because as technology progressed, in the transistor era, once prices of components had come down, it became feasible to have multiple amplification stages rather than the relative few which were used in the valve era.

Another thing with the generation to which the TK819 belongs - reversing mode notwithstanding, the tape transport is quite simple, with no separate brakes - all braking is accomplished by the existing friction in the reel holders. Playback tension is maintained simply by using a friction disc, and as the amount of tape on the supply reel diminishes as the reel is played, the pressure on the friction disc will be lower which compensates for the higher rotational speed, in order to try and maintain a constant tension.

In fact, a principally fully remote controllable tape transport has been implemented with a single motor and three solenoids - even if the remote control actually designed for the TK819 doesn't provide all functions.

I've seen other machines from the same era having similar designs.

In fact, had the TK819 not been reversing, it still would have required a lot of the mechanics (with the exception of the intermediate directional clutches possibly). So adding the reversing function was probably not as a large step as one initially assumes.

Of course, as time progressed, the simple mechanics did not suffice in terms of braking speed etc, and mechanics got more complex with separate brakes and friction clutches for the take up reel. etc But there's something impressive about the relative elegance of early machines such as the TK819.

Ok, enough rambling. (I'm off sick from work so trying to inject some positive energy by reflecting on these things...).