Hello to Germany

Comparison of electron tubes with Pleiades pull up bias

This is just on measuring anode current for low anode potentials.


(Attention to safety. A fuse should always be used in series with a battery.)


E88CC Mullard
Heater voltage = 5V
As cathode temperature is increasing Vg reaches -600mV or more.
With pull up bias resistor Rag=10ΜΩ, Vb=12V
Ia is 300μA
At Vb=24V, Ia=2mA
At Vb=36V, Ia=4.5mA
(Va almost coincides with Vb as on most of these experiments no load resistors are used. Headphones coils, tape recording head coils, transformer could are used. Also since inductors generate EMF output voltage can be double the Vb supply).


Note: As cathode starts emitting the free electrons it begins becoming positive with respect to grid by the positive protons left behind. This is possibly the reason why grid is negative with respect to cathode. The actual negativeness of grid is greater. The readings shown were taken with a 1GΩ to 10ΜΩ digital voltmeter. If a very high impedance electrometer or pH meter is used the grid is measured more negative by a few hundred more millivolts.


In order to make the grid less negative a pull up Pleiades bias resitor of 10MΩ from anode grid was connected. This increases anode current and makes operation possible, see previous posts.


On the following electron tubes 12V were connected to heaters and 10MΩ from anode to grid unless otherwise stated.


ECC83 Tungsram
Vg=-777mV
Vb=12V, Ia=2μA (3μA when cathode is colder before Ia fades to 0)
Other Tungsram ECC83 measured 4μA
Vb=24V, Ia=23μA,
other Tungsram ECC83 Ia=41μA
Vb=36V, Ia=105μA, 108μA at slightly lower cathode temperature
other Tungsram ECC83 Ia=172μA


Mystery electron tube, it was found inside a Tungsram ECC83 box, it reads Philips ECC8?
Vb=12V, Ia=23μA
Vb=24V, Ia=126μA
Vb=36V, Ia=330μA


ECC82 Ei (Yugoslavia)
Ia=500μA
1.6mA
3mA


EF183 triode connected
at Vb=12V
1.5mA with Rag=10MΩ, 2.5mA with Rag=1.5MΩ

DC bias may not be suitable for sound on sound

Accidentally passing the tape again from the recording head (erase head disabled) did not produce very good sound in sound. AC bias seems more suitable.


More experiments needed.

Pleiades Electra 3 with 10MΩ from anode to grid

Caution: protect your ears from high sound pressure levels.


One EF183, triode connected for each channel driving the Sennheiser HD580.


Vb=12V.


Both Rag of 1.5MΩ and 10MΩ sound spectacular. 10MΩ should be of less power. In the first case the DC anode current through headphones is 2.5mA and in the latter case 1.5mA.


How would ECC88 sound with this amplifier?

Hello to France, Ελλας and to Ukraine

Driving a tape recording head directly from anode (part 5)

At your risk. Taking all safety precautions is very important.


The Pleiades V6 schematic is used. Anodes directly connected to Tascam Porta 03 recoding heads. 2 amplifiers for 2 of the 4 tracks. Cassette running at normal speed.


For the time being DC tape bias is used (the anode current itself). It gives thicker noise. But it sounds nice. Like a vinyl record fff... sound when the song comes to the end.


A bias current of about 300μA-400μA seems for the time being desirable.


One of the main problems is designing an amplifier that has a quiescent class A current of 400μA while at the same time able to give much voltage which will be needed for the high frequencies. The head is an inductive load so it will be driven by constant current. This means the driving voltage will increase by 6dB per octave with rising frequency as the amplifier is less and less loaded.


So what drives the magnetic head will have to be of high enough impedance. Electron tubes seem to be perfect for the job.


So far the EF183 electron tube had been tried at up to 24V plate supply. A Pleiades bias or pull up bias resitor from anode to grid makes operation possible. Open grid was also trisd, see previous posts.


The problem is that Ia=400μA cannot be much exceeded or tape will be over biased. But for example EF183 gives much more output and less distortion at high level at Vb=12V when bias is such that anode current is say over 1.5mA. 1500μA can be achieved by a 10MΩ resistor from anode to grid. And in fact connecting the HD580 to anodes gives an excellent sound.


But when 400μA is desired the Vb would better be lower say 5V-6V. Rag can be 4MΩ. Connecting the HD580 headphones at such operating conditions gives a nice sound but with distortion if one wants higher output. When connecting the heads the treble immediately rises or bass decreases due to the constant current inductive effect of the head. In fact the head become like a low cut inductive filter connecting across a microphone (see Pleiades filters).


But the distortion further increases as heard on the headphones. Nevertheless it is surprising that at such conditions a relatively nice recording can be made to cassette tape. Modulation goes up to about +3 VU at playback. Warm noise or hiss is buried in music. Guitars have a high class quality that reminds 60's electron tube recordings such as CBS, for example Jim Reeves recordings. But there is still treble missing.


It became apparent that a different electron tube would be better if it can give less current at higher voltage.


Today the EF89 was tried. Anode current when Philips electron tubes were used was indeed lower. But the sound on the headphones while listening to what will go to the heads sounded much more distorted and much less loud. The EF89 is less sensitive needing a higher signal driving voltage. This was tried too.. Recordings with EF89 were much more distorted Unless very low modulation was used. Treble was higher. Signal was lower. Bass hiss this time could be heard along with the music.


So the setup was returned for the time being to EF183.


After relistening to all recordings done so far the best seem to be those done as follows. EF183, with 1.5MΩ plate to grid resistor, plate supply Vb=3.6V, Ia=250μA. More modulation but less treble is when Vb=5.5V, Ia=500μA. Possibly aiming for a DC bias current between 250μA and 500μA. Voltage headroom of amplifier needs to be increased but without increasing Ia if possible?


Another candidate electron tube may be the EF86.


And there is a gut feeling that the ECC83 will be best. It can have an operating point of 400μA at high voltage. It has a high voltage amplification factor. It has a high anode output impedance. It may have to operate at a high plate supply voltage. Or lower if Pleiades bias is used.


Another consideration may be whether an electron tube can cope directly connected to an indicative load such as a recording head coil. It may. If not the usual high series resitor would have to be used between output and head. See nearly all schematics of recording amplifiers, electron tube or transistor.


Another consideration is whether what has already been done (deliberately not using extra HF boost) is adequate and all now needed is to connect the EF183 anode to the head of a machine running tape at 7.5ips. Eg Uher 4200. Or even at higher speed.

Hello to United Kingdom