DHT filament bias

[vetmed]

Anyone tried this with the 3S4? That is, returning ground side of the cathode through a resistor to ground to establish bias. Negative of D cell would go to ground. For 2 volt bias and 100 ma filament current you would need a 20 ohm resistor. Would dissipate ~ .2W so you could get away with a 1/2 watt resistor but a 1 watt would probably be better. Would eliminate the electrolytic cap in the cathode circuit as well. Thoughts and opinions greatly appreciated

Robert Lees

[Paul Birkeland]

Negative of D cell would go to ground. For 2 volt bias

How do you plan to get 2V of bias from a 1.5V battery?

[Doc B.]

FWIW there are D cell sized 2V SLA batteries.

[vetmed]

Clearly I have not made myself clear With the Quickie you have one 1.5 volt battery. Positive side of this battery goes to pins 1 and 7 of the 3S4. Terminal 5, the negative end of the filament/cathode, connects to ground via a 20 ohm resistor. The negative end of the battery is also grounded. Of course you have to remove the 1K resistor and its accompanying capacitor. 2volts/100ma gives 2 volt bias and no cap in the cathode circuit. Only down side I can see is that such a setup will approximately halve battery life, but paralleling another D cell is easy enough

Robert Lees

[Paul Birkeland]

Clearly I have not made myself clear With the Quickie you have one 1.5 volt battery. Positive side of this battery goes to pins 1 and 7 of the 3S4. Terminal 5, the negative end of the filament/cathode, connects to ground via a 20 ohm resistor. The negative end of the battery is also grounded. Of course you have to remove the 1K resistor and its accompanying capacitor. 2volts/100ma gives 2 volt bias and no cap in the cathode circuit. Only down side I can see is that such a setup will approximately halve battery life, but paralleling another D cell is easy enough

To get 2V of bias, you'd need to start with 3.5V, feed this in to pins 1/7, then use a 20 Ohm resistor to ground from there.   This would give you 1.5V of filament drop at 100mA, and 2V of bias across the resistor.

[vetmed]

PB you say "...feed this in to pins1/7, then use a 20 ohm resistor to ground from there." Not sure I understand, if there is already 1.5 volts there then why not just ground the negative end and call it biased at 1.5 volts? Forgive me if I am being thick about this

Robert Lees

[Paul Birkeland]

You need 1.5V across the filament, and 2V from filament to ground, this is 3.5V.

If you put a 20 Ohm resistor between the filament and ground, then put 1.5V across the filament and resistor in series, there will not be enough voltage across the tube to heat it. 

Perhaps if you draw a schematic of what you're looking to do, then fill in the voltages and currents, things will shape up.

[vetmed]

Well, I did one better, 20 ohm resistor in series does indeed not work So I connected the D cells in series and get 3.08 volts at the + end of the filament, and 1.88 volts at the top of the 20 ohm resistor So I hooked it up and it works Its late and I will try and make sense of this tomorrow Thanks Paul

Robert Lees

[Paul Birkeland]

It's all about learning here!

With four D-batteries, this should be fun to try out.

-PB

[Paul Joppa]

Well, I did one better, 20 ohm resistor in series does indeed not work So I connected the D cells in series and get 3.08 volts at the + end of the filament, and 1.88 volts at the top of the 20 ohm resistor So I hooked it up and it works Its late and I will try and make sense of this tomorrow Thanks Paul

Robert Lees

Cool - I look forward to your report on the sound difference!

Notice that this leaves only 1.20 volts on the tube filaments, which are probably good down to 1.00 volts. As the D cells drain and voltage drops, the filament will run cooler, giving them less resistance, so the filament voltage will drop more rapidly than the D cell voltage.  Running 3 volts power, I would suggest a 15 ohm resistor for bias in order to get reasonable use from the cells.

[vetmed]

Actually the thought occurred to me as well, but it was more of an issue of not flattening the batteries quite as quickly Understand that the setup is not optimal right now with the 2 batteries powering both channels, will be making a trip to the local electronics surplus store for battery holder(s). Still and all sounds very promising and will report back when everything is running more to my liking

Robert Lees

[vetmed]

For me this is an unqualified success. Installed 2 single D cell holders directly underneath the original ones, had to use slightly longer bolts and the CCS now just dangles in space. Put some velcro strap on the holders to keep the batteries from falling out Pretty straightforward and getting the electrolytic cap out of the way has eliminated a slight coarseness that frankly prevented me from using this preamp on a regular basis. Now have VPI TT, Eros, Quickie, 2A3 Paramount and this pre is not out of place with much pricier gear Very pleased with this mod and since it is easy to do I would recommend it. And easily reversed if you don't like the results.

Robert Lees

[johnsonad]

Nice work Robert! Hopefully others will give this a try!

[Hierfi]

My friends all have quickies and I must make amends. I have heard these units and like the simplicity and performance. This inspired some thought that resulted in some experiments on directly heated cathodes connected to ground. This was done mainly to eliminate the electrolytic and simplify the circuit.   

After some testing on the 3S4 I am currently experimenting on a single dual triode tube for stereo. The tube is the directly heated 3a5. I have no idea what reputation this tube has for audio, however it seems that the tube has some merit in just having fewer grids to contribute to the microphonics. After several experiments the power supply voltage used was set to 20 volts with the plate voltage at 10 volts. The plate load resistor was fixed at 470K, requiring a plate current of 20uA. Arguably these parameters are not at all useful as a stand alone preamplifier.

Nevertheless, to obtain such a result the center tap of the heater was attached to ground and the ends connected together (pin 1 and 7) and then connected to an adjustable positive current source. To set the quiescent tube voltage to 10 volts and tube current to about 20uA required the heaters to be set to around 130 ma (65 ma each tube). This resulted in the voltage being around 0.67 volts. Ultimately the heaters are running about 1/2 voltage and 1/2 current, or 1/4 power. Given that electron emissions are a function of temperature this drastically reduces the electrons boiling from the heater.   

Nevertheless, the result was that the output AC gain was about 10X and the output swing was remarkably close to 1 volt off ground and 1 volt from the 20volt supply. In other words the 20 volt supply was adequate to produce almost a 20 volt peak to peak swing. The limitations near the B+ supply and ground was also gradual, resulting predominantly in low order harmonic distortion components.

In further testing the output impedance was estimated to be around 100Kohm, appearing dominated by the characteristic impedance of the output tube, not the plate load resistance. This suggests that the 470Kohm plate load resistor was behaving more reminiscent of a constant current source from the B+ supply. Like a constant current source this offers increased isolation from the noise of the B+ supply getting into the output.
 
Having no aversion to bipolar devices I would like to add a discrete bipolar push pull output stage and keep the total quiescent B+ current to about 50uA per channel As a result I should be able to get away with 3 x 9volt batteries lasting perhaps more than 10x longer than the quickie and a single D cell or Lipo for the heaters.

I don't know what to expect though it looks good on the scope.

cheers 

[Paul Birkeland]

20uA on a 3A5 is nearly 0 current.  You could also reduce the plate current in the Quickie to 20uA to increase battery life, but the circuit's performance would go to hell.

I've used the 3A5 on several occasions, I usually use 2.5V bias, 10mA and ~125V on the plate.

As you can see, with no plate current and no plate voltage, the circuit's output impedance has become so high that it's not able to really act as a preamp.  I would go as far as to say that this qualifies as a class B preamp. 

With one end of the filament grounded and the other end being supplied by a current source, one end of the filament is at zero bias, which will further degrade the performance. (Especially depending on what's feeding the grid of the 3A5)  You actually have a preamp that requires a source with low impedance to drive it (100 Ohms), yet the gain stage itself has incredibly high output impedance. 

I don't mean to sound overly negative, just trying to highlight all the trades you have made to get rid of one capacitor.