• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Parallel feed OT

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi All,
I am considering building something like Kandaudio's SE parafeed preamp. But don't want to order the Lundahl, because I want to see if my tranny builder can build me something like that.

So what do I have to ask for.. in the case of say a 6n1p tube..paralleled both sections Ra of 2200ohm..so do I ask for a 7k SE transformer that does not have to pass a single mA of DC?

Regards,
Bas
 
Hi Bas,

I believe that you are right with the transformer impedance and current specification.
The trouble will come in the way you supply DC to the anodes.
Are you going to use a choke? If so, then you must consider the choke's self capacitance and the transformer's self capacitance to be in parallel.

Cheers,
 
6C41c

Hi Bas ,
CCS loading , to me means low Ra valves , run at high current , low anode volts . I find that CCS loading usually means ridiculous heat , my own MTP2P50E based mosfet sink ran at 70mA and got REALLY hot , and that was with a massive heatsink , and not really dissipating as much as it should (I kept the B+ within safe limits) . Why not try something like stacked 6C41c in an SRPP / SEPP arrangement with fixed bias to the lower valve ? It would be cheap to build , only the power supply needs to have higher HT than with choke loading (400V+) . My usual advice regarding parafeed prototypes and toroids as output transformers applies , could make a nice , cheap project .

316a
 
Why not try something like stacked 6C41c in an SRPP / SEPP arrangement with fixed bias to the lower valve ?

Might very well still try that...I still have the little toroids lying about idly after the 6c45pi spud thing failed..

mmm, come to think of it..I have seen 6c33 preamps..maybe a 6c41c preamp is not even such a bad idea? Nah, 6s19 is wild enough..

Cheers,
Bas
 
Could some engineer translate this message from a tube god? (Paul Joppa of bottlehead fame, and parafeed expert.)

Is that not the Japanese word for engineer..i.e. one who translates the god's work into human language. Question for you Dhaen. ;-)

Don't be fooled by the data sheet. Look at the curves at the operating point you are contemplating, and you will see that the plate resistance is much higher than advertised. At 250v/8mA it is around 8k per triode, for example.
To spec a transformer, determine the load impedance for which the amplifier is designed assuming a resistive load. That is the "nominal" transformer impedance, meaning that the inductive reactance should be larger than the target impedance above whatever minimum frequency you want to operate to. Tell the transformer designer the source resistance (plate resistance) also, so that winding resistances can be chosen for properly damped HF resonances. Tell them the design maximum signal voltage also, so the the core can be optimally sized.

Posted the same question at the AA.
 
Plain English

Hi Bas ,
I think what Paul is saying is that transformer design is a black art that needs to be specified carefully . He is right about the 6N1p (especially Svetlana) curves , particularly misleading . Why use a 6N1p if the Ra is 8k ? Why not use a 6SN7/7N7 instead ?

316a
 
diyAudio Senior Member
Joined 2002
Ra.

Hi Bas,

Don't be fooled by the data sheet. Look at the curves at the operating point you are contemplating, and you will see that the plate resistance is much higher than advertised. At 250v/8mA it is around 8k per triode, for example.

What Paul Joppa means is that Ra varies according to the operating points you choose, hence when you need to specify a xformer, keep in mind these points and don't go for the Ra given in the datasheets as if it were a constant for it is not.

Not that it is necessarily all that critical...I'd need to read the whole article to see if this is actually what he meant.

Hope this clarifies it a bit,;)
 
Hi Frank,

The part you explained I could still understand...it is this part that is greek to me.

To spec a transformer, determine the load impedance for which the amplifier is designed assuming a resistive load. That is the "nominal" transformer impedance, meaning that the inductive reactance should be larger than the target impedance above whatever minimum frequency you want to operate to. Tell the transformer designer the source resistance (plate resistance) also, so that winding resistances can be chosen for properly damped HF resonances.

Cheers,
Bas
 
diyAudio Senior Member
Joined 2002
SPECS.

Hi Bas,

To spec a transformer, determine the load impedance for which the amplifier is designed assuming a resistive load. That is the "nominal" transformer impedance, meaning that the inductive reactance should be larger than the target impedance above whatever minimum frequency you want to operate to. Tell the transformer designer the source resistance (plate resistance) also, so that winding resistances can be chosen for properly damped HF resonances.

From what I read here I assume Paul is referring to a SE OPT:

To spec a transformer, determine the load impedance for which the amplifier is designed assuming a resistive load.

As an example,let's assume the amplifier was designed to work into an 8 Ohm load.

That is the "nominal" transformer impedance, meaning that the inductive reactance should be larger than the target impedance above whatever minimum frequency you want to operate to.

So the inductive reactance should be larger than the 8 Ohm impedance for a given bandwidth,say from 15 Hz to 25 kHz.

Tell the transformer designer the source resistance (plate resistance) also, so that winding resistances can be chosen for properly damped HF resonances.

And in the last part he says it all:

The xformer designer can now choose the winding resistances from the data you provide in order to keep the xformer HF resonances well damped.

In other words: give the designer all the operating parameters you determined for the outputstage + the load into which this should work best for a given bandwidth so he can avoid ringing of the OPT inside this band.

I hope this is what he meant,;)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.