Make that "IN" sign a coupling capacitor, otherwise you might end up with scratching potentiometer.
Is there any particular reason why you insist in capacitor coupling and not using SE OPT ?
Is there any particular reason why you insist in capacitor coupling and not using SE OPT ?
see the discussion beforehand, the jury is out on which is best. If i went SE OPT then the PS would need to be regulated.
You make it sound as if that was a bad thing. You're using two active devices to load each output tube as it is (4 total), simple series "capacitance multiplier" type of regulator would only take one device and shunt circuit to deal with AC distortion one or two more.
Looks about right to me. I'd use 10R plate and cathode stoppers and put a small film cap at the drain of the upper fet to keep the CCS stable. Also, a regulator before the CCS is helpful. Operating point you'll have to play around with.
are the d3a's quite hard to get stable?
I think with the B+ reg, I'll build it so i can insert it with a module and try with and without. I have a few TIP50's around to build a simple reg.
I am going to order stuff... anyone know of a european d3a stockist?
I think with the B+ reg, I'll build it so i can insert it with a module and try with and without. I have a few TIP50's around to build a simple reg.
I am going to order stuff... anyone know of a european d3a stockist?
Also just forget DHT unless you want to pay for the tent supplies.>>
Well, first of all nothing has the delicacy and tone of a DHT, and second nobody needs to use Tent filament supplies. The Ronan Reg uses humble LM1084s and even better is just a choke input supply using a Hammond choke. If you really want the sound this is still the way to go!
Andy
Well, first of all nothing has the delicacy and tone of a DHT, and second nobody needs to use Tent filament supplies. The Ronan Reg uses humble LM1084s and even better is just a choke input supply using a Hammond choke. If you really want the sound this is still the way to go!
Andy
They can be. I've found that teflon sockets on a PCB with SMD resistors right on the pins solves the issue. But, I've also built with them PTP and so long as you are careful with lead dress and getting the stoppers right on the pins (both cathodes), you should be OK.
thanks for your input - extremely helpful. I have ordered 4 x d3a, will order the sowters tomorrow. Do you think 10k on the primary is about right?
tranformers + power tx and choke order from jack at electraprint.
Tubes have arrived. Css modules built and dialled to 15ma.
Project is go
Tubes have arrived. Css modules built and dialled to 15ma.
Project is go
My latest, still on the breadboard...
[URL=http://www.directupload.net]
[/URL]
Even with small vintage Grundig OPT's sounds gorgeous, very detailed and fast... brings you very close to the music. Custom made output transformers arrived today 😉
The whole thing eventually will end up like this:
[URL=http://www.directupload.net]
[/URL]
[URL=http://www.directupload.net]
[/URL]Even with small vintage Grundig OPT's sounds gorgeous, very detailed and fast... brings you very close to the music. Custom made output transformers arrived today 😉
The whole thing eventually will end up like this:
[URL=http://www.directupload.net]
[/URL]Building a similar amp only different output tube, on your 3D image I'm not seeing a heatsink, I think you're going to need big heatsink for those mofset.
Thanks for the picture, our amps will be such similar topology. But looks like you are using only one shunt supply? I was planning on one per channel as I thought I read there would be instability supplying 2 channel with one reg, would be a much less complicated build with one Salas reg.
regal,
there is absolutely no issue with supplying both channels with one Salas reg.
My Aikido (parafeed btw) is also fed by a Salas reg and performing beautifully for almost a year now. Compared to the previous psu which was CRC soundstage with even one reg has improved considerably. I don't really understand why but now noticed the same with my C3G-spud pictured above.
As for the question parafeed or SE... I couldn't tell a favourite. In a tube amp there are so many variables that the choice of output topology is just one of many... for sure I prefer transformers over anything OTL now... 😉
there is absolutely no issue with supplying both channels with one Salas reg.
My Aikido (parafeed btw) is also fed by a Salas reg and performing beautifully for almost a year now. Compared to the previous psu which was CRC soundstage with even one reg has improved considerably. I don't really understand why but now noticed the same with my C3G-spud pictured above.
As for the question parafeed or SE... I couldn't tell a favourite. In a tube amp there are so many variables that the choice of output topology is just one of many... for sure I prefer transformers over anything OTL now... 😉
Hello,
The power supply is a important piece of the amplifier. The higher the impedance of the power supply the more the voltage dances with the music. The impedance converts current to voltage. The voltage change is audible.
I have been using a TubeCad PS-2 Solo to power a SET headphone amplifier. The PS-2 is a huge improvement over the CLC approach. Plus there is no hum or buzz.
Next is to try a Salas HV shunt regulator. However the shunt will require a transformer with a much greater VA rating. Shunts use 3 to 4 times more amps.
DT
All just for fun!
The power supply is a important piece of the amplifier. The higher the impedance of the power supply the more the voltage dances with the music. The impedance converts current to voltage. The voltage change is audible.
I have been using a TubeCad PS-2 Solo to power a SET headphone amplifier. The PS-2 is a huge improvement over the CLC approach. Plus there is no hum or buzz.
Next is to try a Salas HV shunt regulator. However the shunt will require a transformer with a much greater VA rating. Shunts use 3 to 4 times more amps.
DT
All just for fun!
Last edited:
This is only true for crappy designs that employ brute force approach rather than considering the problem. There is no point in burning multiple times the DC current. What a decent shunt regulator does is that it eliminates (= burns) any AC riding on top of DC. DC can be obtained at considerably lower 'cost' in energy by series regulation but superimposed AC cannot be avoided due to nonlinear consumption of device, nonideal conductors and so on.
I am pretty sure one of the linear IC manufacturers has a very good appnote posted on this subject (Linear or Analog ? I'm 95% certain it is one of these two).
Hello,
Please tell me where I go south. I do this for fun; it is not my day job.
Voltage regulators are voltage dividers. Series regulators adjust the top resistor. Shunt regulators adjust the bottom resistor. Shunt regulators can maintain better regulated cleaner voltage than a series regulator because they can shunt more current to ground. Reduce the current in the shunt and the quality of the regulator is reduced, there is a trade off. Perhaps there is a balance point between green and good enough.
I will put one together and play with it.
DT
All just for fun!
I now have most bits, jack got the transformers done quickly, and they look superb.
Does anyone have any recommendations of wire for the power umbilical. Could i get away with mulitcore (16 gauge) , or would you advice multiple single cores with a large gauge for the ground?
I am going with these connectors as they seem good value. going 5 way, B+, G, DC Heater x2 + spare (just incase...)
Does anyone have any recommendations of wire for the power umbilical. Could i get away with mulitcore (16 gauge) , or would you advice multiple single cores with a large gauge for the ground?
I am going with these connectors as they seem good value. going 5 way, B+, G, DC Heater x2 + spare (just incase...)
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
- Status
- Not open for further replies.