Hello,
The nemesis amplifier designed by Jean Hiraga ( one fet, single power supply, output transformer) uses the same kind of technique.
One single thick copperr wire is al it takes. Usually i will take wire from an old crossover coil.
The virtue amp i did buy will come with a 4 pole panasonic cap. Normally it includes a seperate switching power supply. But the person that is selling did use batteries. I don't wanna invest big money in caps. One reason i don't have to much space in the outboard power chassis and it will only be used with an 80 hertz 6 db filter at its input.
The chokes will give me clean power i hope. I think i will go for a few 2200 or 4700 mf caps. Maybe later one or two 22.000mf. Please take a look here.
RR#002 - DC Filament Supply Test
I did include a copy from the internet about nemesis. I have all the magazines but i think it is possible to find it on the net too
Many thanks for the advice, Edward
The nemesis amplifier designed by Jean Hiraga ( one fet, single power supply, output transformer) uses the same kind of technique.
One single thick copperr wire is al it takes. Usually i will take wire from an old crossover coil.
The virtue amp i did buy will come with a 4 pole panasonic cap. Normally it includes a seperate switching power supply. But the person that is selling did use batteries. I don't wanna invest big money in caps. One reason i don't have to much space in the outboard power chassis and it will only be used with an 80 hertz 6 db filter at its input.
The chokes will give me clean power i hope. I think i will go for a few 2200 or 4700 mf caps. Maybe later one or two 22.000mf. Please take a look here.
RR#002 - DC Filament Supply Test
I did include a copy from the internet about nemesis. I have all the magazines but i think it is possible to find it on the net too
Many thanks for the advice, Edward
Attachments
post13, amp layout 1 -wiring diagram.
The Live from the IEC must go direct to the fuse. NOT strung around most of the chassis until it eventually goes to the fuse.
It is better/safer to connect the IEC earth wire direct to chassis very close to the IEC socket using a very short length of grn/yel cable.
Fuse each transformer with it's own close rated fuse.
In a dual mono, each channel should be built entirely separate from then other.
This requires that the Main Audio Grounds are NOT connected together.
Each MAG is required, for safety reasons, to be connected to chassis. Use a Disconnecting Network (DN) for each channel, ie. two DN are required for a two channel amplifier.
Consider adding fuses in the DC supplies to the amplifiers. Note that some amplifiers will offst to rail if only one fuse blows. Find out if your amplifier does this. And put the fuse in the other rail to help avoid blowing up your speaker. Use a double rated fuse in the rail that you don't want to blow.
Bring the rectifiers much closer to the transformer secondary output wires. Then connect the rectifier to ONE end of the capacitor bank. Connect the amplifier to the OTHER end of the capacitor bank.
All the wiring on each route from mains input to transformer to rectifier to smoothing bank to amplifier must be twisted as flow and return pairs.
The Live from the IEC must go direct to the fuse. NOT strung around most of the chassis until it eventually goes to the fuse.
It is better/safer to connect the IEC earth wire direct to chassis very close to the IEC socket using a very short length of grn/yel cable.
Fuse each transformer with it's own close rated fuse.
In a dual mono, each channel should be built entirely separate from then other.
This requires that the Main Audio Grounds are NOT connected together.
Each MAG is required, for safety reasons, to be connected to chassis. Use a Disconnecting Network (DN) for each channel, ie. two DN are required for a two channel amplifier.
Consider adding fuses in the DC supplies to the amplifiers. Note that some amplifiers will offst to rail if only one fuse blows. Find out if your amplifier does this. And put the fuse in the other rail to help avoid blowing up your speaker. Use a double rated fuse in the rail that you don't want to blow.
Bring the rectifiers much closer to the transformer secondary output wires. Then connect the rectifier to ONE end of the capacitor bank. Connect the amplifier to the OTHER end of the capacitor bank.
All the wiring on each route from mains input to transformer to rectifier to smoothing bank to amplifier must be twisted as flow and return pairs.
Last edited:
Hello,
Now i know how to connect the ground wire but i got confused trying to use psud. The transformer is 300 va with one 30 volt secondary winding. Serie restance of this winding is not measurable withj a normal dnm so i guess it is well below 1 ohm. I wanna use a bridge with 4 mur 860 ( maybe double them to upgrade the current rate )
The virtue amp has one 27000mf pansonic t-ha on the circuit board.
I wanna make it choke input with the above mentioned choke and bleeder across the first cap. I wanna put a 27000 there as well. Then a second choke and then another 27000 which will be in parallel with the one on the board. The 2 chokes are identical but one is in a choke input position with the extra current draw because of the bleeder. So i think there behaviour will not be indentical also because the second stage has double the capacity.
Is there anyone who can give me an idea if this set up will work properly. Maybe i should reduce the capacity after the first choke?
Many happy thanks in advance, Edward
Hello,
For those who can read French ( i can ) a cpoy from an article from Jean Hiraga about building the famous Hiraga 20 watt from the early eighties. One of the first amps they did connect the positive and negative leads coming from the first capacitors to one of the 2 big caps ( in fact they are all big, the 30 watts versions did use 330000mf caps ) and make it leave at the other cap which are connected by a copper bar.
Then they did connect the wires in the middle of the bar and it was better. They say because the caps are really big one should take extreme care how and where to connect the wires.
Does it make sense? But the french audiophiles which in fact did get much of there ideas from Japan did do a lot of good things for diy audio.
Happy reading, Edward
For those who can read French ( i can ) a cpoy from an article from Jean Hiraga about building the famous Hiraga 20 watt from the early eighties. One of the first amps they did connect the positive and negative leads coming from the first capacitors to one of the 2 big caps ( in fact they are all big, the 30 watts versions did use 330000mf caps ) and make it leave at the other cap which are connected by a copper bar.
Then they did connect the wires in the middle of the bar and it was better. They say because the caps are really big one should take extreme care how and where to connect the wires.
Does it make sense? But the french audiophiles which in fact did get much of there ideas from Japan did do a lot of good things for diy audio.
Happy reading, Edward
Attachments
Hiraga's two PSU have completely different grounding schemes.
The left most (fig6a) is completely wrong.
Try tracing the route of the charging current passing from transformer through the rectifier to the smoothing capacitors and back to the transformer. It is terrible. I hope the french is telling us that message.
The right most (fig6) shows the first smoothing caps connected directly to the centre point (PSU Zero Volts) much much better, but still flawed.
The left most (fig6a) is completely wrong.
Try tracing the route of the charging current passing from transformer through the rectifier to the smoothing capacitors and back to the transformer. It is terrible. I hope the french is telling us that message.
The right most (fig6) shows the first smoothing caps connected directly to the centre point (PSU Zero Volts) much much better, but still flawed.
Andrew, fully agree with the discussion in this thread on earthing points of smoothing caps, but my drawing is just on two different setups for rectification, one of which is better than the other. Did so because the Hiraga drawings show the least preferred option.
Hope your summer on the Scottisch border is better than ours on the shores of the North Sea.
vac
Hope your summer on the Scottisch border is better than ours on the shores of the North Sea.
vac
Hello Andrew,
Yes, they are telling the right one is better than the left one. They say it is not about the low frequency noise getting reduced but it is like a bzzz sound as they describe it that is being reduced.
In the eighties i did buit the 30 watt Hiraga Le classe A with 100000uf as the first cap, then a 200mhenry 2.5 A choke and then 2 times 330.000 uf. The biggest part of the cost were in the power supply.
As far as i remember i did built it like the right attachment. At that time i did need some power because i was using big transmission line loudspeakers. I remember it a very musical effortless amplifier.
Now i am using altec lansing voice of the theatre so i can drive them with single ended tubes.
I don't know if there have been other attempts to improve the power supply. The improvement i did make was using the choke that the French did develop for the nemesis and later also did use for a 50 watt classe a amplifier.
Sincere greetings, Edward
Yes, they are telling the right one is better than the left one. They say it is not about the low frequency noise getting reduced but it is like a bzzz sound as they describe it that is being reduced.
In the eighties i did buit the 30 watt Hiraga Le classe A with 100000uf as the first cap, then a 200mhenry 2.5 A choke and then 2 times 330.000 uf. The biggest part of the cost were in the power supply.
As far as i remember i did built it like the right attachment. At that time i did need some power because i was using big transmission line loudspeakers. I remember it a very musical effortless amplifier.
Now i am using altec lansing voice of the theatre so i can drive them with single ended tubes.
I don't know if there have been other attempts to improve the power supply. The improvement i did make was using the choke that the French did develop for the nemesis and later also did use for a 50 watt classe a amplifier.
Sincere greetings, Edward
BTW
Directly on the topic of this thread: small inductors in the PS can be used to get rid of switching pulses or RF, but beware of using large inductors in a power supply that has to deliver fluctuating current. The more efficient an amplifier, the more current demands fluctuate.
Therefore, in a power supply for a class A, inductors may be used, for class AB it is not recommendable, for class D it could be suicide. The reason is that coils store energy in the form of a magnetic field, the strength of which is determined by the current running through it. This not only means that a ps with a heavy coil will sag if current is demanded (energy is required to build up the magnetic field), but if the current demand suddenly diminishes, there will be a voltage spike. The energy stored in the magnetic field is suddenly released, causing this. Since class D usually is operated close to its maximum supply voltage, this spike could easily fry your device.
vac
Directly on the topic of this thread: small inductors in the PS can be used to get rid of switching pulses or RF, but beware of using large inductors in a power supply that has to deliver fluctuating current. The more efficient an amplifier, the more current demands fluctuate.
Therefore, in a power supply for a class A, inductors may be used, for class AB it is not recommendable, for class D it could be suicide. The reason is that coils store energy in the form of a magnetic field, the strength of which is determined by the current running through it. This not only means that a ps with a heavy coil will sag if current is demanded (energy is required to build up the magnetic field), but if the current demand suddenly diminishes, there will be a voltage spike. The energy stored in the magnetic field is suddenly released, causing this. Since class D usually is operated close to its maximum supply voltage, this spike could easily fry your device.
vac
Last edited:
Power supplies must be designed. This simple fact seems to escape some people. The design needs to take account of things like desired hum suppression, likely signal frequency and envelope frequency currents. Even the signal source may need to be considered: a subsonic resonance in the PSU which coincided with arm-cartridge resonance (or 33rpm) might be a problem for a system with a turntable but of no consequence for a CD-based system.
The PSU choke is intended to block hum and signal currents, but pass envelope currents. This means it has to be roughly the right inductance, not that it has to be omitted. Design?
The PSU choke is intended to block hum and signal currents, but pass envelope currents. This means it has to be roughly the right inductance, not that it has to be omitted. Design?
Sure. That must be the reason that ClassA needs such fat heatsinks. Plse check your facts.
vac
Hello,
There is design in everything of course. Ikea has design but will that table be helpful if the owner needs to change a light bulb. It isn't designed to support the owner
Most of the members here know something that can be helpful to others. But there are just a few who have all the knowledge it needs to make a state of the art design.
I know about making stainless steel or any other material to make a sturdy chassis. Most diy is housed in a far from good '' kind of box ''
I depend upon others to have a clue about how to implement parts in a future project. For me it is a long time hobby not a means to make a living.
Still trying to read between the lines in real books and on the internet to get some guidance.
There is design in everything of course. Ikea has design but will that table be helpful if the owner needs to change a light bulb. It isn't designed to support the owner
Most of the members here know something that can be helpful to others. But there are just a few who have all the knowledge it needs to make a state of the art design.
I know about making stainless steel or any other material to make a sturdy chassis. Most diy is housed in a far from good '' kind of box ''
I depend upon others to have a clue about how to implement parts in a future project. For me it is a long time hobby not a means to make a living.
Still trying to read between the lines in real books and on the internet to get some guidance.
Signal current is usually dealt with by the last PSU cap, or the decoupler nearest the amp, or the OPT (for P-P). Envelope current, which I think is what you two are arguing about, is handled by some combination of last PSU cap and the feed to there, which may include a suitably sized inductor. Even Class A generates some envelope current from second-order distortion, although not as much as AB or D.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.