Picture of the fair maiden?
Chalk,
Is that red+ black through the 2pole terminal block to blue+ brown the mains feed to the transformer?
I would recommend two further and quite separate stages of improvement:
performance
1.) twist the leads together to minimise the loop area.
2.) move the cable and secure it further away from the audio circuits.
safety
3.) double insulate the mains cable. A wrap of sticky insulating tape would do. I go for 55% overlap in bright red for mains voltage cables. The ends of the tape should be mechanically secured (tiny tie-wrap), so that over the years they do not unwrap.
secondary winding leads
4.) the three feeds from transformer to PSU should be twisted as a triplet to minimise loop area.
Is that red+ black through the 2pole terminal block to blue+ brown the mains feed to the transformer?
I would recommend two further and quite separate stages of improvement:
performance
1.) twist the leads together to minimise the loop area.
2.) move the cable and secure it further away from the audio circuits.
safety
3.) double insulate the mains cable. A wrap of sticky insulating tape would do. I go for 55% overlap in bright red for mains voltage cables. The ends of the tape should be mechanically secured (tiny tie-wrap), so that over the years they do not unwrap.
secondary winding leads
4.) the three feeds from transformer to PSU should be twisted as a triplet to minimise loop area.
Last edited:
It is far easier to tack an extra resistor alongside the existing CCS resistor/s.
It also makes it much easier to predict the new CCS current.
If the CCS resistor have 1.9V across them and you have 68r//68r then adding 47r will reduce the Vr very slightly, I'll guess and suggest it drops by 10mV.
The original CCS current is 1.9/68r + 1.9/68r = 27.9+-1% +27.9+-1% = 55.9mA +-1%
New CCS current is approximately 1.89/68+1.89/68+1.89/47 = 27.8+-1% + 27.8+-1% + 40.2+-1% ~= 95.9mA +-1%
If you remove the 68r and replace with a 5W +-5% resistor you don't know how far the CCS voltage will drop. I can assure you it will be much more than 10mV.
The 10r current be be very approximately <1.89/10+-5% =~<189mA +-5%
BTW,
the original DCB1 PCB (both Hyp & Mes) has locations for upto 6 resistors in the CCS. For a 2.4Vdrop (the most that the stack up of tolerances will allow) you can use 47r to dissipate a maximum of 123mW (<25% of a half watt resistor).
6off 47r in parallel give ~7r8. That gives a very wide range of CCS resistor to experiment with.
It also makes it much easier to predict the new CCS current.
If the CCS resistor have 1.9V across them and you have 68r//68r then adding 47r will reduce the Vr very slightly, I'll guess and suggest it drops by 10mV.
The original CCS current is 1.9/68r + 1.9/68r = 27.9+-1% +27.9+-1% = 55.9mA +-1%
New CCS current is approximately 1.89/68+1.89/68+1.89/47 = 27.8+-1% + 27.8+-1% + 40.2+-1% ~= 95.9mA +-1%
If you remove the 68r and replace with a 5W +-5% resistor you don't know how far the CCS voltage will drop. I can assure you it will be much more than 10mV.
The 10r current be be very approximately <1.89/10+-5% =~<189mA +-5%
BTW,
the original DCB1 PCB (both Hyp & Mes) has locations for upto 6 resistors in the CCS. For a 2.4Vdrop (the most that the stack up of tolerances will allow) you can use 47r to dissipate a maximum of 123mW (<25% of a half watt resistor).
6off 47r in parallel give ~7r8. That gives a very wide range of CCS resistor to experiment with.
Last edited:
Found a packet of 51R 0.5W resistors in my parts bin. So in they have gone. The sound is great and the heat sink is still cool.
The voltage drop across one set if the 51r resistors is 1.89V, but only 1.55V on the other set. The voltages at the test points are 9.36V and 9.73V.
I don't have a 15V trannie as specified in the BOM, but I do have a 12V one. So I thought I would try it. It may be the cause of the voltage imbalance. However, I'm not too worried as the sound is brilliant!
The voltage drop across one set if the 51r resistors is 1.89V, but only 1.55V on the other set. The voltages at the test points are 9.36V and 9.73V.
I don't have a 15V trannie as specified in the BOM, but I do have a 12V one. So I thought I would try it. It may be the cause of the voltage imbalance. However, I'm not too worried as the sound is brilliant!
Don't worry, the Vo imbalance in the Mez is designed in by me by voicing it. Its normal. On the resistors its a matter of Vgs and leds. PMOS & NMOS don't have the same ID/VGS curves. You can try have about same current there by trimming the resistors (Vrset/Rset=ICCS), but it won't change anything audibly at that range.
I have taken measures for when a volume pot goes bad, there is a parallel resistor to keep the input referenced, only disclaimer in a full DC system is if a hard failure ever happens and one PSU dies. Then full opposite rail comes to output. Then your DC protection must show its worth. Same if a one sided failure happens in any complimentary output solid state power amp. I.e. 99% of solid state amps out there. Also all source components must be checked are producing no DC before connected to a full direct system like yours.
Yes, this is a deliberate choice. For me there is no point in protecting the system, if there's a chance a component after the protection will fail. A coupling cap won't help me if the output of the power amp hugs one of the power rails due to a failure.
The actual circuit takes care of what it needs to, thank you Salas. I still have 1 more of them to make, actually
The actual circuit takes care of what it needs to, thank you Salas. I still have 1 more of them to make, actually