advertisement


QUDOS - the brilliant new amplifier boards from Avondale

misterc6 said:
We're getting a long way off topic here but the diagram in post #657 is correct. I suggest removing the diagram in post #655 in case it confuses anyone.
Click to expand...

yes sorry - my fault - i figured it was a lesser evil to 'spam' this thread than making new topics ... ok - thanks so much for your help Malcolm - gives me the confidence i need to proceed. (and ... sorry)
 
TheManWhoBuiltTheMoon said:
I’ve noticed I’ve had to turn my volume up more than usual when running these boards. Normal would be -35db now it’s more -25 to -20. I run a mdac straight into the 250.
Click to expand...

i'm not really finding such a difference at all really ... finding my maximum volume comfort zone still around 12 o'clock on the dial ...
 
OK ... here's a question for the qudos initiates ... what about speaker cable ...? has anyone compared NAC A5 to other cables with the qudos boards ... or would the inductor at the output stage kind of obviate the need for it ...? or make it undesirable even? currently using Canare 4S6 ...
 
Last edited:
I’m using Avondale Blacklink and have been for years through all my different amps I’ve purchased. Recently been thinking of moving to Chord epic. As for interconnects, I use a Chord 3pin XLR to phono that goes straight into a Mdac as my pre amp.
 
I use blacklink too. However, I am sure Les has stated in the past that the NCC200 modules have no speaker cable restrictions a la Naim, so no need to use A5.
 
So OK to connect high level active subwoofer signal feed directly to amplifier speaker outputs then. Main speakers connected by 5m of A5. BK sub currenly fed from connections at each speaker, not elegant IMHO.
 
misterc6 said:
Looking good Glenn!

I’d advise a temporary fix of the BD237s and doing the test between TP3 and TP4, there’s no need for the heat sinks until you install the output transistors. The ability to adjust the voltage between the test points is more important than the exact values quoted. Set the voltage to its minimum value before proceeding to fit the output devices.
Click to expand...

for other users - there really doesn't seem to be ANY reason to attach the BD237 to a heat sink ...i've explored this under all running conditions and it seems it runs mostly cool as a cucumber ... far more than the ZTX's which themselves aren't sinked ... so ...
 
i found myself in the curious position of having ordered (2)100nf SMR capacitors (big chunky things!) and cannot for the life of me figure out where i was planning on putting them (ohhhh ... the responses i can imagine!). I was thinking maybe a replacement for C16 though that's a 220nf (!). I could well have got them for my NAC 62 ... so i guess i'll throw this out there as a question to the public ... can you think of any fantastic uses for them ..(?) I've been all through the Qudos, NAC62 mod threads and the McBride and acoustica pages and nothing's really jumping out at me ...
 
Jonathan said:
for other users - there really doesn't seem to be ANY reason to attach the BD237 to a heat sink ...i've explored this under all running conditions and it seems it runs mostly cool as a cucumber ... far more than the ZTX's which themselves aren't sinked ... so ...
Click to expand...

The purpose of attaching the BD237 to the main heat sink has nothing to do with keeping it cool. The benefit of placing it there is that it enables the bias chain to track temperature changes on the output devices.
 
It needs to be connected to the output devices by a thermally conductive path.
Being in proximity, i.e. coupling by convection, is not good enough.
 
S-Man said:
It needs to be connected to the output devices by a thermally conductive path.
Being in proximity, i.e. coupling by convection, is not good enough.
Click to expand...

EXCEPT for the fact that it's 'old' position was in the middle of the board and ZERO connection to any 'thermally conductive path' to the MJ transistors (?) - how does that make sense then?
 
The "old" position was for use with quasi-complementary o/p stages. For the "Naim" method to work, the box ideally should be closed (no vents). Even so, there is no way to properly compensate a Q-C ouptut stage, so it's a bit of a fudge.

With a fully complementary o/p stage, the bias current can be properly stabilised. The better the thermal coupling and the lower the thermal time constant (between the o.p stage and the Vbe multilier) the more accurate the compensation will be**.


** This assumes all o/p devices and the drivers are on the same heatsink. This is not the case with the NCC200.
 
S-Man said:
The "old" position was for use with quasi-complementary o/p stages. For the "Naim" method to work, the box ideally should be closed (no vents). Even so, there is no way to properly compensate a Q-C ouptut stage, so it's a bit of a fudge.

With a fully complementary o/p stage, the bias current can be properly stabilised. The better the thermal coupling and the lower the thermal time constant (between the o.p stage and the Vbe multilier) the more accurate the compensation will be**.


** This assumes all o/p devices and the drivers are on the same heatsink. This is not the case with the NCC200.
Click to expand...


hmmm .. i see ... i hadn't realized. i'd assumed, based on previous discussion, that there was actually no real preference in position between the two ...
 
S-Man said:
The "old" position was for use with quasi-complementary o/p stages. For the "Naim" method to work, the box ideally should be closed (no vents). Even so, there is no way to properly compensate a Q-C ouptut stage, so it's a bit of a fudge.

With a fully complementary o/p stage, the bias current can be properly stabilised. The better the thermal coupling and the lower the thermal time constant (between the o.p stage and the Vbe multilier) the more accurate the compensation will be**.


** This assumes all o/p devices and the drivers are on the same heatsink. This is not the case with the NCC200.
Click to expand...


It could only be properly stabilised if the bias, driver and output transistors were on the same silicon, the big advantage a Sziklai pair has is that the driver transistors set the bias current, not the driver transistors plus output stage as in the emitter follower Qudos type designs. Having the driver stage set the bias currents means that the driver transistors and bias transistor can be mounted very close to each other thermally resulting in much better thermal tracking, if this method is done the emitter resistor of the driver transistors that it shares with the collector of the output transistor can be reduced to as low as 0.05 Ohm, I have a class A amplifier with a standing current of 1.25 amps and it's stable with this value, the circuit is similar enough to the Qudos design that it could probably tried quite easily.
 
Chops54 said:
I built a pair of Voyager spec NCC200s a few years back which my son uses in his music system. The NCC220s I've nearly finished are pretty much identical except for the amplifier boards. Same transformers, same VBEs etc and pretty much the same make of components used to build the amplifiers. He's coming over at Christmas so I'll ask him to bring them with him.

I doubt I'll upgrade to the HCR200 as I'll start building the NCC300s in the new year and the 220 Voyagers will end up with my son I suppose. I still can't work out how he blagged my original Voyagers o_O
Click to expand...

Hi Chops, did you get to compare the ncc200 voyager to the ncc220 voyager?
:)
 
You must log in or register to reply here.


advertisement


Back
Top