QUAD 405 Next Gen

Jump to Latest
A

altus fidelitas

Member
Joined 2012
MarcelvdG said:
Anyway, if there are audible differences, it doesn't necessarily have to be due to something non-linear
Click to expand...
I did consider the frequency roll offs. I have R12 = 2.7k / C6 = 1nF which should give a low pass of 48khz and C1 = 0.68uF / R3 = 22k for a high pass of 10Hz.

But realistically, my speakers roll of at 50Hz and my ears can barely discern anything above 10khz and definitely nothing over 15khz. So I suspect the frequency response limits may not explain the 405's signature in the ABX tests I've done.

I'm still curious, so planning to do a frequency and distortion sweep once I have the time
 
Hennady Kovalsky

Hennady Kovalsky

Member
Joined 2009
ejp said:
Only if you ignore the DC servo, R4, R5, and C2. There's another LF pole there.
Click to expand...
Exactly
such a simple servo has a very high quality factor, because the effect of the reactive load is much greater than the active value of the load, with the peak at 2.4 kHz, although the time constant seems to be very large. What this means in practice is the increased non-linearity under the NFB and the dependence of the sound on the parameters of the capacitor in the servo, and in the most sensitive range of human hearing.
In later versions of Quad amplifiers, the engineers abandoned such a servo by using an active circuit with an even longer time constant. Although it is possible to reduce the quality factor of this servo (405th) by damping the storage capacitor in the integrator circuit - this will slightly reduce the depth of the NFB, but increase the linearity of the gain over the entire frequency band.
 
Last edited:
M

MarcelvdG

Member
Joined 2003
MarcelvdG said:
2. Using an op-amp with FET input, you can easily redesign the DC loop such that you don't need an electrolytic capacitor in the DC feedback anymore, see https://www.diyaudio.com/community/threads/redesigning-quad-405-dc-bias-loops.339643/ Mooly's simulation results in that thread are with an incorrect resistor value.
Click to expand...
Mind you, my calculations are for the inverting circuit, like the original QUAD 405 and 405-2. I had overlooked that the circuit in post #1 has been modified into a non-inverting circuit.
 
A

altus fidelitas

Member
Joined 2012
MarcelvdG said:
Mind you, my calculations are for the inverting circuit, like the original QUAD 405 and 405-2. I had overlooked that the circuit in post #1 has been modified into a non-inverting circuit.
Click to expand...

Not had the chance to look into the details on how it works, but will definitely check it out the next time someone brings me a QUAS 405 or a clone to fix. Thanks
 
K

karvalha

Member
Joined 2015
ljm_ljm said:
I can give suggestions for modifying quad405. Is to reduce the gain. This can significantly reduce thd.
In addition, reduce the capacitance of LPF. It can reduce the high-frequency attenuation. The details are better.
In addition, do not replace the operational amplifier.
Click to expand...

ljm_ljm said:
I would apreciate if you can give me those sugestions and instructions as i want to assemple your 405-2 boards.
Click to expand...
I have a quad 44 original, between 405-2, L12-2 and mx50x2 wich one you think will be best for me sound wise
 
You must log in or register to reply here.
Top Bottom