Don't forget to give the valve DAC main board a termination of about 750 ohm. Without it, the signal level gets quite large (though not large enough to blow up the buffer - unlike what I wrote originally) and the DAC's noise modulation gets much worse.
Assuming 15 nF for C28 and C50 on the main board, and assuming that the negative output pins of the main board are grounded and the positive pins are connected to 750 ohm to ground in parallel with one of these filter boards, the whole response will be roughly a cascade of a first-order filter at 16042 Hz and a second-order Butterworth filter at 118627 Hz - not very good, but good enough to test if it works. With C28 and C50 reduced to 2.2 nF, it gets close to third-order Bessel at 80 kHz or so, which may work well for the higher DSD rates.
Assuming 15 nF for C28 and C50 on the main board, and assuming that the negative output pins of the main board are grounded and the positive pins are connected to 750 ohm to ground in parallel with one of these filter boards, the whole response will be roughly a cascade of a first-order filter at 16042 Hz and a second-order Butterworth filter at 118627 Hz - not very good, but good enough to test if it works. With C28 and C50 reduced to 2.2 nF, it gets close to third-order Bessel at 80 kHz or so, which may work well for the higher DSD rates.
Last edited:
Thanks Marcel. I'm certain I have one of the buffer boards leftover from a previous project but I can't put my hands on it at the moment - may not be quite so cheap if I can't find it!
I will try and configure it with an approximation of the filter you have suggested.
On other matters, today the smd parts that were out of stock arrrived so it shouldn't be too long before the digital sections of the Valve DAC board get soldered up.
Ray
I will try and configure it with an approximation of the filter you have suggested.
On other matters, today the smd parts that were out of stock arrrived so it shouldn't be too long before the digital sections of the Valve DAC board get soldered up.
Ray
Thanks, however, I'm cheating as I have an experienced friend who is taking care of it for me.
I got a message earlier to say my chum has done all of the smd soldering apart from two parts I forgot to order 
I should get the PCB back on Monday when we get back to work.
I've made some good progress soldering up various power supplies today so it should start coming together fairly soon.
I should get the PCB back on Monday when we get back to work.
I've made some good progress soldering up various power supplies today so it should start coming together fairly soon.
Yes, looking forward to getting the board back to have a look as he sounds pleased with the work he's done for me - I'll post a picture when I get it.
I'm actually in the process of ordering up some more parts at the moment.
Thanks for the update Marcel. I'm thinking I may glue some LEDs into the holes so I can have some disco lighting
If I were to do that I would play safe and run the LED wires inside some teflon tubing through the PCB holes.
Hello again Marcel.
I'm slowly working through the BOM and schematics to get some of the through hole parts ordered. Looking at the resistors, there are quite a few that have a value but no power rating specified; unless it specifically states otherwise is it a reasonable assumption that 0.25W resistors will be OK. I'm assuming a tolerance of 1% for the resistors.
With the SMD parts soldered, for my next step I want to get the power supplies assembled so that I can check them out and look to test the clock/data lines and Vref before I spend more time and money. How does that sound?
Ray
I'm slowly working through the BOM and schematics to get some of the through hole parts ordered. Looking at the resistors, there are quite a few that have a value but no power rating specified; unless it specifically states otherwise is it a reasonable assumption that 0.25W resistors will be OK. I'm assuming a tolerance of 1% for the resistors.
With the SMD parts soldered, for my next step I want to get the power supplies assembled so that I can check them out and look to test the clock/data lines and Vref before I spend more time and money. How does that sound?
Ray
For the through-hole resistors, I used 1/4 W carbon film resistors unless otherwise specified. If they fit in the footprints (both the vertical and the horizontal version), then 1/2 W is also good.
It sounds like a very good idea to first check whether the data and clock input section does what it is supposed to do according to post #262 before proceeding, because this part is new and might need tweaking of component values.
The Vref part is very similar to what I built, but you can check it separately, so why not? If for whatever reason the voltage between Vref and the -300 V rail should be much too high, it could damage the E88CCs, so it is nice if you can prevent that.
It sounds like a very good idea to first check whether the data and clock input section does what it is supposed to do according to post #262 before proceeding, because this part is new and might need tweaking of component values.
The Vref part is very similar to what I built, but you can check it separately, so why not? If for whatever reason the voltage between Vref and the -300 V rail should be much too high, it could damage the E88CCs, so it is nice if you can prevent that.
I'm going to have to get some help with the measurements in post #262 as I don't have the test equipment, nor the experience to be confident of doing it correctly.
Closely related, another question please Marcel - what are the functions of P8, P9 & P11?
P9 & P11 appear to be for alternative connections, like a changeover switch function and I assume P9 should be the opposite of P11?
Thanks
Closely related, another question please Marcel - what are the functions of P8, P9 & P11?
P9 & P11 appear to be for alternative connections, like a changeover switch function and I assume P9 should be the opposite of P11?
Thanks
