£230+ for the HQPlayer licence 😱
Looks a good solution, does the player work with system audio too? (ie browser streamed audio, simply to have everything neatly through one thing)
I didn't pay that as I've been using HQ Player for a while and had upgrade rates available, but your's is not an uncommon response. It always amuses me that people (not necessarily you) will pay a fortune for a resistors, capacitors and the like, but baulk at paying for software.
BTW, I run AudioLinux as my operating system, it works very well (it also has a small cost, because of all the add-on developments).
I think it's possible though I've never tried. I think probably the best way to complement HQ Player is with Roon (chargeable software again!), which'll make connecting streaming services very simple.
Another step forward...
This morning I finished off my new amplifier and have been listening to it with my Valve DAC as the source;
The amp is a SE-OTL design using 13E1 output tubes - just one per channel for around 1.5W per channel, which is plenty for my 15ohm Lowther horns. The 13E1s are pentodes, run in triode mode.
I'm feeling pretty chuffed! 🙂
This morning I finished off my new amplifier and have been listening to it with my Valve DAC as the source;
The amp is a SE-OTL design using 13E1 output tubes - just one per channel for around 1.5W per channel, which is plenty for my 15ohm Lowther horns. The 13E1s are pentodes, run in triode mode.
I'm feeling pretty chuffed! 🙂
Cheers Felipe.
Some more pictures here;
https://www.diyaudio.com/forums/tubes-valves/360764-se-otl-amplifier.html#post6355520
I'm actually working on another SE-OTL amp too, this time using 6SN7s to drive 6C33Cs.
Some more pictures here;
https://www.diyaudio.com/forums/tubes-valves/360764-se-otl-amplifier.html#post6355520
I'm actually working on another SE-OTL amp too, this time using 6SN7s to drive 6C33Cs.
Morning Ray,
Very nice amp, congrats.
I have finished the bottom side of dsd valve dac pcb during Sunday.
A lot of work. I will build topside during the week, I hope.
On the other hand soldering PPY's recklock pcb seems impossible to me.
Does anybody has an idea?, recklock is a very handy solution for this dac.
Periklis
Very nice amp, congrats.
I have finished the bottom side of dsd valve dac pcb during Sunday.
A lot of work. I will build topside during the week, I hope.
On the other hand soldering PPY's recklock pcb seems impossible to me.
Does anybody has an idea?, recklock is a very handy solution for this dac.
Periklis
Very quick, and probably dumb question being a novice..
Just thinking about the reconstruction filter - would it be possible to use the Miller effect with a tube+resistor rather a capacitor+resistor in a low pass filter config?
Just thinking about the reconstruction filter - would it be possible to use the Miller effect with a tube+resistor rather a capacitor+resistor in a low pass filter config?
In principle yes, but I don't really see the advantage of that. The input capacitance of a valve with Miller effect is far less linear than a good normal capacitor and besides you get an effective series resistance equal to the reciprocal of the transconductance of the valve, which makes the filtering less effective.
You do use a hot air soldering station and solder paste? After some first tries it gets easier.
Though I own a hot gun soldering station I prefer to use a Dremel Versatip, Butane Soldering station, and soldering paste, for SMD. The problem is with the 0603 format which I can not handle not even see, easily. Anyway, thanks, since it is a rather small board I'll give it a try
A stencil helps a lot too.
Maybe try one of these as well.
8LED 1000X 10MP USB Zoom Digital Microscope Endoscope Magnifier Camera + Stand | eBay
I was just working through the idea of the raw-DSD design with the thought of making into a lower impedance (ie 50ohm - those AKGs I have measure 50-120ohm but 84ohm average) capable OTL style headphone dac/amp by adding/doubling the EC88s to provide more current. Headphones with a max of 1.8V and 36mA max. Given the EC88 output, one option is to double up for additional current output with some tweaks to the output filter.
Naturally would need addition power supply and regulation.
Thoughts?
Naturally would need addition power supply and regulation.
Thoughts?
Do you mean ECC88? You could drive two channels with the same data and connect the outputs, but then you are still not close to 36 mA RMS or even 36 mA peak, at least not without a step-down transformer.
Ok, I was just reading the Phillips ECC88 and it was suggesting a 20mA maximum, so I would have expected it capable of a 50-75% of that. Hmm back to the thinking board.
GB Update
Just a quick update to say that my friend has now moved house and will be in a position to start the smd soldering next week. Apologies for the delay.
Is anyone who has already received bare PCBs close to getting a build finished - it will be nice too hear what others think of this DAC.
Ray
Just a quick update to say that my friend has now moved house and will be in a position to start the smd soldering next week. Apologies for the delay.
Is anyone who has already received bare PCBs close to getting a build finished - it will be nice too hear what others think of this DAC.
Ray
Late3st updated GB spreadsheet attached.
All bare PCBs that I despatched have now been delivered.
I'm waiting on two final payments for the bare PCBs that have been despatched, highlighted in blue.
As soon as they become available I'll start sending on the populated PCBs in the order shown on the spreadsheet.
Ray
All bare PCBs that I despatched have now been delivered.
I'm waiting on two final payments for the bare PCBs that have been despatched, highlighted in blue.
As soon as they become available I'll start sending on the populated PCBs in the order shown on the spreadsheet.
Ray
Attachments
Right now the tail current is about 10 mA, so the current through each of the lower triodes switches between 0 and 10 mA. You could try doubling that by halving R8...R13 and R79...R81, R83, R84. I didn't because the anode to cathode voltage is only about 80 V and I didn't want to get too close to grid conduction, but there is probably some room for increasing the current.
Suppose you have two DAC cores running at 20 mA tail current each and a DSD signal that varies between 25 % ones and 75 % ones (as prescribed in the Scarlet Book). The average current (averaged over a large number of clock cycles) through the output branches then varies between 5 mA and 15 mA, so +/- 5 mA peak variation around a 10 mA DC level. With two of those DAC cores, the total peak signal current becomes 10 mA, so 7.071 mA RMS when you play a sine wave. You can increase it further if you use a sigma-delta modulator that can be driven further than 25 %...75 %, for example a second-order sigma-delta or one with an embedded pulse width modulator. Still, it won't reach 36 mA.
By the way, if you should make a circuit like that, it is better to drive one of the two DACs with data that has been delayed over one bit time, rather than with the exact same data as the other DAC. With delayed data, you effectively increase the FIRDAC length to four taps and further reduce the effect of jitter.
Thank you for that 🙂
Just reading an interesting piece on D-Class amp design, one of the issues they have is (as any digital switching) shoot through on the short rising time. Just wonder increasing the voltage too high would see any problems.
In a push-pull configuration, that shoot through hits the opposing diode within the D class amp. As the rise time gets shorter so the shoot through increases so I suspect there would be a limit before shoot through starts placing enough energy on the plates to cause problems.
(Link here)
Hi Ray, sorry for my late payment, been busy for the last few days. The remaining was paid for. Thanks again for your help!