Ping... Pong as ITs' say
Saw the other thread... Yes thank you with pleasure ! ?
As we are building the same project, what do you think about continuing the discussion in the Amelioration thread of GaryB about the Subbu ES9023 Xenified with the Buffer (The OP talk already about it....) ?
Any way I PM you as you ask... Ping !
Saw the other thread... Yes thank you with pleasure ! ?
As we are building the same project, what do you think about continuing the discussion in the Amelioration thread of GaryB about the Subbu ES9023 Xenified with the Buffer (The OP talk already about it....) ?
Any way I PM you as you ask... Ping !
I've built one of the JG filter/buffers and think that it sounds quite good. But this thread had me take a look at the choice of inductor and I now have questions about the Fastron inductor. Mouser's website lists the self resonance for the Fastron part as 525khz but I think this is a mistake. If you look at the datasheet from Fastron http://www.fastrongroup.com/image-show/37/07M.pdf?type=Complete-DataSheet&productType=series you see that Fastron specifies the SRF as only 79khz. This is consistent with earlier comments that the filter shows a peaking at around 80khz.
If this is true, then the shielded Bourns 18mH part with a SRF of 190khz is actually a better part. Or if you want a much higher SRF then Bourns 8250 shielded RF chokes have an SRF of 720khz. The resistance of these inductors is a bit high but shouldn't really affect the filter.
As Joachim pointed out, there are lots of choices and people don't have to wait for the Fastron part to be restocked.
---Gary
Gary,
As per the datasheet 79kHz is the measurement frequency for the Q of the Fastron inductor, not the self resonant frequency (SRF is due to parallel [interwinding?] capacitance). AFAIK the 525kHz SRF was specified too on the Fastron page, however I can not find it at the moment.
I did some simulations and actually the 190kHz SRF Bourns doesn't look too bad especially compared to the standart ES9023 1st order lowpassfilter as per the datasheet. The Bourns 8250 looks like a very good alternative! Its axial type but that should not be a big problem.
kind regards, Daniel
As per the datasheet 79kHz is the measurement frequency for the Q of the Fastron inductor, not the self resonant frequency (SRF is due to parallel [interwinding?] capacitance). AFAIK the 525kHz SRF was specified too on the Fastron page, however I can not find it at the moment.
I did some simulations and actually the 190kHz SRF Bourns doesn't look too bad especially compared to the standart ES9023 1st order lowpassfilter as per the datasheet. The Bourns 8250 looks like a very good alternative! Its axial type but that should not be a big problem.
kind regards, Daniel
Sorry about misreading the label on the Fastron datasheet. It is a bit odd that the datasheet no longer lists the SRF.
In any case, I agree that the Bourns 8250 looks like a very good replacement for all those people who are waiting for the Fastron to show up. Or you could just use a lower value inductor - say the Fastron 15mH and then adjust the filter resistor to give the same cutoff frequency.
---Gary
Yes, the curryman DAC is available again from miniDSP. In Europe you can also buy them from audiophonics.fr. Furthermore DIYaudio member Rondadon will soon announce a GB for european customers where they can probably safe some money 
All those versions are ready build, PCBs are not available. From miniDSP you can also download some instructions
cheers, Daniel
All those versions are ready build, PCBs are not available. From miniDSP you can also download some instructions
cheers, Daniel
Basically you have three options:
1st: run each DAC from the DSPs MCLK. In this case you have to check the drive capability of the MCLK line of the DSP board since you will parallel several DACs (maybe a clock buffer is a good idea if not already implemented). You have to disable the local XO on the DACs (remove a ferrite bead in the corresponding local LDO reg) and add a 0603 jumper to enable external clocking.
2nd: run each DAC asynchronous from the DSP off of it's local MCLK (ESS DACs feature an asynchronous reclocking).
3rd: run the DACs synchronous from one clock but asynchronous from the DSP. This seems only feasible if using two stereo DACs and would require a bit more rework if using more than two DACs.
With the miniSHARC I am using option 2 with good results.
@all: just want to mention that the GB for curryman DACs has just been started
cheers, Daniel
Just finished one 
A very straightforward build, without any problem. A big thank to Daniel for designing that board and making it available. And another thank to Mullet who suggested me this DAC.
It sounds really good. The built was for a friend (he pays the parts, I play with them), so it'll go away soon but I could certainly live happily ever after with it.
The supporting parts are rather straightforward, the total cost was somewhere around 220€ iirc, shipping included:
- Power supply is based on C-L-C filters (470uF-10mH/14r dcr-1000uF) followed by lm317-337.
- USB input is provided by the CM6631A board from diynhk, which has isolated I2S outputs. Getting the driver to install was however a bit of a pain, no automatic setup would work (on vista business).
- Case is a galaxy gx247, rca are from Neutrik. I tried to make the build a bit cleaner than my usual wire nests.
Btw, DC offset is 2 and 3mV, which is quite ok. It's a pcb bought on audiophonics last week, from the new minidsp batch.
I attach a bad phone pic with bad lightning to illustrate the description. I'll try to make better pics at daylight.
A very straightforward build, without any problem. A big thank to Daniel for designing that board and making it available. And another thank to Mullet who suggested me this DAC. It sounds really good. The built was for a friend (he pays the parts, I play with them), so it'll go away soon but I could certainly live happily ever after with it.
The supporting parts are rather straightforward, the total cost was somewhere around 220€ iirc, shipping included:
- Power supply is based on C-L-C filters (470uF-10mH/14r dcr-1000uF) followed by lm317-337.
- USB input is provided by the CM6631A board from diynhk, which has isolated I2S outputs. Getting the driver to install was however a bit of a pain, no automatic setup would work (on vista business).
- Case is a galaxy gx247, rca are from Neutrik. I tried to make the build a bit cleaner than my usual wire nests.
Btw, DC offset is 2 and 3mV, which is quite ok. It's a pcb bought on audiophonics last week, from the new minidsp batch.
I attach a bad phone pic with bad lightning to illustrate the description. I'll try to make better pics at daylight.
Attachments
Attachments
I am using the Amanero USB to I2S together with an isolator and it works really good.
However can you use USB interfaces with Squeezebox touch? Isn't it SPDIF out? In the latter case I'd use the Twisted Pear SPDIF Transeiver
cheers, Daniel
However can you use USB interfaces with Squeezebox touch? Isn't it SPDIF out? In the latter case I'd use the Twisted Pear SPDIF Transeiver
cheers, Daniel
Hi Ernst,
Yes, synchronous is the way to go in this case.
Therefore you can disable the on board clock oscillator of the curryman DAC and feed the clock from Ians reclocker MCLK out as suggested in post #271.
The following steps are necessary:
1. Remove L31 in order to unpower the onboard oscillator
2. Add a jumper (solder bridge or 0 ohms resistor 0603) at J2 position. This will connect the MCLK header to the DACs clock input
Ideally you can add the U.Fl headers to the DAC in order to use the coax connections.
Kind of ES9023 on steroids ;-) Looking forward to hearing your report J
Kind regards, Daniel
Yes, synchronous is the way to go in this case.
Therefore you can disable the on board clock oscillator of the curryman DAC and feed the clock from Ians reclocker MCLK out as suggested in post #271.
The following steps are necessary:
1. Remove L31 in order to unpower the onboard oscillator
2. Add a jumper (solder bridge or 0 ohms resistor 0603) at J2 position. This will connect the MCLK header to the DACs clock input
Ideally you can add the U.Fl headers to the DAC in order to use the coax connections.
Kind of ES9023 on steroids ;-) Looking forward to hearing your report J
Kind regards, Daniel
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.