Thanks Daniel for help, but I had one to my lab.
Already, I have done the change but the behavior is identical.
The only things that changed is the THD, now is almost identical btw the two channels. I am happy for that.
The dac-chip that I removed says:
ES9023P
GJVA1879A
J442
I am going to looking more about this in the future...I'' make some measurement to mambo dac (same 9023 dac-chip).
It was my fault that I haven't talk about sonic characteristics of your dac, Daniel.
It is very nice, detailed with the very deep low frequencies and superb analysis.
I have done a lot of acoustic comparisons between direct output (after the filter of 9023) and buffer output.
To my system the direct output is more balanced, to high frequencies.
For that reason, I have done some measurements via RightMark Audio Analyser Pro between both of outputs.
Look at the attachments, especially to frequency response. As easy seems the buffer output has a more flat response to very high frequencies than the direct output that has a small roll-off.
The results are very interesting and it is my question if this is by design of filter-buffer or setup of filter-buffer.
Already, I have done the change but the behavior is identical.
The only things that changed is the THD, now is almost identical btw the two channels. I am happy for that.
The dac-chip that I removed says:
ES9023P
GJVA1879A
J442
I am going to looking more about this in the future...I'' make some measurement to mambo dac (same 9023 dac-chip).
It was my fault that I haven't talk about sonic characteristics of your dac, Daniel.
It is very nice, detailed with the very deep low frequencies and superb analysis.
I have done a lot of acoustic comparisons between direct output (after the filter of 9023) and buffer output.
To my system the direct output is more balanced, to high frequencies.
For that reason, I have done some measurements via RightMark Audio Analyser Pro between both of outputs.
Look at the attachments, especially to frequency response. As easy seems the buffer output has a more flat response to very high frequencies than the direct output that has a small roll-off.
The results are very interesting and it is my question if this is by design of filter-buffer or setup of filter-buffer.
Attachments
The Filter has a slight peak of 1.9dB arround 80kHz. At 24kHz the amplification is approximately 0.3dB. That corresponds to your measurements.
However 0.3dB is not really that much (at 24kHz!)
You can dampen the peak shunting the inductors with a 47kOhms resistor. Maybe you prefer that variant
kind regards, Daniel
However 0.3dB is not really that much (at 24kHz!)
You can dampen the peak shunting the inductors with a 47kOhms resistor. Maybe you prefer that variant kind regards, Daniel
The Filter has a slight peak of 1.9dB arround 80kHz. At 24kHz the amplification is approximately 0.3dB. That corresponds to your measurements.
However 0.3dB is not really that much (at 24kHz!)
kind regards, Daniel
From what I understood with my poor english, if I changed the L1,L2 with 47KOhms resistors, this "lift" will be changed?
Hi,
I don't know if I am in the right place for this question about the JG buffer :
There are two passive parts in serie with the signal before the fets. A 18 mH Fastron bulk inductor and a 470 r smd resistor.
What is the job of the inductor ? DC blocker or/and glitch removal ? As it is not a state of the art inductor (Sovtek, Lundhal, ...eg does it woerth to go with expensive one or stay with the Fastron (another cheap brand possible?) ?
Idem for the resistor , shall a Rhopoint bulk wirewound better than the smd one. Can I populate with 500 ohms instead 470 if Rhopoint.
Sorry for these 2 cents questions. I'm sure it's good enough wxith the original BOM but can spend ten dollars more if better (and of course don't spend if there is no difference
I don't know if I am in the right place for this question about the JG buffer :
There are two passive parts in serie with the signal before the fets. A 18 mH Fastron bulk inductor and a 470 r smd resistor.
What is the job of the inductor ? DC blocker or/and glitch removal ? As it is not a state of the art inductor (Sovtek, Lundhal, ...eg
does it woerth to go with expensive one or stay with the Fastron (another cheap brand possible?) ?Idem for the resistor , shall a Rhopoint bulk wirewound better than the smd one. Can I populate with 500 ohms instead 470 if Rhopoint.
Sorry for these 2 cents questions. I'm sure it's good enough wxith the original BOM but can spend ten dollars more if better (and of course don't spend if there is no difference
The coil acts as an interator so it supresses higher frequencies that can be aliasing products.
It is in series with the signal so does not dump HF junk to ground.
The first prototype was build with an expensive Sowter coil so i could compare.
To my surprise the Fastron coil does the job very well.
At least at the levels present the coil does not add audible distortion so a better, more expensive coil is not nesassarry.
It is in series with the signal so does not dump HF junk to ground.
The first prototype was build with an expensive Sowter coil so i could compare.
To my surprise the Fastron coil does the job very well.
At least at the levels present the coil does not add audible distortion so a better, more expensive coil is not nesassarry.
thanks Joachim,
Cool. I will go with the Fastron.
As it can be found only at Mouser, do you think for example an equivalent value with a Bourn inductor from costa Rica can match the Fastron ? (less than 90 mA) ; because can be found at Digikey and minimal Fee with small BOM !
What about the resistor in serie after the inducor ? Same conclusion, if not tested with exotic expensive resistor like Rhopoint bulk wirewoung 510g (Farnell), any idea if it could be better (I know that Thorsten love it both for digital signal and to the ground for passive i/v too) ?
Cool. I will go with the Fastron.
As it can be found only at Mouser, do you think for example an equivalent value with a Bourn inductor from costa Rica can match the Fastron ? (less than 90 mA) ; because can be found at Digikey and minimal Fee with small BOM !
What about the resistor in serie after the inducor ? Same conclusion, if not tested with exotic expensive resistor like Rhopoint bulk wirewoung 510g (Farnell), any idea if it could be better (I know that Thorsten love it both for digital signal and to the ground for passive i/v too) ?
Last edited:
Well, the Bourns is quite simailar but has a lower self resonance ( the point where the inductance drops and the coil gets capacitive ) as by the spec sheet.
Never measured or verified by a second source though, not by me ether.
The group buy more or less eroded the world stock of that particular Farstron 18mH coil C: !!!!!!! and my buffer-filter is used in other sucessfull DIY projects like the Curryman DAC so in total several hundred are in use.
You can use another value though that comes close. This is no holy cow.
Never measured or verified by a second source though, not by me ether.
The group buy more or less eroded the world stock of that particular Farstron 18mH coil C: !!!!!!! and my buffer-filter is used in other sucessfull DIY projects like the Curryman DAC so in total several hundred are in use.
You can use another value though that comes close. This is no holy cow.
Coilcraft RFB0807 Leaded Power Inductors
dunno if they get across the pond. Ignore the 330 in the picture which is just for reference.
dunno if they get across the pond. Ignore the 330 in the picture which is just for reference.
An externally hosted image should be here but it was not working when we last tested it.
Which Power supply for JG Buffer ?
Hi,
Don't see any build thread for JG Buffer, maybe it's so simple to need it. But as I don't want to waste the design, I have please some questions about it:
I will wait for the Fastron coil to respect the design even it's not written on marble (end of April according Mouser...Damned theses deutschs colmpanies have to much work ) . Thanks for the others ref given above... didn't know !
As I have time to finish my BOM, I would ask please which minimal powershematic and technic I have to apply for it.
The Buffer is the matched version of EUVL/Mark (so my understanding is I don't need The RC (R2+C3) : 470 R + 150 pF ?)
Do I need :
1- the +/- 5V shunt PS shown of the SEN-CEN-ZEN thread ?
2- linear PS with a coil Pi filter and Micrel LDO for stable voltage ?
3- double batteries PS with phantom rail and big caps for impedance with Micrel Ldo. Is there a lake of dynamic in Low frequency in relation to a linear or shunt PS ?
4) Something else ?
Last point about the big caps on the Buffer : is there better than the 220 uF Panasonic FC (I think maybe about KZ muse ?!). Saw than Gerrad used 470 uf Panasonic as well ? Detail or important ?
I try to do the cheaper but minimal quality not to waste the combo ES9023 DAC + JG (matched EUVL) Buffer. And got already a regulated 5v for the DAC. Can make myself the assembly if shematic on a bare board.
Thank you,
Eldam
Hi,
Don't see any build thread for JG Buffer, maybe it's so simple to need it. But as I don't want to waste the design, I have please some questions about it:
I will wait for the Fastron coil to respect the design even it's not written on marble (end of April according Mouser...Damned theses deutschs colmpanies have to much work
) . Thanks for the others ref given above... didn't know !As I have time to finish my BOM, I would ask please which minimal powershematic and technic I have to apply for it.
The Buffer is the matched version of EUVL/Mark (so my understanding is I don't need The RC (R2+C3) : 470 R + 150 pF ?)
Do I need :
1- the +/- 5V shunt PS shown of the SEN-CEN-ZEN thread ?
2- linear PS with a coil Pi filter and Micrel LDO for stable voltage ?
3- double batteries PS with phantom rail and big caps for impedance with Micrel Ldo. Is there a lake of dynamic in Low frequency in relation to a linear or shunt PS ?
4) Something else ?
Last point about the big caps on the Buffer : is there better than the 220 uF Panasonic FC (I think maybe about KZ muse ?!). Saw than Gerrad used 470 uf Panasonic as well ? Detail or important ?
I try to do the cheaper but minimal quality not to waste the combo ES9023 DAC + JG (matched EUVL) Buffer. And got already a regulated 5v for the DAC. Can make myself the assembly if shematic on a bare board.
Thank you,
Eldam
Hi Eldam,
there is a dedicated thread to the JG Buffer here -->
http://www.diyaudio.com/forums/group-buys/216669-joachim-gerhard-filter-buffer-es9022.html
In this thread you will find the schematic and other information... like the JG Filter Buffer has nothing to do with Sen-Cen-Zen
and needs a bipolar supply of 15V. Power consumption is low at about 25mA per rail when I remember correctly.
I have tried the JG Buffer with a bipolar battery supply of 9V for testing purposes, my final build will use a supply with "super
regulators" à la W. Jung.
EUVL has advocated to use a filtered battery supply...
yes know it and already find the BOM and shematics about the buffer but not the way to feed it with a power supply.
I didn't want to post in the link you give because it's yet an active incremental group buy thread and don't want to bother Mark (or Patrick about the PS which seems not to be supported in this thread) without to have the permission to do it.
Thank you Stixx for your input and voltage rails value (I was thinking about 5 V ).
batteries PS can be hard to do for gound sounding with the little experience I have and you need strong value of caps (because of its bad impedance), off course regulators as the volatge move as they get empty. Lead batteries have better result with sound for me but I saw than Gerrard used simple 9V batteries ??? Maybe just a test before a more acurate PS design with batteries ?
Anyway, I suppose it's I will found here the Super Voltage Regulator of Mr W. Jung without any difficulty.
I didn't want to post in the link you give because it's yet an active incremental group buy thread and don't want to bother Mark (or Patrick about the PS which seems not to be supported in this thread) without to have the permission to do it.
Thank you Stixx for your input and voltage rails value (I was thinking about 5 V
).batteries PS can be hard to do for gound sounding with the little experience I have and you need strong value of caps (because of its bad impedance), off course regulators as the volatge move as they get empty. Lead batteries have better result with sound for me but I saw than Gerrard used simple 9V batteries ??? Maybe just a test before a more acurate PS design with batteries ?
Anyway, I suppose it's I will found here the Super Voltage Regulator of Mr W. Jung without any difficulty.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.