I just wanted to mention that my NDK clocks that I got from analogresearch.co were always preferred to the Crysteks, some will agree that clocks do make a difference. I will do a comparison again when I get my boards and will report asap. I wish he was selecting 45/49 ones as well but it is what it is...
Hm, I wonder if he would be willing to measure two clocks I have from the Wall Tempered Master Clock project? They are based on AT cut crystals so should be worse than the ones with SC cut but it would be interesting to see how they fare. Andrea never got them measured...
45/49 clocks will probably perform worse, that's most likely the reason he doesn't sell them.
45/49 clocks will probably perform worse, that's most likely the reason he doesn't sell them.
Interesting.
Can you describe what you mean with “more real” does it portray more colour, dynamics, more transparant?
Regards,
D
Deleted member 537459
D
Deleted member 537459
Well, I tried Ians dual 9038 dac with the TP Mercury and found I needed a lot of gain after it, compaired to the TP Buffalo 9038PRO dac. I'd say the Merc is set up to expect the massive current output of the PRO dac. The junior 9038 versions current output is much lower. So, lower current input equals lower voltage output.
I'm no expert but that's my call.
The Merc could possibly be setup to put out more voltage.
Is there a table somewhere listing the magnitude of the current output of the different ESS90xx DACs?
es9038pro has much higher output current than es9038q2m dac. You need to change the resistor.
E.g.: for es9038pro 60R is recommened to get ~2Vrms. For es9028pro four times more, because (es9038pro is four es9028pro in paralel), so you would need around 240R. Then es9028pro is four es9038q2m, but it is dual mono, thus you multiply only by two, which is around 480R to get ~2Vrms from Ians dac with mercury.
Last edited:
A tip and a Challenge
The tip(s) pertain to sound quality from the NDK SDA clock. (perhaps pertains to others). I experience a bump in sound quality from 2 tweaks.
1. Place a small blob of adhesive putty right on the clock to dampen vibration
2. An even better improvement by placing a cap right on the power pin to ground. I use BG NX HiQ .33u 50v.
I know BlackGate caps are controversial, and certainly not worth the asking price... but if you happen to have one, put it on your 45mHz clk.
Now the challenge. I use a well sorted (IMHO) linear supply, not battery to power the FIFOPi. I wonder if there would be any improvement for someone using battery supply. One would think no, but it would be interesting. Challenge goes out to Greg as I know you are also a BG enthusiast.
FYI the linear supply is a small Hammond 229 transformer, super soft recovery rectifiers, CLCLC filtering with 9mF low ESR caps, a 5v salas shunt as a pre reg followed by Ian's tps4700 based discrete reg also with BG caps. So likely not in the same league as battery, but not a bad supply.
Pic of caps on clock below
The tip(s) pertain to sound quality from the NDK SDA clock. (perhaps pertains to others). I experience a bump in sound quality from 2 tweaks.
1. Place a small blob of adhesive putty right on the clock to dampen vibration
2. An even better improvement by placing a cap right on the power pin to ground. I use BG NX HiQ .33u 50v.
I know BlackGate caps are controversial, and certainly not worth the asking price... but if you happen to have one, put it on your 45mHz clk.
Now the challenge. I use a well sorted (IMHO) linear supply, not battery to power the FIFOPi. I wonder if there would be any improvement for someone using battery supply. One would think no, but it would be interesting. Challenge goes out to Greg as I know you are also a BG enthusiast.
FYI the linear supply is a small Hammond 229 transformer, super soft recovery rectifiers, CLCLC filtering with 9mF low ESR caps, a 5v salas shunt as a pre reg followed by Ian's tps4700 based discrete reg also with BG caps. So likely not in the same league as battery, but not a bad supply.
Pic of caps on clock below
Attachments
Have you populated the bottom areas of the adaptator ?
If not, challenge your BG nx with an acrylic smd cap at the bottom on the 0807 area without populating the two other littliest areas...
Tip 2 : beginn with the acrylic without the BG first, beginn also with just one adaptator not both; check with ears then only after try the two adaptaors.
If not, challenge your BG nx with an acrylic smd cap at the bottom on the 0807 area without populating the two other littliest areas...
Tip 2 : beginn with the acrylic without the BG first, beginn also with just one adaptator not both; check with ears then only after try the two adaptaors.
Last edited:
Impressed more and more...
Ian,
I thought what I was listening to with my my stable of other DAC hats was good, but your DAC and associated support boards has elevated that to the 'great' category. And if it's this good out of the box with the standard supplied parts, I can't wait to do the upgrades - already planned.
I spent a good amount of time last night listening to many of my favorite tracks, and was literally amazed by what I was hearing on my humble system. Almost every track (using roon), perked up my ears with a new sense that this sounds different - better! There was a such a 'detailed presence' to the sound. The highs seemed clearer and notes more distinct. The mids are so well executed that the vocals came through with clarity, yet smooth and effortlessly. And even though I'm using a powered sub with my system, the lows were solid with a bold and commanding feel. I clearly was hearing a much better presentation of the music. The imaging, while in the past seemed focused on the speakers, the imaging really opened up nicely across the front of my room.
This was across the entire genre of my collection, from CD rips from the 80s right up until this year. Including redbook 44/16, all the way up to DSD tracks from this year. I'm simply impressed with what this DAC project, and have all but abandoned some of my other DAC projects.
Note: While moving my temporary hookup on to a chassis for moving into my listening room, a little scary, but minor mishap occurred. This was all my fault. One of the those little green power connectors(with a plug) I had used on the dual 12v rails PSU for the I/V stage got connected incorrectly. Even though it is 'loosely' keyed to fit a particular way, it CAN BE inserted the wrong way without any effort. And without me doing a serious re-check of everything, I turned on the system. No sound. I immediately touched the opamps and they weren't warm, they were HOT. I pulled the plug within about 4 secs. Thinking I had fried the output board, I re-inserted the cable into the PSU correctly. Fortunately, I did not see any magic smoke or smelled anything weird. I gave it a couple minutes to cool off and retried everything. Voila! It worked, and all seemed well and sounded fine. You can't say I wasn't warned, as throughout the docs there are plenty of reminders to observe CORRECT polarity! Lesson learned.
Anyway this DAC project sounds terrific with everything I've thrown at it to play so far. Thanks again for all your effort to bring this to the DIY community with boards and products that well thought out and professionally executed, for a very fair price too.
Now I can start on the upgrades.
Below is a pic of my temp setup.
Rick
Ian,
I thought what I was listening to with my my stable of other DAC hats was good, but your DAC and associated support boards has elevated that to the 'great' category. And if it's this good out of the box with the standard supplied parts, I can't wait to do the upgrades - already planned.
I spent a good amount of time last night listening to many of my favorite tracks, and was literally amazed by what I was hearing on my humble system. Almost every track (using roon), perked up my ears with a new sense that this sounds different - better! There was a such a 'detailed presence' to the sound. The highs seemed clearer and notes more distinct. The mids are so well executed that the vocals came through with clarity, yet smooth and effortlessly. And even though I'm using a powered sub with my system, the lows were solid with a bold and commanding feel. I clearly was hearing a much better presentation of the music. The imaging, while in the past seemed focused on the speakers, the imaging really opened up nicely across the front of my room.
This was across the entire genre of my collection, from CD rips from the 80s right up until this year. Including redbook 44/16, all the way up to DSD tracks from this year. I'm simply impressed with what this DAC project, and have all but abandoned some of my other DAC projects.
Note: While moving my temporary hookup on to a chassis for moving into my listening room, a little scary, but minor mishap occurred. This was all my fault. One of the those little green power connectors(with a plug) I had used on the dual 12v rails PSU for the I/V stage got connected incorrectly. Even though it is 'loosely' keyed to fit a particular way, it CAN BE inserted the wrong way without any effort. And without me doing a serious re-check of everything, I turned on the system. No sound. I immediately touched the opamps and they weren't warm, they were HOT. I pulled the plug within about 4 secs. Thinking I had fried the output board, I re-inserted the cable into the PSU correctly. Fortunately, I did not see any magic smoke or smelled anything weird. I gave it a couple minutes to cool off and retried everything. Voila! It worked, and all seemed well and sounded fine. You can't say I wasn't warned, as throughout the docs there are plenty of reminders to observe CORRECT polarity! Lesson learned.
Anyway this DAC project sounds terrific with everything I've thrown at it to play so far. Thanks again for all your effort to bring this to the DIY community with boards and products that well thought out and professionally executed, for a very fair price too.
Now I can start on the upgrades.
Below is a pic of my temp setup.Rick
Attachments
I just wanted to thank Ian for his fabulous work on the latest group buy. My package arrived wednesday and the highlight for me was the new Fifo pi reclocker and isolator (given that I need to wait for a new batch of the lifepo power supplies before Ian is able to send me one). I bought the fifo pi for a separate project but decided to give it a go on my main DAC. This would mean that the I2S signal coming into my dac would be reclocked twice with an increasing ‘quality’ of clock. I’ve not read of anyone trying this, so I thought I’d give it a go! I have to admit I wasn’t expecting anything positive given that I already have a pair of Pulsar 45/49 clocks in my DAC but the improvement was well worth it and the board will likely stay in the DAC!
Previously my signal path has been this:
Pi 2 powered by mezzanine power board > I2S out to Ian’s pi isolator board > Is2 u.fl’s to Ian’s Mc FIFO (45/49 pulsar clocks) > dual mono 9024
Last night I replaced the isolator board to make:
Pi 2 powered by mezzanine power board > I2S out to Ian’s FIFO pi isolator + reclocker from this year’s GB (stock clocks then 957s) > Is2 u.fl’s to Ian’s Mc FIFO (45/49 pulsar clocks > dual mono 9024
None of the power supplies changed, in fact nothing changed bar the isolator/reclock board itself. I initially used the supplied clocks and the results surprised me. For starters there was change to the depth of presentation which I wasn’t really expecting and secondly a quite positive change to the separation of instruments. So, I then swapped out the standard clocks for a pair of new 45/49 Crystek 957s and was a little disappointed that there wasn’t any real further improvement. I thought I’d leave it all in place and listen to some tracks I know well. As the evening wore on (and presumably the 957s started to warm up / run in)….things just got better and better. (I managed to hold off the wine, too!) The little expansion to the soundstage was still evident but what was really apparent was the extra texture detailing individual sounds and instruments that started to appear after a couple of hours. I hadn’t really expected anything good to come of this experiment - let alone a tweak that turned out to be one of the 2 most significant I've done this year, so I can’t believe it’s expectation bias (i'll listen more tonight)…but I would appreciate someone else doing a similar test please.
In other news I’ve followed Greg’s idea of an ultra capacitor as close to the load as possible and soldered one of those little Panasonic ultracapacitors onto the end of the supply cable to the pulsar clocks (mounted on Ian's McFIFO clock board). Not tried it yeat, so we’ll see tonight whether that improves anything further. If anyone can recommend a similarly small packaged ultracap but with higher working voltage I’d appreciate it. I don’t like running components at their limit but I would like to try the same trick to the supply of the Twisted pear dac boards…5v … and even the IV circuit, but I might be pushing things there as that needs +/- 32V !
Cheers,
Crom
Previously my signal path has been this:
Pi 2 powered by mezzanine power board > I2S out to Ian’s pi isolator board > Is2 u.fl’s to Ian’s Mc FIFO (45/49 pulsar clocks) > dual mono 9024
Last night I replaced the isolator board to make:
Pi 2 powered by mezzanine power board > I2S out to Ian’s FIFO pi isolator + reclocker from this year’s GB (stock clocks then 957s) > Is2 u.fl’s to Ian’s Mc FIFO (45/49 pulsar clocks > dual mono 9024
None of the power supplies changed, in fact nothing changed bar the isolator/reclock board itself. I initially used the supplied clocks and the results surprised me. For starters there was change to the depth of presentation which I wasn’t really expecting and secondly a quite positive change to the separation of instruments. So, I then swapped out the standard clocks for a pair of new 45/49 Crystek 957s and was a little disappointed that there wasn’t any real further improvement. I thought I’d leave it all in place and listen to some tracks I know well. As the evening wore on (and presumably the 957s started to warm up / run in)….things just got better and better. (I managed to hold off the wine, too!) The little expansion to the soundstage was still evident but what was really apparent was the extra texture detailing individual sounds and instruments that started to appear after a couple of hours. I hadn’t really expected anything good to come of this experiment - let alone a tweak that turned out to be one of the 2 most significant I've done this year, so I can’t believe it’s expectation bias (i'll listen more tonight)…but I would appreciate someone else doing a similar test please.
In other news I’ve followed Greg’s idea of an ultra capacitor as close to the load as possible and soldered one of those little Panasonic ultracapacitors onto the end of the supply cable to the pulsar clocks (mounted on Ian's McFIFO clock board). Not tried it yeat, so we’ll see tonight whether that improves anything further. If anyone can recommend a similarly small packaged ultracap but with higher working voltage I’d appreciate it. I don’t like running components at their limit but I would like to try the same trick to the supply of the Twisted pear dac boards…5v … and even the IV circuit, but I might be pushing things there as that needs +/- 32V !
Cheers,
Crom
Hi diyiggy
Great suggestions..
Yes I did populate the 3 small caps in the bottom of the board.. I thought this was a good design by Ian as I understand the caps should be really close to the consumer which I expect he accomplished. If I mount new clocks sometime, I'll try all your excellent suggestions. The one adapter vs the other is also a great idea for rapid A B test. One could have the same song at 2 different clock rates and learn the difference in sound just from the clock rate before installing one cap. So I'll leave the AB testing to the next challenger for now
super caps
Crom,
Interesting feedback re sound quality.
Where exactly are you going to place your super caps? Right at the clk?
I have some 5v super caps. I was thinking of putting one right at the input to the +3.3v clean power on the FIFOPi. So it would be right at the output of the 3.3v regulator/input to the FIFO. I need to do some digging and find Greg's suggestion.
Maxwell has modules up to 48v. Probably very expensive. As I understand it, they series connect individual cells and balance the current flow until they get the target voltage.
Interested in your report.
Crom,
Interesting feedback re sound quality.
Where exactly are you going to place your super caps? Right at the clk?
I have some 5v super caps. I was thinking of putting one right at the input to the +3.3v clean power on the FIFOPi. So it would be right at the output of the 3.3v regulator/input to the FIFO. I need to do some digging and find Greg's suggestion.
Maxwell has modules up to 48v. Probably very expensive. As I understand it, they series connect individual cells and balance the current flow until they get the target voltage.
Interested in your report.
Thanks TioFrancotirador,
So if I understand you correctly the current output (per chip) is like this:
9038Pro = 4x 9028Pro = 16x 9038Q2M
What about the 9028Q2M and the venerable 9018?
desoldering smd caps is not difficult.
Hi, you missunderstood me, sorry I was not precise enough
I just wanted to say as I already told before to Greg Steewart :
You might prefer a standalone acrylic cap instead the three ceramic (which imho are reducing the BG NX goodies) but not with // ceramic - the idea behind that is ceramic gave to me not as good results and by far.
Populating only the bottom of one adaptator profits also to the other unpopulated adaptator. It could be enough (too much of a good thing may waste the sound and that's why I wrote to first test one adaptator : the 48K Hz side first - and yes you will hear a difference on the 44.1 K Hz materials despite its adaptator is unpopulated)
- One can start with something between 0.47 to 1 uF smd acrylic (but not the Panasonic PPS range which are too big for the decoupling area: CDE FCA are ok in 16V). then try // the BG NX, etc...
- So the idea behind was not for rapid A B test
Now ceramics are perhaps just good with the BG NX and the FiFoPi, also there is already an acrylic cap at the bottom of the FiFoPi board (which is a great idea), so my tip may not be relevant. But notice I have experienced with CCHD-957 and than the so called smd adaptator was in fact first for the CCHD-957 in the older iancanada projects (as the values of the caps of this kit, so maybe not the good values for the NDK !). YMMV with the NDK-Z A grade.
Anyway if people want to stay with ceramic caps on the adaptator, I should first try the adaptaor without these ceramic caps first, then only one socket, then both...
Let your ears decide what you most prefer - changes here are not at the margin.
regards
Hi diyiggy
Great suggestions..
Yes I did populate the 3 small caps in the bottom of the board.. I thought this was a good design by Ian as I understand the caps should be really close to the consumer which I expect he accomplished. If I mount new clocks sometime, I'll try all your excellent suggestions. The one adapter vs the other is also a great idea for rapid A B test. One could have the same song at 2 different clock rates and learn the difference in sound just from the clock rate before installing one cap. So I'll leave the AB testing to the next challenger for now
Hi, you missunderstood me, sorry I was not precise enough
I just wanted to say as I already told before to Greg Steewart :
You might prefer a standalone acrylic cap instead the three ceramic (which imho are reducing the BG NX goodies) but not with // ceramic - the idea behind that is ceramic gave to me not as good results and by far.
Populating only the bottom of one adaptator profits also to the other unpopulated adaptator. It could be enough (too much of a good thing may waste the sound and that's why I wrote to first test one adaptator : the 48K Hz side first - and yes you will hear a difference on the 44.1 K Hz materials despite its adaptator is unpopulated)
- One can start with something between 0.47 to 1 uF smd acrylic (but not the Panasonic PPS range which are too big for the decoupling area: CDE FCA are ok in 16V). then try // the BG NX, etc...
- So the idea behind was not for rapid A B test
Now ceramics are perhaps just good with the BG NX and the FiFoPi, also there is already an acrylic cap at the bottom of the FiFoPi board (which is a great idea), so my tip may not be relevant. But notice I have experienced with CCHD-957 and than the so called smd adaptator was in fact first for the CCHD-957 in the older iancanada projects (as the values of the caps of this kit, so maybe not the good values for the NDK !). YMMV with the NDK-Z A grade.
Anyway if people want to stay with ceramic caps on the adaptator, I should first try the adaptaor without these ceramic caps first, then only one socket, then both...
Let your ears decide what you most prefer - changes here are not at the margin.
regards
Last edited: