Yes it's hard to find. I saw it just at Mouser, Digikey and Farnell.
Isnt there a local Farnell dirtributor at your country? I found few weeks ago fortunately that, and they can get parts without shipping.
So I did test yesterday, switching between SC Cut Driscoll and AT Cut Clapp.
The difference very hard to hearing, its just like feeling.
The Clapp is sounded a little detailed but a little metallic. Driscoll warmer, rounder, I feel a little wilder soundstage, a little near to my head. But the clapp a little airy, I hear better the echoes finish.
I say the differences are very tiny , both oscillator are very good, the factory NZ2520 is washed, misty, and the highs more hissy than the 2 boards, it is easy to hearing the differences.
The difference very hard to hearing, its just like feeling.
The Clapp is sounded a little detailed but a little metallic. Driscoll warmer, rounder, I feel a little wilder soundstage, a little near to my head. But the clapp a little airy, I hear better the echoes finish.
I say the differences are very tiny , both oscillator are very good, the factory NZ2520 is washed, misty, and the highs more hissy than the 2 boards, it is easy to hearing the differences.
I had similar impressions (post #1438) when I did compare the AT-Cut Clapp vs the SC-Cut Driscoll, although the difference was more pronounced.
My Driscoll uses 11.2896 SC-Cut crystal (Q higher than 1M), I think that this is the reason because the difference was remarkable.
So thats good news, not so bad my ears
) Do you think the 11MHz Driscoll could be much more better than 45Mhz? Is it worth to update with a 11MHz in the future the 45MHz?
So thats good news, not so bad my ears
Do you think the 11MHz Driscoll could be much more better than 45Mhz? Is it worth to update with a 11MHz in the future the 45MHz?
I believe the best way is using a frequency as low as possible, since usually the lower the frequency the higher the Q, and theoretically the lower the phase noise.
As I wrote in the GB thread I'm working on a discrete DAC and since the frequency I need is 44.1 to 192 kHz I don't see any reason to start from very high oscillator frequency and then divide down by a large factor. I think is better to start from the lower oscillator frequency as possible, from 5/6 MHz to 11/12 MHz, not higher.
Sc-Cut crystals at 11/12 Mhz have a Q higher than 1M, at 5/6 Mhz the Q could be around 2M.
With the Driscoll oscillator we get very high loaded Q, that tipically means low phase noise, and the higher the Q of the crystal the higher the loaded Q of the oscillator.
Thanks! So I sometimes play music of 352-384KHz, I needed a higher frequency oscillator, but I listen to lower SR music in 70-80%, I think its worth to buy lower frequency SC cut. If there will be group buy possible join for it.
The 5/6Mhz would be too low I think, the FIFO reclock I remember can just minimum 128x divider, isnt it? (88/96 HD music cant play)
I think you are right, 128x is the minimum, please ask Ian to confirm.
In my discrete DAC I don't need Ian's stuff because I have an internal FIFO and data is reclocked directly on the DAC.
Any plans for GB with your discrete DAC?
It will take long time to go, since I have very little free time for diy audio.
But sure, as soon as my discrete is ready I will share the project and I will offer the PCB and the firmware for the FPGA.
The 5/6MHz can works with Fifo (Ian wrote) on 44,1/48K (128X).
Do you think the 5/6MHz crystals can reach the 2M of Q? The datasheet on the page of Laptech write 1,3M for the HC-43 SC cut maximum, I think its the lowest frequency.
If it is real the 11MHz not very lower Q than 5MHz (on the measure sheet pdf what you attached before about 1,1M).
A 5MHz Sc-cut 3rd overtone can easily reach a Q of 2M, the problem is that it cannot fit in HC/43U package.
The 5/6 MHz SC-cut crystals offered by Laptech are fundamental and tipically the real Q is Greater than specified, so I expect at least 1.5M.
I am thinking now just the HC-43 package
But yes I see the HC-47 can reach 2M. You start now GB with just HC-43 crystals, dont you?
Hi Andrea,
I replaced the MMBT with BF182 and it will only oscillate at 26.667Mhz with C3 installed. I tried 5pf, 10pf, 12pf or 20pf on C3, but none of them will help. Without C3, the frequency will be mostly 24.576Mhz. However, if the probe touches any point "after" 74LVC04 (like right hand side of R9 or R10) while powering up the circuit, the freq will be 26.667Mhz. If the probe stays at left hand side of R9 (before 74LVC04) while powering up, it will be 24.576Mhz! This is so weird that I have no clue about it. I wonder how many people have similar problem like this. It seems so close. Do I miss any thing?
Poting
Last edited:
I've built at least 5 + 5 oscillators with SC-Cut crystals at 22/24 MHz, I've used both MMBT5179 and BFR182 and they worked without any issue.
The last I'm building for a member, 2 x 22 MHz + 2 x 24 MHz, use the MMBT5179 and start properly at the right frequencies.
It's very strange, seems in your crystals the B-mode is dominant and more difficult to suppress.
But it should work tuning some component values.
Firstly, are you using exactly the component of the BOM?
If yes, you should try a little more tuning.
You could move back to the MMBT5179 (oscillator not starting) and try increasing the value of capacitor C2 to 6.8 or 8.2 pF, keeping in place the 12pF C3 capacitor.
If you have some MMB5179 you could try to replace it to find a device with higher gain.
Maybe a little tuning of the resonant LC tank L4/C1 could help. You could replace C1 with a 5-15 or 5-20 pF trimmer capacitor to change the resonance of the tank.
In the meantime I'm investigating about other bjt to replace Q1.
A few weeks ago Andrea sent me the Driscoll-Oven SC- Cut 22-24 Mhz assembly board,But I had previously for more than 6 months, the Clapp -AT Cut 22-24Mhz,The latter is a set of high quality,far superior to Crystek CCHD 957, the Crystek offers aggressive listening,projected forward related to the AT Cut-Clapp; with the Driscoll-Oven-SC Cut we enter In a listening class still superior,extremely dynamic,accurate without ever being aggressive,the sound scene becomes wider,From low to high It is that pleasure,
Thanks again Andréa for all this work, this patience which allows me today to have a strong system.
Thanks again Andréa
for all this work, this patience which allows me today to have a strong system.Great
Have anyone tried the 5/6 AT on Clock II board from Ian please ? Manual write it needs 11/12 XO minimum ?!
I have the choice wit AT 5.6 or AT 21.xxx but I'll like to favor the higher quality of the 5.6 XO (I have mainly 44.1 Library and I'm a TDA1541A user!
I wrote post few days ago, Ian wrote me 5/6MHz works with FIFO, but just 128x multiplier the minimum, so it will work on 44/48 KHz.
(PS: But I havent trying it yet)
Thanks gabor80,
I plan to use the 5.6xx then the 24.xxx AT-Cut for the second area of the Clock II:
TWTMC-C directly soldered on the TWTMC adaptor made for the DIL14.
I planed at the beginning to use the big pads made on the Kali Reclocker 2 crystal areas by removing its little smd NdK Crystal, but there are no vias on the 4 pads...
I plan to use the 5.6xx then the 24.xxx AT-Cut for the second area of the Clock II:
TWTMC-C directly soldered on the TWTMC adaptor made for the DIL14.
I planed at the beginning to use the big pads made on the Kali Reclocker 2 crystal areas by removing its little smd NdK Crystal, but there are no vias on the 4 pads...
- Status
- Not open for further replies.