Hi xaled,
Yes, you are right.
For my FIFO solutions, all XOs will be interchangeable. That's a transition. Or you can call it "DIY friendly"
.Regards,
Ian
Hi Greg,
All on-board regs are LT3042. So, you can use 7V DC power as input without any problem.
Ian
Another view
My take on this is somewhat different... as Ian is combining the functionality from what is now 3 separate boards (each larger than the standard RPi DAC HAT) into 1 board, I'm assuming he is wrestling with compromises. IF one trade-off is overall performance versus easily interchangeable clocks, I vote for performance. I am of the mind that RPi players are on the cusp of moving from upper mid-fi to lower high-end based on some of the products introduced in the last year or so and some to be released soon and this will be a good addition to that movement.
Thanks for that confirmation. Curious if it'll make any noticeable difference. I DON"T expect to hear one, but reality often conflicts with my expectations and surprises me!
Greg in Mississippi
My take on this is somewhat different... as Ian is combining the functionality from what is now 3 separate boards (each larger than the standard RPi DAC HAT) into 1 board, I'm assuming he is wrestling with compromises. IF one trade-off is overall performance versus easily interchangeable clocks, I vote for performance. I am of the mind that RPi players are on the cusp of moving from upper mid-fi to lower high-end based on some of the products introduced in the last year or so and some to be released soon and this will be a good addition to that movement.
Thanks for that confirmation. Curious if it'll make any noticeable difference. I DON"T expect to hear one, but reality often conflicts with my expectations and surprises me!
Greg in Mississippi
Last edited:
I can answer that... no. Adjusting the volume in Volumio changes the volume using a software process in Volumio and the associated device drivers. Changing it with Ian's controller changes it using the firmware-based processes in the ESS ES9028Q2M chip. People have said that the ESS on-chip volume control is VERY good. I personally haven't compared much, so I cannot comment, but my experience with driver/software volume controls has generally been they work best when at full volume (that is not engaged at all).
Greg in Mississippi
Last edited:
@Madds1 & @Rod Coleman, what is the formula for calculating the capacitance value in response to the paralleled resistor value for this circuit?
TIA!
Greg in Mississippi
I thought there was a Linux driver available to control the ESS controller? Via ALSA? Deal breaker for me otherwise.
Once the driver that invokes the HW volume control is written and in a distro, it should be available for use. BUT AFAIK, there is not one for the ES9028Q2M that will work with this DAC board. If there is, I apologize for the confusion and mis-information. Very interested if anyone has contrary info.
@Madds1 & @Rod Coleman, please creatively ignore my question above. I went back and re-read Madds1 post quoted above and realized I had not included the ES9028Q2M output impedances, the “(403*2” above. Once I included that, my calcs for the rolloff frequency now match yours and I can tweak my setups as needed. DUH!!!!
I had first tried a 330R/44nF combo on the LL1570XL, but that sounded like it was rolling off too low. Now calculating the rolloff, I get around 16K… too low. Going to 330R/33nF is much better (and calcs at about 20K), but now seems a bit high. I’ll try some parallel trimming resistors to see if I can achieve the Goldilocks effect of ‘Just Right’.
I also cleaned up the wiring on the Onetics 600R:600R with the 22uF PPS SMD grounded CT AND listened more. I do prefer the now-corrected LL1570XL w/Joe R/et al input filter, so I’ll convert the Onetics to that configuration and give it a new listen.
I also tried 7V input into the board. THAT DID NOT WORK, no sound. I haven’t tried to track down why, but as soon as I realized what was going on I disconnected it and hoped I hadn’t permanently damaged it. Going back to 5V input, it works again, but I’m still concerned something may have gotten damaged… at some point I’ll try to determine where the 7V went that caused it not to work and assess whether it could have hurt something on the board or whether I can arrest my paranoia.
Next trials will involve setting up the other 3 transformers I have with the input filter configuration… a pair of LL1684, of Cinemag CMLI-15/15B, and of Sellarz (from EBay) I/V transformers (20R input winding, 1:5.477 winding). AND build-up an OpAmp I/V using the OP1632 (which will give a reference similar to other OpAmp I/Vs I've heard). I'm not sure the Sellarz will work... the 330R I used has the overall output at the ragged edge of sufficient volume for my systems. I'll also try the Lundahls configured for 1:2 step-up... then I can go lower on the input resistor.
More next week.
Greg in Mississippi
Last edited:
Another DUH!, I mean '... some parallel trimming capacitors...'.
Greg in Mississippi
Code:
I added 470pF to the 33nF on the Lundahl LL1570XL and after a quick listen, that is sounding better. At some point I'll replace the 470pF with 1000pf, but before that I'll first listen a bit more. ALSO I've reconfigured the Onetics 600R:600R with the Joe R / Madds1 / Rod Colemen setup using 330R & 33nF + 470pF and I'll listen to that first to help triangulate the sonics.
Sounding good now.
As above, there's lots of other alternative output stages / I/Vs to try along with additional separate regulation stages (instead of just 1 for the DAC rails) AND also trying it in Sync mode instead of ASRC. I found I always preferred the previous generation of ESS chips in Sync mode, but some have suggested the difference is smaller to un-noticeably with this generation. Of course, I'll have to try it myself.
Greg in Mississippi
Madds1,
No arguments from me on that. I'm currently using Wima polyprops as that is my standard go-to small-value filter cap line. I will also try some SMD PPS, but both don't have them here in the right values currently AND am trying to determine if there are significant sonic differences between the lines available from the different manufacturers I've used in the past... Panasonic, Rubycon, and NIC Components.
I've not been a big fan of silver mica over the years, but am willing to try them again. AND I've never tried PIO, so willing to try them too.
AND once I've settled on a configuration (output transformer and input filter cap value combo), I've also considered buying some upgrade caps such as the Amtrans AMCH polypropylenes or RelCap polystyrene or Teflon).
Remember, my signature is 'Everything matters!'. I don't discount the importance of ANY capacitor in a setup.
Greg in Mississippi
No arguments from me on that. I'm currently using Wima polyprops as that is my standard go-to small-value filter cap line. I will also try some SMD PPS, but both don't have them here in the right values currently AND am trying to determine if there are significant sonic differences between the lines available from the different manufacturers I've used in the past... Panasonic, Rubycon, and NIC Components.
I've not been a big fan of silver mica over the years, but am willing to try them again. AND I've never tried PIO, so willing to try them too.
AND once I've settled on a configuration (output transformer and input filter cap value combo), I've also considered buying some upgrade caps such as the Amtrans AMCH polypropylenes or RelCap polystyrene or Teflon).
Remember, my signature is 'Everything matters!'. I don't discount the importance of ANY capacitor in a setup.
Greg in Mississippi
Breaking news on Sync versus ASRC modes for the latest generation of ESS chips... there was some discussion over the past several days of getting that to work as expected on the Buffalo-IIIPro thread starting here:
Introducing the Buffalo III-SE-Pro 9028/9038
After some troubleshooting, TP released a trial version of their firmware to fully enable Sync mode and deactivate the DPLL. This is a comment from TP after listening to the fully-enabled Sync mode:
Introducing the Buffalo III-SE-Pro 9028/9038
Hmm... I'm now MUCH more eager to try this DAC in full Sync mode with the DPLL off, but I suspect that'll require an update to Ian's Controller's firmware... assuming that this option is even available with the ES9028Q2M. I don't have the Q2M datasheets, Ian can you check and confirm if that option is available?
Greg in Mississippi
Hi Greg,
I have a "break news" for you too:
The ESS controller that you are having right now supports the new ESS true SYNC mode already. Yes, you are right, this new mode is available with ES9028Q2M.
To set this mode, you will need:
Manual - "MCLK SYNC mode" - select "SYNC MODE stop DPLL", then confirm.
You can use either the encoder or the Apple remote.
Please make sure you have already had external sync MCLK connected before doing above setting.
Good luck!
Ian
Last edited:
Hi guys,
Is it worth to buy a pair of LL1544A?
The price from KK audio is $85 each.
K&K Audio | Prices
I have a pair of LL1674 now, I'm really interested in getting them compared.
Thanks,
Ian
Is it worth to buy a pair of LL1544A?
The price from KK audio is $85 each.
K&K Audio | Prices
I have a pair of LL1674 now, I'm really interested in getting them compared.
Thanks,
Ian
Ian,
YAY!!! You are the BEST!
Hmmm... what clock did you use on the DAC board? I have some of the SDA versions of the NDK clocks coming and planned to wait until there were here and I could swap out the clocks on the Kali before trying Sync mode. BUT if you did not use an SDA-level clock on the DAC board, it would be worthwhile to try it sooner rather than later.
Let me know.
In other news, while I cannot swap from one output board to another for a close to direct comparison, I CAN swap them quickly enough that my PS Audio BHK-250 does not fully cool down from operational while in standby, which means it is back up to operational goodness in 10 minutes or less rather than the 1 hour plus it takes from a long standby.
I mention this as I swapped in the Onetics 600R:600R with the 330R/33nF+470pF input filter this afternoon and listened for 20-30 minutes. They are definitely the winners to-date in this configuration, sounding more alive, detailed, dynamic, and delineated than the Lundahl LL1570XLs (which I expected to be behind the amorphous-core LL1684s anyway). Very rewarding time listening to them, I'm now pumped to build up a couple more of your LT3042 boards, setup separate regulation for the different rails, and try it in Sync mode. Also now more curious on how the other output setups will sound.
Likely not much more action this week besides listening sessions, got a pretty heavy work week. BUT I'm pumped after both knowing the setup is full-Sync-Mode-no-DPLL ready AND hearing the Onetics with the Joe R/Madds1/Rod Coleman input filtering setup.
Ian, again THANKS!
Greg in Mississippi
P.S. I bought the Onetics over 7 years ago thinking to use them as output transformers for a Buffalo-II setup I never got running. AND I was discouraged by some of the comments from people who pooh-pooh both using transformers for outputs on the ESS DACs AND running them in voltage-out mode. But this experience has opened my eyes/ears! These ended up being a super bargain, based on what I heard this afternoon!
Last edited:
Ian,
One of us really needs to do this. I had planned to do so... we really need to compare them to the LL1674 & LL1684 as well as the other options. AND maybe the Sowter I/V transformers such as the 1465 & 1495 too? I'll contact you via email to coordinate.
Greg in Mississippi
Last edited:
I’ve been running my BII in voltage mode with trafos for years with great results.
I’m wanting to try the LL1544A as well.
Is there any reason I shouldn’t try a “hotrod” (raw output from the dac) connection to my nCore 400 mono blocks?
Differential amps, so a volt or two of dc shouldn’t matter?
P.S. Once I have it up and running I’ll be able to cannibalize my current rig for LL1527 trafos
I’m wanting to try the LL1544A as well.
Is there any reason I shouldn’t try a “hotrod” (raw output from the dac) connection to my nCore 400 mono blocks?
Differential amps, so a volt or two of dc shouldn’t matter?
P.S. Once I have it up and running I’ll be able to cannibalize my current rig for LL1527 trafos
Last edited:
LL1544A are cheaper than LL1674 but LL1674 are rated to +29dbu for 1% @50hz distortion vs LL1544A is at +20dbu. Our line levels at much lower than that so not sure if the differences will be audible.
Windings also matter but there is no information on that - except that LL1544A has more resistive losses (higher DCR)
As an aside - something useful - all the specifications for all Lundahls in one PDF
http://www.lundahl.se/wp-content/uploads/catalogue/Catalogue_20170118.pdf
Windings also matter but there is no information on that - except that LL1544A has more resistive losses (higher DCR)
As an aside - something useful - all the specifications for all Lundahls in one PDF
http://www.lundahl.se/wp-content/uploads/catalogue/Catalogue_20170118.pdf