I cant explain more and why, and it is hard to measure with commom equipment, but i made some non-inductive Riv from isotan wire.
It is about 1ohm/cm and compare against common R types. It was huge differences. Amplification after Rivs usualy was test JFET circuit with 2SK170, with JFET buffer...
Also maybe even btter was Riv made from grafite pencil core. But it is pain in the *** to match and made phisical grip termination for wire to connect...
(Working best in the loudspeaker crosovers too)
Me too? On the simulations with current transformation it is disaster... Compared with voltage transformation with same xfrm model, souce and load...
And it is total reactance loading into the DAC out...
Even galvanic isolation is not present, because in 99.9% users are employing same Power Suply with same ground plane from DAC PS. That means there no galvanic islation from P to S because same ground is used...
Probably it has additonal filterihg HF effect, reducing anoying digital atrefacts in HF in that configuration.
😅 I know, I usually populate the ESD sensitive parts last. Even though I use ESD safe bracelet connected to the earth of a wall outlet.
Consider adding a D flip-flops to reclock the latch, bck and data signals for DAC ICs like in the attachment (SN74LV574).
It does not matter JLsound or York will be used, it is good to re-sync the signals close to DACs anyways.
That should have good impact on jitter but I do not have measurements data to confirm, unfortunately.
@miro1360 ,
I have even less simplier : Ã srandalone resistor . Music for the peeps before the end of the year or the end of the world or the end of your pocket (choose the item related to your situation)
🙂
Thanks Miro for your wonderfull thread, merry Christmas too or St Nicolas for the beginning of 2025.
I have even less simplier : Ã srandalone resistor . Music for the peeps before the end of the year or the end of the world or the end of your pocket (choose the item related to your situation)
🙂
Thanks Miro for your wonderfull thread, merry Christmas too or St Nicolas for the beginning of 2025.
Opa1641A is almost as good as your tpa6120 I/V !Guys try this my very cheap I/V and compare it to pricey opamps, it is time for cheap christmas presents 🤣
On miro board directly, you can't. You can make an adapter board and look at the datasheet to see which pin is which. Two for I/V, other two can be buffers, which you can push into class since you would be making an adapter board so you have freedom. What is the point in using that opamp anyway? Have it on hand?
or74f74 (2 needed)as pointed out Thorsten Loesh in the big thread (I have made a pcb for the TDA1541A with T. guidance but brijac should hits one as I stopped because of a troll and mods' banning, so I keeped it for I and few luckys)
It is a good idea eclipsevl. I wondered about that. But that little board above is made more for testing all sort of op amps, some needing buffers, and also compatible ad811 in smd if you dare, ahaha ! It mostly permits to have no current crossings, only feasible with 4 layers and some cares about current flows and digital lines. It is designed to avoid straycapacitance the best possible and the use of very fast op amps as well as not decompensated ones. adapter headers has some limits alas with high bandwidth op amps. So I just made JL Sounds + I2StoPCM (IanCanada) inputs to cut the costs (but anyway the layout is great : third iteration,I had to develop it in the cave or I would have been banished from Cristmass at home 🤣!)
But indeed a bigger pcb should have such reclocking and some should work as well to see if slew rate attenuation has some positive effect here on a Cmos chip. If I did, I 'd making it sligthy different though. (isolation between your board and the reclocking for instance. Not the same crystals as well. Or maybe MCLKslaving from Mori clock.
Last edited:
The biggest differences between the various 5532/5534 op-amps are noise and offset.
DC Offset:
Signetics without a line (series 8709D): -3.5 mV and -4.5 mV. This specific op-amp was removed from an old Yamaha equalizer, where we also salvaged the PCM63P-Y.
Signetics with a line (series 8825VG): -2.0 mV and -2.1 mV on both channels. This unit was taken from an old CD player.
TI’s 5532: Offset values ranged between 10 mV and 12 mV.
In my opinion, the combination of the Signetics NE5532 and Nichicon Muse UES capacitors is a match made in heaven.
The synergy is undeniable. The warm, musical presentation of these capacitors complements the precision of the Signetics op-amps, enhancing the overall listening experience.
It's a pity that all of this is now obsolete.😕
Yeah I wish I had some to test as buffer as well. Many considers the JRC then the TI are not jumping above the knee of the Signetic !
AD8597 is a good bet as it is very near ad797, cheap, better sounding with a darker character. Not the thin sound of a LM4562 in some layouts.
The Ues are not warm at all : they are crystal clear at the opposit of usual Nichicon sound. So if it sounds so, it is elswhere you did the good job. Kz and eS are veery two non typical Nichi caps 🙂
AD8597 is a good bet as it is very near ad797, cheap, better sounding with a darker character. Not the thin sound of a LM4562 in some layouts.
The Ues are not warm at all : they are crystal clear at the opposit of usual Nichicon sound. So if it sounds so, it is elswhere you did the good job. Kz and eS are veery two non typical Nichi caps 🙂
It is interesting that all decoupling capacitors on Yamaha's equalizer are Elna Cerafine 47uF 16v Blue😀
There are a lot of old DACs that either don't have a USB input, or have a bad USB input. And as a rule, they have more SPDIF inputs, which are very good. Then the JLS board is used as a USB input connected to one of the SPDIF inputs. It can be in a separate box or it can be built into an existing DAC. There are other devices that perform this conversion, only for this one we are sure it is good.
Similarly, I added Bluetooth to a friend's DIY DAC. Children like to play, but they only know about BT and mobile phones.
Last edited:
