https://www.diyaudio.com/community/...st-tht-i2s-input-nos-r-2r.354078/post-7578753Give us a schematic please. I/V has to exist somewhere (active or passive I/V), the DAC chip is with Iout. I don't remember exactly what I told you.
Attachments
Please explain how I can do that with my JLsounds v3, I have the normal version without the extra isolation...
I have both versions of V3. I cut that jumper at the bottom of the PCB and connected the power supply for XMOS where it says in the manual. Since I have a fixed +5V regulator (LM1117T-5.0), I put a 1N4001 diode in the plus line from the regulator to JLS. The voltage drop across the diode at that current (slightly less than 400mA) is about 900mV. A 2.2-2.5ohm 2W resistor can be placed instead of a diode. This reduced the dissipation on the BD139 by 0.3-0.4W. The power supply for the XMOS has a separate secondary 8-9VAC so that it is galvanically isolated from the rest. I grounded supply GND via 100nF + diode bridge 10A to the IEC connector GND (or to the grounded chassis for metal boxes).
Last edited:
I've got devices that are 10+ years, no issues whatsoever. So shouldn't be an issue here. You can always fix usb port from jlsounds to chassis via rubber. It will add to rigidity.
I remember now. That immediately seemed to me like a better option. I hope it can't affect the Iout badly when switching on and off. Connect the instrument to the output of that circuit and monitor the voltage during transients, on&off. Let's say a classic OPA when it is turned on quickly reaches around 0V at the output (with a small DC offset of a few mV), but when the power is turned off and the supply voltage drops below the minimum for the operation of the OPA (say below +- 3V), a DC voltage of the order of 2V appears at the output. And the output of the OPA is connected to the Iout leg of the DAC chip via an I/V resistor. So far, I haven't lost a single DAC chip because of that OPA behavior, but the question is what would happen in case of higher voltages at the output?
Last edited:
Thanks! I power the JLsounds v3 from the USB and use only one extra PSU for the clean side. So that means I would need to drop the voltage coming from the USB to around 4V? Since USB has varying voltages I would need to use an LDO to get the 4V, or is it safe to just insert a diode in the VUSB line before the JLsound board?
So you don't want to make a separate power supply but use +5V from the USB port?
For lowering the voltage to +4V, the LDO is certainly a best option. And a standard Si diode (or resistor) will also work, I don't expect any problems there. You can also make a RC filter there, for noise suppression.
A separate +4V internal power supply is probably the best thing.
For lowering the voltage to +4V, the LDO is certainly a best option. And a standard Si diode (or resistor) will also work, I don't expect any problems there. You can also make a RC filter there, for noise suppression.
A separate +4V internal power supply is probably the best thing.
You might have already known but i just want to say that you do not need the shifter ICs on the DAC board if you are using the JLSounds board.
Ok, I will try. I already tried to use a dedicated regulator from my existing PSU for the USB part, but that sounded worse, probably because GND was shared with the other regulators. A separate PSU should solve this, but for now I will power it from VUSB.
So I guess I can use the jumper at the underside of the JLsounds board to wire in a diode or RC filter...?
ADP7142 or ADP7182 are excellent regulators for these purposes. ADM7150 is much better, but it is also worse than Walt's shunt.
I was annoyed by the too strong light from the LCD display, so instead of the current limiting SMD resistor of 100 ohms (R8 on the PCB) I put a serial connection of 100 ohm resistor and a trimmer in front of the LED. The ideal trimmer value is 500 ohms. I didn't have that one, but I put a 2k2 trimmer and a 680 ohm resistor in parallel. It is not linear but it can be adjusted. That's why you should take a 500ohm trimmer to make a nice and smooth lighting adjustment. It is a 1602 LCD display that comes with the AK4113/4118 SPDIF receiver card. Or bought separately, it doesn't matter, they are all the same. Yellow/green is in kit by default. You can buy separately other colors, blue one is the best.
Attachments
Last edited:
I have just changed one of Miro Dacs using JLS and bypassing the shift registers, very noticeable improvement in SQ.
It should be like that. That jumper is on the +5V power supply from the USB port. It was only necessary that the voltage does not drop below 4V in order for the voltage regulators on the PCB to work properly.
