I made and tested AD1862 (1.3 with I2S ufl) and PSU2 today, but there is a problem and I want to get help. I changed the Jumper resistor from 0 to 2.2R because I didn't read the build guide carefully.
The output voltage of PSU2 is OK.
I connecting Combo384 (USB to I2s) and testing the playback, there is no sound at all from the left channel and sound from the right channel, but the volume is very low. Even when the volume of the power amp is turned up, it is very low. Perhaps the DAC output is very small.
R8 and R9 are 2.4k. OP IC is LM6171.
For the left channel, I will check the soldering first, but I don't know how to find the cause of the weak right channel output.
I would appreciate it if you could give me a guide on how to do trouble shooting.
@soonmo
- desolder away R4, R5, R6, R7, TR1, TR2 (I assume you do not have adequate setup equipment for this feature, so it is better to let it empty)
- Combo384 configuration: Stream format: I2S, Word length: 24 bit (64fs) ... maybe 32 bit can work
- in PC settings - Control panel - Sound - configure the device for 24-bit 48kHz first ... and try another configurations as well
I tested the stopped clock circuit, it works 
I wired my existing glue logic (configured for 20bit output as shown in this post) with the additional 4 ICs shown in Miros schematic. Attached some scope shots: yellow is LR, green is clock output, red is data out left and blue is data out right. If fed with 24bits it shows some pulses on the data lines before the clock starts, but this should be OK according to Miro. 24/96 seems to work as well!
I wired my existing glue logic (configured for 20bit output as shown in this post) with the additional 4 ICs shown in Miros schematic. Attached some scope shots: yellow is LR, green is clock output, red is data out left and blue is data out right. If fed with 24bits it shows some pulses on the data lines before the clock starts, but this should be OK according to Miro. 24/96 seems to work as well!
Attachments
I just looked on mouser and the ada4627’s they offer are 36 volt. I thought the opamps were supposed to be 12 volt or 5? Please explain. Also interested in what other opamps are best to try? I would like to try a few to hear what’s best for me.I can confirm that ADA4627 is worth the price
Your knowledge is amazing and I appreciate your help
Regards
@intojazz There are many opamps which I have not tried and I want to (one of them is THS4031)
From many tested opamps I know only a few. The best in price/sound is LM6171, the ratio is not beaten yet.
Other opamps: ADA4627, OPA627 (pricey as hell), AD797 (with mod from @Hidy: #2687 ), ... while testing I would pick something from High-speed opAmps or Instrumentation opAmps. Maybe some Renesas opAmps can sound good, like EL5160 or EL5162 ... or no one tested my OPA627 discrete clone #2201
From many tested opamps I know only a few. The best in price/sound is LM6171, the ratio is not beaten yet.
Other opamps: ADA4627, OPA627 (pricey as hell), AD797 (with mod from @Hidy: #2687 ), ... while testing I would pick something from High-speed opAmps or Instrumentation opAmps. Maybe some Renesas opAmps can sound good, like EL5160 or EL5162 ... or no one tested my OPA627 discrete clone #2201
Last edited:
We can expect a nice PCB with stop clock 🥳Our regiment has arrived. Finally I started the DAC from the respected Miro. So far from Amanero which is connected to BBB with Pure firmware. For some reason, only a crackle is heard directly from BBB + Recklocer, you have to conjure with the output data, maybe it's the 45MHz generators instead of 24MHz.
BUT!!! Even so, he plays very, very well. Very melodic and detailed. I started to hear in the tracks what I hadn't paid attention to before. I like.
The only thing I wanted to clarify is, if I have 11.5 or so on the +/- 12V power supply, can this affect the sound and the operation of the AD1862 or only the output signal level?
I use NE5534 op-amp, it is interesting to try LM6171 and LT1468.
There are also LT1028, LT1363, LT1227, AD817, AD811, AD586, LME49886, AD4627, AD825, AD823, AD827, AD825, TL071, you need to try them as it is ...
Miro thank you for your hard work, great device. Super!!!
I will also listen, I will compare it with a professional sony on 1702, and if you like it more on AD, you have to think about an 8-channel device in the car ...
BUT!!! Even so, he plays very, very well. Very melodic and detailed. I started to hear in the tracks what I hadn't paid attention to before. I like.
The only thing I wanted to clarify is, if I have 11.5 or so on the +/- 12V power supply, can this affect the sound and the operation of the AD1862 or only the output signal level?
I use NE5534 op-amp, it is interesting to try LM6171 and LT1468.
There are also LT1028, LT1363, LT1227, AD817, AD811, AD586, LME49886, AD4627, AD825, AD823, AD827, AD825, TL071, you need to try them as it is ...
Miro thank you for your hard work, great device. Super!!!
I will also listen, I will compare it with a professional sony on 1702, and if you like it more on AD, you have to think about an 8-channel device in the car ...
No, this is according to this scheme, I drew for myself both unipolar and two-polar scarves. These stabilizers have very good parameters. And the components are not expensive. Who is interested I will send gerberas.
On the positive side, the stubs are right next to the consumer.
Linear power
Keep it, try it, you might like how they work.
From the latest recommended by the developer.
C1, C3 electrolytes. C6 is, in principle, better than ceramics, but it will work fine with an electrolyte.
C4 can be both electrolyte and ceramics. This is essentially an output air conditioner for a regular stub at 317. In principle, 317 will work slightly differently for different capacitors, but I don’t remember anything causing instability.
If C6 and C4 are electrolytes, preferably with not very high ESR. Well, probably no more than 100 milliohms.
Denominations, yes, in general, it does not matter. C1, C3-value affects mainly the suppression at the lowest frequencies. And the launch speed.
C6- in general, it is not very important, it affects the suppression of interference at medium-high frequencies. But the main goal is rather to provide a relatively low impedance at the MF / HF in order to reduce the influence of parasitic inductance of the input wires.
This can affect stability. Well, in front of it, you can turn on some kind of small resistor or ferabid to increase passive suppression at the RF. You just need to take into account the voltage drop across this resistor.
The C4 value affects the transient and phase margin 317. But in principle, the range is very wide.
Those. in principle, you can change almost all the ratings of the air conditioners every 10 times and everything will work. But the parameters will change.
Sopno, in order to assess what and how changes and need a SPICE model. The data in the hardware may of course differ, I do not pretend to have a very high accuracy of the model. But qualitatively, the impact of the changes will persist and easily
to estimate what influences what.
Attachments
Hello,
I’m going to build a very similar dac. What smd adapter did you use?
My experience is small and I’ve never used one of these before but see it’s the only way to incorporate smt devices on this board.
thank you
@miro1360:
Thank you.
I tested it by connecting the raspberry pi and i2s instead of the pc and installing moOde. I installed moOde and changed the settings to output in 24bit/48khz.
After re-soldering and playing music, the left channel plays normally. This is a sound I like very much.
There is no sound in the right channel. I suspect 74ahct164d on the right channel, but I would like to proceed with troubleshooting for more accurate diagnosis.
I think I could check something with the points marked DL and RL on the board using oscilloscope. I would like to know what this point is used for.
Thank you in advance.