Thanks Vunce for your detailed description. The adjustments are added on your tips
No worries, DIY things increase over time
Miro, Would you consider a layout for multiple PCM56, which has the same pinout as ad1851. If you make a jumper for the glue logic i.e. the 74hc164 ic's, for 16 bit as well as 18 bit dacs, the same board can be used for the AD1860/PCM61 dac chips.
All these dac chips are still available and the PCM56 is especially inexpensive.
See for example 10PCS Audio D/A Converter IC BB/TI DIP-16 PCM56P-J PCM56P J PCM56PJ PCM56P-JG4 | eBay
you already have glue logic for 18 bit dacs, which is 13 bit delay for data to the right channel and a 32 bit buffer for data to the left channel.
For glue logic for 16 bit dacs such as PCM56, you would have to delay data to the right channel by 15 bits and the same 32 bit buffer for data to the left channel.
This can easily be achieved in your current designs by a jumper. What really piques my interest is if you made a pcb for 8 dac ic's 4 for the left channel and 4 for the right channel. It is said in technical documents that every doubling of dac ic's reduces the errors and noise floor.
Of course to do this, the pcb board would have to leave out the i/v stage. The dacs could be designed to utilise the internal opamps for voltage output and have an alternative for current output by use of a jumper
I think a separate i/v pcb would not be a big problem. More importantly, the pcb's could be parralled for multiple pcb's. In that case, the sky is the limit.
I hope you give this suggestion some serious consideration.
Many thanks in advance.
@mllum
@Guqin7
These DAC chips are on my to-do list and I create something in the future. Giving it parallel is a nice approach and the option will be considered for cheap DACs
Look forward to it, I happened to have four pcm63 and ad1862 ,and wants to ab with ad1862
Hi guys,
I went this route already and I warmly advice you to stay on the AD1862. Far superior to the AD1865 K version I have. Both my Painkiller ad1862 discrete I/V DAC (project here by Painkiller fellow) and the Miro's with chips sound better, no doubt.
Anyway I have original Rochester AD1862 from the Painkiller project, but the one I bought lately for Miro's project on Ebay were fake and one had a leg broken inside the body (unfixable)... So if Paddy one day is going again for a AD1862 GB, count me in, story to have the two AD1862 DAC working without dac chips swap.
I really think and experienced better result without smd/Dil adaptators for the fastest and best oaps for the Miro's project.
As the boards drawing are nicely offered by Miro and Vunce (thanks guys) for some options, I really advice and share my experience on that : better your energy and money focused on better PS and typologies, different caps and passive parts experience that makes really sound a dac board different ways to adapt to the rest of your hifi system and make a DAC to sounds bad or good at iso layout whatever say a scope deaf fanatic that never listen to and finally go the route to the caps tweaking himself at the end following my own advices -I can not laugh more about the Soekris DAC whatever the viking wet powder zelote I that pursue me from threads to threads are saying... hey I was sinbined for that cause I should accept to be named a tweak mania guy without answering... I really hesitated to come back to diyaudio due to some unfair judgment and not so clean souls-
My two cents...
I went this route already and I warmly advice you to stay on the AD1862. Far superior to the AD1865 K version I have. Both my Painkiller ad1862 discrete I/V DAC (project here by Painkiller fellow) and the Miro's with chips sound better, no doubt.
Anyway I have original Rochester AD1862 from the Painkiller project, but the one I bought lately for Miro's project on Ebay were fake and one had a leg broken inside the body (unfixable)... So if Paddy one day is going again for a AD1862 GB, count me in, story to have the two AD1862 DAC working without dac chips swap
.I really think and experienced better result without smd/Dil adaptators for the fastest and best oaps for the Miro's project.
As the boards drawing are nicely offered by Miro and Vunce (thanks guys) for some options, I really advice and share my experience on that : better your energy and money focused on better PS and typologies, different caps and passive parts experience that makes really sound a dac board different ways to adapt to the rest of your hifi system and make a DAC to sounds bad or good at iso layout whatever say a scope deaf fanatic that never listen to and finally go the route to the caps tweaking himself at the end following my own advices -I can not laugh more about the Soekris DAC whatever the viking wet powder zelote I that pursue me from threads to threads are saying... hey I was sinbined for that cause I should accept to be named a tweak mania guy without answering... I really hesitated to come back to diyaudio due to some unfair judgment and not so clean souls-
My two cents...
Last edited:
Hi diyiggy, nice to hear about you again
Thank you for your experience and I'm glad the DAC didn't disappoint you.
>>I really hesitated to come back to diyaudio due to some unfair judgment and not so clean souls
You can come in my thread at any time, you are more than welcome here. Let the others hearing die in endless delta-sigma tweaks
My thread is something other, is based on experience and everyone here can share own opinion, there is a good mood. We will enjoy real analog sound not redrawn by digital filters.
Join us back and don't leave a hole here, digital sound would happen to flow through it
Thank you for your experience and I'm glad the DAC didn't disappoint you.
>>I really hesitated to come back to diyaudio due to some unfair judgment and not so clean souls
You can come in my thread at any time, you are more than welcome here. Let the others hearing die in endless delta-sigma tweaks
My thread is something other, is based on experience and everyone here can share own opinion, there is a good mood. We will enjoy real analog sound not redrawn by digital filters.
Join us back and don't leave a hole here, digital sound would happen to flow through it
Thanks Miro,
I did few things during the last month. I almost finished the cabinets of a clone of the BBC 30.2 monitor from Harbeth. I just clone the looking as I have of course not the main original plastic cone driver or the filter plan either. As it's a difficult gambling, I called mine The Hard-bet... story to swap my pain for a pun, sort of... And yes this is Eucalyptus outside py as the H. Anniversary Eucalyptus veenered serie.
I'm boxing in the same time your DAC, the Painkiller one and a TDA 1541A more I'm working on to try to make a BOM with easier passive parts as my main original which sounds beautifull have some difficult parts to source which are not here for the snobism but because the synergy is good and it may help one guy or two here to wake up a normal TDA1541A chip and not the crown one or the late South Asia made I usually play with and listen to every day the cosmos makes !.
I have still the bunch f resistors and caps to continue to test of the Miro's v1.1 but I was a little stopped by the lack of a free USB to I2S board that is easy to connect as I don't like as you know definitive soldering of the front end, but more uf-l cables... I have been purchasing the JLSound board I'm waiting for... I already have the Wave I/O board and all the stuffs by IanCanada...
This hobby is mad, do I need a fourth pair of Loudspeaker... of course not, but this one is just for the fun and is my firt DIY Loudspeaker as I am very impress by all the good work of many members here.
Ah, the Dacs... just the ... , nah, you don't want to know the number I already still have and the number of CD players/Dacs I had, I'm here for a group therapy about that 
!
I did few things during the last month. I almost finished the cabinets of a clone of the BBC 30.2 monitor from Harbeth. I just clone the looking as I have of course not the main original plastic cone driver or the filter plan either. As it's a difficult gambling, I called mine The Hard-bet... story to swap my pain for a pun, sort of... And yes this is Eucalyptus outside py as the H. Anniversary Eucalyptus veenered serie
.I'm boxing in the same time your DAC, the Painkiller one and a TDA 1541A more I'm working on to try to make a BOM with easier passive parts as my main original which sounds beautifull have some difficult parts to source which are not here for the snobism but because the synergy is good and it may help one guy or two here to wake up a normal TDA1541A chip and not the crown one or the late South Asia made I usually play with and listen to every day the cosmos makes !.
I have still the bunch f resistors and caps to continue to test of the Miro's v1.1 but I was a little stopped by the lack of a free USB to I2S board that is easy to connect as I don't like as you know definitive soldering of the front end, but more uf-l cables... I have been purchasing the JLSound board I'm waiting for... I already have the Wave I/O board and all the stuffs by IanCanada...
This hobby is mad, do I need a fourth pair of Loudspeaker... of course not, but this one is just for the fun and is my firt DIY Loudspeaker as I am very impress by all the good work of many members here.
Ah, the Dacs... just the ... , nah, you don't want to know the number
I already still have and the number of CD players/Dacs I had, I'm here for a group therapy about that !Attachments
Last edited:
PSU 1 not only for AD1862
PSU 1 is a simple DIY power supply based on MC78xx/MC79xx regulators. These regulators are underestimated by many people due to the higher noise specs compared to other „more advanced“ regulators. This regulator is very stable even for a pulse load. For audio applications it is sufficient form my experience ... and is very cheap
In the basic configuration, components are set to output +/-5VD and +/-12VA for AD1862 DAC.
Since PCB is universal, you can solder components for any other DAC voltages.
The PSU needs 2 separate transformer windings. Voltage is rectified through schottky diodes bridge and goes to a big CRC bank, which makes the voltage smoother for regulators. CRC combination works as a filter with frequency cut of 23 Hz. It should help to the purity.
An external regulator can be sourced from the CRC (PCB contacts V+, GND, V-).
Output from the regulator includes another big low ESR capacitor for the final filtration. The 78/79 regulator has an emitter folower stage with no gain, so the big output capacitance makes no difference to the phase shift, thus stays stable
The output power and heatsink is a DIY question, but since the AD1862 DAC consumption is small, it should go without heatsink.
PCB size is 75x150mm and can be cheaply ordered from jlcpcb or any other PCB manufacturer (gerbers included in the post).
(it was not tested at the time of writing this post)
Example list of components (copied from BOM picture, except resistors)
mouser:
603-FMP300FRF73-1R
R79IC3100Z340J, 505-MKP2D031001FJO00
505-MKP20.33/100/5
647-UHE1E472MHD
647-LKG1V682MESYAK
604-L483GDT
511-1N5822
863-1N4007G
863-MC7805ACTG
863-MC7812ACTG
863-MC7905ACTG
863-MC7912ACTG
534-4628
576-0216001.MXR20SPP
651-1729128
651-1729131
ebay: 233823483402, 233670142926
PSU 1 is a simple DIY power supply based on MC78xx/MC79xx regulators. These regulators are underestimated by many people due to the higher noise specs compared to other „more advanced“ regulators. This regulator is very stable even for a pulse load. For audio applications it is sufficient form my experience ... and is very cheap
In the basic configuration, components are set to output +/-5VD and +/-12VA for AD1862 DAC.
Since PCB is universal, you can solder components for any other DAC voltages.
The PSU needs 2 separate transformer windings. Voltage is rectified through schottky diodes bridge and goes to a big CRC bank, which makes the voltage smoother for regulators. CRC combination works as a filter with frequency cut of 23 Hz. It should help to the purity.
An external regulator can be sourced from the CRC (PCB contacts V+, GND, V-).
Output from the regulator includes another big low ESR capacitor for the final filtration. The 78/79 regulator has an emitter folower stage with no gain, so the big output capacitance makes no difference to the phase shift, thus stays stable
The output power and heatsink is a DIY question, but since the AD1862 DAC consumption is small, it should go without heatsink.
PCB size is 75x150mm and can be cheaply ordered from jlcpcb or any other PCB manufacturer (gerbers included in the post).
(it was not tested at the time of writing this post)
Example list of components (copied from BOM picture, except resistors)
mouser:
603-FMP300FRF73-1R
R79IC3100Z340J, 505-MKP2D031001FJO00
505-MKP20.33/100/5
647-UHE1E472MHD
647-LKG1V682MESYAK
604-L483GDT
511-1N5822
863-1N4007G
863-MC7805ACTG
863-MC7812ACTG
863-MC7905ACTG
863-MC7912ACTG
534-4628
576-0216001.MXR20SPP
651-1729128
651-1729131
ebay: 233823483402, 233670142926
