I was lucky to get some AD1865N-K dacs and plan to build dual mono dac with it.
Did anyone do the same and can you please share the output circuit? I would prefer to go with passive I/V (resistor) stage. While it is a simple resistor between the current output end the ground for each channel using single chip, what is the best approach using two chips (dual mono)?
Is it possible at all to use passive resistor also for dual mono differential current output?
Did anyone do the same and can you please share the output circuit? I would prefer to go with passive I/V (resistor) stage. While it is a simple resistor between the current output end the ground for each channel using single chip, what is the best approach using two chips (dual mono)?
Is it possible at all to use passive resistor also for dual mono differential current output?
All is posible, but the simple resistor way will not give you the best sounding result. You might want a lowish impedance to be seen at the outputt, so it ask few more parts. There a lot shematic here and elswhere, google andrea Ciufolli diy site for instance.
Hope that helps
Hope that helps
Thanks diyggy. I think Andrea is using single AD1865 with output transformer and tube buffer.
I think there are basically two options - use somehow differential current output from each chip or connect output current pins in parallel? Either using a resistor for I/V conversion or maybe an output transformer?
both have the same purpose, increasing the curent to lowish the resistor value. traffo adds galvanic isolation.
you can also use one chip by chanel, try unbalanced...
look for ad1862 Miro1360 thread that has acheived an I/V something simple as you plan.
Also : Passive output and I/V passive stage with Lundhal by K&K
with active stage there is also Sen/Zen shematic or good shematic by Pedja rogic with discrete or with aop with no feedback : opa861.
These AD chips diserve a little care but will give you a lot imo.
you can also use one chip by chanel, try unbalanced...
look for ad1862 Miro1360 thread that has acheived an I/V something simple as you plan.
Also : Passive output and I/V passive stage with Lundhal by K&K
with active stage there is also Sen/Zen shematic or good shematic by Pedja rogic with discrete or with aop with no feedback : opa861.
These AD chips diserve a little care but will give you a lot imo.
Last edited:
This is what I would like to do - one AD1865 chip per channel. I will probably start with resistor i/v and parallel output current pin to be able to use lower resistor value. I wasn’t able to find any schematic with parallel current output - is it as simple as connecting both current oins together and through resitor (i/v) to the ground?
Not sure how differential passive i/v can be dome using resistor if unballanced out is needed (grounded)
output trafos do that.
If I'm not mistaken : Lchip 1 + Lchip2 then the resistor 'half the value of one chip) ; idem for the Right chanel.
The thing you should try to look at is if the I output has an internal protection in the chip, if not use maybe reversed signal diodes in your experiments.
Dual chips stands also for more complex pcb for the ground. Not sure it worths the effort to divide by two the i/v resistor : better to use traffo then if you want to reduce the R value but staying with passive parts. Anyway you need to focus on the outputt impedance you will have at the end and/or think to a buffer.
I'm pretty sure there are plenty of shematics here and elswhere.
In your shoes if I had the joy t ohave two AD1865K (I have one and not sure it's genuine as it was an Ebay Kit from China, certainly more a std rebadged K, but who knows?) : I will follow with one the passive road : trafo + Resistor and the active one OPA861 (no feedback)+Resistor, or one of the latest Ti oaps with feedback. I would let the diferential job to a traffo at the end if you need it over a single ended solution.
If I'm not mistaken : Lchip 1 + Lchip2 then the resistor 'half the value of one chip) ; idem for the Right chanel.
The thing you should try to look at is if the I output has an internal protection in the chip, if not use maybe reversed signal diodes in your experiments.
Dual chips stands also for more complex pcb for the ground. Not sure it worths the effort to divide by two the i/v resistor : better to use traffo then if you want to reduce the R value but staying with passive parts. Anyway you need to focus on the outputt impedance you will have at the end and/or think to a buffer.
I'm pretty sure there are plenty of shematics here and elswhere.
In your shoes if I had the joy t ohave two AD1865K (I have one and not sure it's genuine as it was an Ebay Kit from China, certainly more a std rebadged K, but who knows?) : I will follow with one the passive road : trafo + Resistor and the active one OPA861 (no feedback)+Resistor, or one of the latest Ti oaps with feedback. I would let the diferential job to a traffo at the end if you need it over a single ended solution.
Last edited:
I'm just curious if anyone has ever seen an AD1865 datasheet that has the N-K version listed on it, I never have which makes me question what the K version really is.
Besides Rochester, who have mins of 10 or 11, where are people buying known good 1865s.
Besides Rochester, who have mins of 10 or 11, where are people buying known good 1865s.
Actually the only dac that I can find in the AD lineup with a N-K version is the AD1864, so I suspect if you guys are buying AD1865N-Ks then they might not be what you think they are, otherwise they would be in the AD1865 datasheet as well as in the AD1864 datasheet.
Yes I know Rochester shows them but why doesn't Analog Devices show them? Or did they just forget to add the suffix to the datasheet, hmm, I think not.
Yes I know Rochester shows them but why doesn't Analog Devices show them? Or did they just forget to add the suffix to the datasheet, hmm, I think not.
Back in the day (mid 1990s) Analog Devices were sending samples of AD1865N-K ICs. I received several from them and replaced the stock AD-1865N in my early Audio Note DAC kit with one. So yeah, Analog Devices did not update their spec sheets.
That said, no one should currently buy from Asian sources.
The AD-1865 has cophase current (and voltage) outputs between each internal channel. That means there is a latency between R channel and L channel of half the sampling rate frequency. In other words, R and L outputs do not dump parallel current in sync and do not natively sum without some engineering to address that.
The information when the dac has to go to conversion is in the LR Latch digital line.
When the latch is the same for booth sections, they will go to conversion in the same time and will output io at the same time.
I think there will be no delay between the channels.
Digital data input format is given in the data-sheet.
When the latch is the same for booth sections, they will go to conversion in the same time and will output io at the same time.
I think there will be no delay between the channels.
Digital data input format is given in the data-sheet.
Yes, and you can parallel one chip channels in this way, no problems, it works and is practically tested. If using 200 ohm resistors for bohth channels you get output resistance little less than 100 ohms and full scale signal voltage about 200 mV RMS. But reasonable is to add some filtration, at least prallel capacitor to the resistor. This is not enough if you use AD1865 in NOS mode, but at least something.