Possible Mods to Audio Alchemy DITB dac

[arupg]

Hello All:

The Audio Alchemy DITB dac uses a AD1860 dac chips, 1 per channel and then a dual OpAmp as a buffer for the final RCA outputs.

I was advised to add an external I/V section using the THS4031/2 and directly terminate to the RCAs without any other circuit/s in between. Am told this would yield in better results than what I'm currently getting using OPA2107 as an OpAmp. I was actually contemplating to using the OPA627 OpAmp on browndogs in the original circuit.

I sincerely am hoping for help from the Forumers.

Thanks in advance!!



Link to the Circuit/Schematic:

img297.imageshack.us/my.php?image=aaditbcircuitld7.jpg

[Deleted]

arupg

Personally, I would remove the existing opamps and fit quality gold flashed , turned pin I.C. sockets. I would either remove the muting transistors, or replace their capacitors with polypropylene or polystyrene types. Same goes for the capacitors at the opamp input pins if ordinary types are used.
Mike will undoubtedly have further suggestions.
I think the choice of opamp has far more influence on SQ than a few Vishay resistors. Better to play with opamps initially, and THEN try fine tuning with the exotic stuff. Get a trusted friend to listen before and after all mods have been done. Have a cup of coffee or something, and relax, before each session. Also , give a replacement I.C. time to "burn in"
SandyK

[leo]

THS4031/2 make excellent current to voltage conversion on a dacs output, you need to be sure the layout is suitable for these op-amps though as they can be picky and suffer stability problems.
Ideally you need to look at the pins on a scope checking for oscillation.
If you get oscillation the op-amps tends to overheat and sound bad.
Nice op-amps though if you can get them to behave

I can't seem to find the datasheet for the AD1860, just had a quick look at the DITB dac and it seems its a voltage out dac not current, I'm not sure if you can bypass the internal I/V in this chip like you can a few of the other AD based dacs

[arupg]

Hi Sandyk:

Thanks for your suggestions. A high Quality socket is already in there. Most caps are MKP of high quality. Fav brand is Vishay 1837 MKP. Power Supply caps are all upgraded to Nichicon Muse FG types & bypassed with 0.1uF Vishay MKPs.

Get a trusted friend to listen before and after all mods have been done.

Thanks for this suggestion but I do honestly trust my ears. I use a reasonably good system - Audio Synthesis ProPassion Passive Pre / Quad 606 / B&W 802s and can easily hear small changes that occur in the system.

[arupg]

Hi leo:

Thanks for your post. The AD1860 is indeed a voltage out dac. I/V is done internally.

Here's the AD1860 chip details:

diyhifi.org/forums/download.php?id=2833&mode=view


I am currently using BB OpA 2107 a DiFET and am wondering if moving to a 627 may yield better results. Any thoughts?

[Deleted]

arupg
Leo's warning about possible overheating due to oscillation also applies to the OPA627. It has a much faster slew rate and wider bandwidth than what you are using now. .If they sound slightly grainy,or feel more than luke warm,they may be oscillating. If you have access to a 100MHZ CRO, it would be a good idea to look at their outputs.

SandyK

[arupg]

Unfortunately, I don't have access to a scope but will have to rely on temp buildup of the opamp and its sound in particular. What I was thinking is whether the OpA circuit is ok or do some resistors etc need adjustment/omission etc.

Detail of such can be seen here:
img169.imageshack.us/my.php?image=daccircuitio5.jpg

Thanks for looking!

[Deleted]

arupg
I would leave passive components as they are. As this schematic is much clearer than the previous, I realise that the "muting" transistor I referred to is actually more likely to be for de-emphasis. As you are already using quality components, leave as is. Using fast devices like the OPA627 on Browndog adaptors may make them more likely to have stability problems, so it may be adviseable to connect a capacitor e.g. 100nF, between pins 4 and 8 of the IC position UNDER the main PCB. Try it without initially.Don't forget to check that DC out from the ICs is less than a couple of mV. This should not be a problem with the OPA627

SandyK

[arupg]

Hi SandyK:
Thank u for all the suggestions. I already have 0.1uF MKP between Pins 4 & 8 so I suppose it should be ok.

Thanks again


----------------

Arup

[arupg]

Someone suggested that I do a passive I/V of the AD1860 dac chips, (bypassing the internal OpA & resistor within the dac chips) and grounding the Iout with a 500 ohms resistor. They claim this would give 1mA that amounts to 500mV of output. Then, take a OpAmp (maybe with a Gain of 5 or more) like the OPA637 and terminate to RCAs.

Will this be better than trying a better buffer OpA?

Please advise.

[Deleted]

arupg
Sounds not only messy to implement, but risky. Are you willing to take the small (?) risk of
destroying a DAC, sourcing a replacement, and installing it ? Perhaps, even double trouble.
Also, have you considered using a Burson discrete opamp in the O/P stage ? Many people report great results with them.
SandyK

[leo]

Yes you can bypass that dac chips internal op-amp and use either passive I/V with an external gain/buffer stage or an external active I/V stage but its not the easiest of things to do.
Only do it if your confident, results can be very good but please be careful as these chips can be delicate and easily destroyed

[arupg]

Sandyk:

I tried a Burson discrete OpA and found it very nice BUT also a bit too sharp, slightly brittle and thinning of the sound. It robbed the musicality.


----------------------------------------------

Leo:

Thanks for giving me the confidence. Unfortunately before I take this route, I need to source a pair of the ancient AD1860 chips - 'just in case things go wrong.....'

[Deleted]

"I tried a Burson discrete OpA and found it very nice BUT also a bit too sharp, slightly brittle and thinning of the sound. It robbed the musicality."

arupg
I haven't personally tried them, but many people rave about them. Again, it proves that there is no single solution that works for all pieces of equipment.
SandyK

[leo]

The Bursons also need good quality regulation and work better on higher than normal voltage, +/-15v is the minimum IMO

[rickcr42]

Missed this during my "absence" but in a utshell the largest single improvement you can make to the DITB is right at the power supply which is the main "choke point" holding back the analog stage from providing a wide scale dynamic output due to the opamps being powered by a piddling +/- 5VDC.By making these +/- 12VDC you will improve the SQ across the board but this means your "power station" is not up to driving the new V-regs so you will need to make one that puts out a well filtered +/-15VDC @ 500mA minimum (1 A even better ) which means the digital section V-Regs will now be asked to "get rid of" the extra voltage and the way V-Regs get rid of volts ?

HEAT !

so heat sink the digital section V-Regs !


DITB Schematic courtesy of Kal Rubinson

www.audioasylum.com/images/DITBschematic.pdf

DITB Mod Guid also courtesy of Kal

www.audioasylum.com/images/DITBMOD.pdf

[rickcr42]

the DITB uses the AD1860/AD1851/AD1861/PCM56 (all pin compatible) DAC chip and yes,has both a voltage output and a current output pin which in the case of the Audio Alchemy DITB means you need to "snip" the pcb trace (see AD1865 data sheet and DITB schema) to separate out the raw IV output which is normally strapped to the voltage output.

DITB Schematic via the Audio Asylum (FAQ Section) and posted by Kal Rubinson:

www.audioasylum.com/images/DITBschematic.pdf

Kal Rubinson suggested mods,also via the Audio Asylum

www.audioasylum.com/images/DITBMOD.pdf



My suggestion is to do the "quick and dirty" Kal Rubinson upgrade mods first then later if you feel you need to go all out in the analog section,it really does make the DAC sound a LOT better for not a lot of loot or work-no small thing in such a cramped chassis where even a simple cap upgrade would be a bitch (luckily AA used good and plentiful bypass caps though you may want to upgrade the aging electrolytics to Panasonic FC's )

More ?

Look at the next level of audio alchemy DAC the DDE3.While an entirely different animal in the digital section (uses a DAC chip for instance many consider to be one of the best ever designed) if you look at the analog section you will see where not only is the more powerful +/-12 VDC supply used but the Op-Amps (single Op Amp Version of the OP275 Dual Op-Amp) are followed by what Audio Alchemy called "Class-A Current Pumps" that if one was able to cob up on a small perf board would maybe be the ideal output stage for the DITB if using the Iout mode because it incorporates a nice active IV conversion stage (both the AD1862 and AD1860 have a +/-1mA output )

Kal again

www.audioasylum.com/images/DDE3.pdf

the "players"

supernovae.in2p3.fr/fp/done/CCDchain/AD1862.pdf

www.lampizator.eu/lampizator/LINKS%20AND%20DOWNLOADS/DATAMINING/AD1860.pdf

www.analog.com/static/imported-files/Data_Sheets/AD1851_1861.pdf

focus.ti.com/lit/ds/symlink/pcm56.pdf

ezphysics.nchu.edu.tw/prophys/ael/File/Datasheet/op176.pdf

www.analog.com/static/imported-files/data_sheets/OP275.pdf

BTW-you knuckleheads really need to be aware of the circuit before advising on upgrades since in the case of the DITB "unmodified" you have limited current and only plus and minus 5 VDC available to power any analog stage "upgrades" and trust me,most Op-Amps sound like pure crap running under such "starved" conditions that low so the selection needs to be on what is an ACTUAL improvement and what is just throwing parts at it.

If I have a beef with most "upgrades" it is that most suggest the same exact parts or part types without ever even considering the actual circuit being modified which means it is not a given but a crap shoot when using such a "shotgun" method to audio circuit upgrading meaning it is NOT always a step forward unless suited to the actual device being modified and that means taking the ENTIRE design into consideration.

So far in THIS design the OP275 beats every Op-Amp I have tried in the "stock/12 volt upgraded) DITB except for the pain in the ass AD825 which MUST have an adapter board (SMD Design,single OA) AND an IC socket for that adapter with any "improvements" to my ears more of a slight sonic shift than a day/night improvement so to be honest more trouble to implement than any "potential" for better SQ and with most other options not only nothing special but again would mean adding an IC socket which to be blunt is best left out of ANY IC based design and is more asking for trouble by its inclusion on the increased stray capacitance,decreased reliability (it IS between the part and the board and IS a connector and we all know connectors ARE the weak link in most designs) and by being there means most would spend more time rolling in opamps than they would listening just because...


I think the above are workable with the only difficulties being there WILL be increased heat buildup (and the DITB already runs really warm) so you must use proper heat sinking (it is tight as a bitch in there so no easy task unless you "skew" the heat sinks) and the external "Power Station" must now be at +/- 18 VDC instead of the former +/- 8 VDC and even better,make it have 1-1.5 A per polarity available and use a "Pi" filter after the rectifier section.

do these things and the DITB holds up very well against much more expensive competition

[clint]

Rick,
is there any difference between the Audio Alchemy DITB and the version 1.1? I'm asking this because i have the V1.1 for many years and i find it just great, even in nowadays standarts.

[rickcr42]

From what I can tell just the opamps are different and yes,the "el cheapo" DITB entry level DAC is a real sleeper that when you compare it to all these "wonder kid" DIY attempts holds up very well indeed (8X OS,AD1860 DA Chip) but this is no accident.open it up and you will see how small and tidy should be done and with the only concessions to cost being the single power supply and the underpowered opamp section (five volts)-with the latter easily rectified with a soldering iron and a high current "power station" of -/+ 15VDC Raw (unregulated) and a change to 12 volt V-Regs feeding the op amps

As it it still beats a damn lot of the "modern" competition so these updates are not by any means an essential

[clint]

Sorry Rick but i'd mention the version 1.1 but i was wrong. The AA version i have is in fact the V 1.0. Anyway tanx for your quick reply and i concur with you the low end responce & detail of these units are in fact amazing.

[rickcr42]

considering what passes for a bargain these days i still marvel at just how much performance vs. cost the AA products of old offered PROVING you don't need to have fancy face plates and "the best" everything cobbed in to have a thing play actual music and no,i will not give up my DITB and no,it has not yet been "bettered" in any meaningful way with redbook CD or 16 Bit USB-to-spdif output computer streaming so until or unless I have no more need for 16 bits,and that isn't happening any time soon and if it does i will adress the need THEN, I don't see any reason to upgrade in my lifetime

[saber7]

Hello all. I am new to this forum but have read all the posts regarding upgrading the AA DITB. I am going to perform the mods in the Kal Rubinson pdf. I am an engineer but new to electronic components and can use some help. Most of the parts listed in the mods pdf are not available anymore so I am trying to put together a list of current parts for this mod. Below I listed all the parts and a link to the part I might use but was hoping to get confirmation that these would be the correct/recommended parts. Also, hopefully when complete, this list can help other people wanting to perform this mod. I have also listed some questions I have at the bottom. Thanks in advance for any help!

------------------------ PARTS / LINKS ---------------------------

AN7905 5V, 1A Neg. Voltage Regulator (U24) for -VD
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM7905CTFS-ND

AN7805 5V, 1A Pos. Voltage Regulator (U18) for +VD
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM7805CT-ND

AN7912 12V, 1A Neg. Voltage Regulator (U23) for -VA
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM7912CTFS-ND

AN7812 12V, 1A Neg. Voltage Regulator (U17) for -VA
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM7812CTFS-ND

330uF, 25V electrolytic capacitors, MUSE, Panasonic Z or HF (C71, C72)
search.digikey.com/scripts/DkSearch/dksus.dll?PName?Name=P10275%2DND&Site=US&Lang=EN

.47uF, Panasonic P-series (PP, 50V) or V-series (SF, 63V) (R24, R27)
search.digikey.com/scripts/DkSearch/dksus.dll?PName?Name=P4544%2DND&Site=US&Lang=EN

OP-275
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=OP275GPZ-ND

1K, 1% metal film resistor for OP-275 (replacing 1.5K)
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=PPC1.00KZCT-ND

14-16 pin heatsinks (33HS016) to mount on DAC chips (U14 and U15)
search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=HS179-ND

------------------------ QUESTIONS ---------------------------

1) Is the 330uF, 25V electrolytic capacitor lead spacing 0.197" (5mm) or 0.138" (3.5mm)?
2) Is the lead spacing of 0.295" (7.50mm) ok for the 0.47uF cap?
3) Would there be any difference in using the polypropylene vs stacked film for 0.47uF? Can't find the PP.
4) I noticed the voltage regulator model numbers all match except the 7805, its missing the FS from CTFS-ND. Does this matter?
5) Does the OP-275 need to be surface or through mount?
6) Is the OP-275 link I posted the correct op-amp that everyone talks about sounding so good? Is there other manufactures that make a better one?
7) What wattage does the 1K OP-275 resistors need to be? Is 0.6W like the link I provided ok?
8) I have no idea what size heat sink to get. I cannot get the 33HS016. I googled it and found 1 in stock at one place. Does the one posted seem like it would fit?
9) Does Thermal Tape Adhesive or Heat-conductive epoxy work better?