Audio-gd Reference 1 DAC (56 K warning)
Jul 5, 2010 at 5:43 PM Post #2,221 of 2,441
I've spent the last couple of days thoroughly enjoying what I'm hearing from the freshly done RE1. I think it's still too early to definitively comment on the changes since burn in will likely alter the sound scape further but what has changed in the last couple of days is truly noteworthy.
 
In a week's time I will outline these changes in detail. Right now there are substantial improvements in the mid range/upper mids (presence region) and the highs. The bass changed the least but that region also shows improvement (control, body, texture).
 
I'm pleased by the results to say the least.
 
Peete.
 
Jul 6, 2010 at 2:00 PM Post #2,222 of 2,441
That's pretty cool Peete! I hope you can keep on making gains in terms of sound quality without coloring the sound too much.
 
Jul 6, 2010 at 6:10 PM Post #2,223 of 2,441


Quote:
Interesting mod, could you elaborate on what you did (and the parts used) please. I hope you get your R channel problem sorted ASAP. I know it sucks not having the DAC in the signal chain. I've not listened to my ref rig since I pulled the RE1 ( for the MAX mod) last week. I'm itching to get it back into the system.
 
Peete.



Hello to everyone,
 
My mods are:
- Legenburg OCC copper wire (red ones) to replace the silver plated on RCA and XLR outputs.
- 2 x 4700µF BC component 056 serie on the digital board to replace the 2 nover 3300 µF.
- 3 x 2200 µF pana FC to replace each nover 3300 µF.
- WBT RCA connectors on the digital and analogs boards.
- SUN V2 discrete OPA to replace the Burr Brow ones under the CAST modules, with 330 µF pana FM caps (on V+ and V-) to feed them instead of the 68 µF Oscon ones.
- CANARE BCJ-JR BNC connector (no solder at each end)
- CANARE L-2.5CFB and BELDEN 1855-A 75 ohms cables inside.
 
That's all.
 
What I have noticed, not during elaborated tests, is that sound stamp (?) is more realistic.
Low frequencies are better sculptured. More refinement and nuances too.
Not the improvement of the century, but quite audible.
Now my right channel is OK, I continue with the burn in.
Sorry for the shorts impressions. I'm not a good reviewer.
 
Djilbe.
 
Jul 6, 2010 at 9:34 PM Post #2,224 of 2,441


Quote:
That's pretty cool Peete! I hope you can keep on making gains in terms of sound quality without coloring the sound too much.
 

 
The overall voicing remains neutral with no emphasis in one region vs another. The changes are of clarity, low level cues and details and freq extension PG. The upper mids and the highs are better revealed which has led to an even greater level of resolution (and air). I purposely steered clear of coloration (as much as possible) without changing the overall character of the RE1. The Mundorf wire is the only spot where actual signal pathway influence could directly assert itself. The caps in the RE1 are not in the signal pathways...all I did was improve the various PSU components (which was my goal). Cheers PG
beerchug.gif

 
 
Quote:
Hello to everyone,
 
My mods are:
- Legenburg OCC copper wire (red ones) to replace the silver plated on RCA and XLR outputs.
- 2 x 4700µF BC component 056 serie on the digital board to replace the 2 nover 3300 µF.
- 3 x 2200 µF pana FC to replace each nover 3300 µF.
- WBT RCA connectors on the digital and analogs boards.
- SUN V2 discrete OPA to replace the Burr Brow ones under the CAST modules, with 330 µF pana FM caps (on V+ and V-) to feed them instead of the 68 µF Oscon ones.
- CANARE BCJ-JR BNC connector (no solder at each end)
- CANARE L-2.5CFB and BELDEN 1855-A 75 ohms cables inside.
 
That's all.
 
What I have noticed, not during elaborated tests, is that sound stamp (?) is more realistic.
Low frequencies are better sculptured. More refinement and nuances too.
Not the improvement of the century, but quite audible.
Now my right channel is OK, I continue with the burn in.
Sorry for the shorts impressions. I'm not a good reviewer.
 
Djilbe.


Sounds good and thanks for the parts listing....I'm curious, why did you swap out the BB IC OPA's ? They are only used (IIRC) to control DC offset (servo duty) and keep that offset at 0. Maybe I'm wrong about that.
 
How do you like the Legenberg wire ? I've used it for the ground runs in my ACSS and DC line cables for the Phoenix but opted for Mundorf silver/gold for the RE1/Phoenix internal hookup and the + - runs in the aforementioned cables. Your observations mirror my own WRT RE1 changes but the degree of these improvements (your mod vs mine) is impossible to judge. It would be cool to compare them side by side.
normal_smile .gif

 
Good work Djibe ! Report back every once in while with your impressions. I think since the BG and Mundorf caps are very similar in construction the burn in period for my mod may actually be closer to 700 hours than the usual 4-500. I can wait but hopefully this number is closer to 500 for my electricity bill's sake. Oh what was the issue that had your R channel MIA ? Good to hear you have it fixed.
 
Peete.
 
Jul 7, 2010 at 2:24 AM Post #2,225 of 2,441
Peete is correct, the opamps only serve to keep the DC at 0.  There's no point in replacing them.
 
Jul 7, 2010 at 4:09 AM Post #2,226 of 2,441
From engineering point of view there is no point changing the dc servo op amp because of its purpose to keep dc at zero. But, u can change it and hear the differences by using different op amp to further tweak the sound. It all depends on how good is the dc servo circuit. The better the dc servo circuit, the less the differences that you can hear. Anyway, this is based on my experienced. YMMV. However, Kingwa don't recommend to use the discrete op amp for servo circuit. I've tried the SUN V2 on a dc servo circuit b4, and it can't keep the Dc at 0. Maybe it is a different case on Djilbe's dac.
 
 
Cheers!
 
 
Jul 7, 2010 at 5:27 AM Post #2,227 of 2,441
i'm thinking of upgrading the entire middle board of my dac9mk3 to the dsp1 board. currawong and other guys who have replaced their ref 1 middle board before, could you guys tell me how difficult the replacement is?
 
i'm wondering if it's possible, after upgrading, to use the old board as the PSU for a modded hiface, hmmm
 
Jul 7, 2010 at 7:52 AM Post #2,228 of 2,441
It's tedious, but not difficult.  Other than the digital connections you have to remove from the back and sides, under the plate over the transformer there are the solder points for the power wires to remove from that cover plate.  Other than those, the really PITA point is unscrewing the FETS from the bottom of the case.  That is hell as they are screwed on very tight.  I did it by putting the machine on its side. The board change took me 4 hours overall, but I go slowly when doing something for the first time.
 
Jul 7, 2010 at 8:55 PM Post #2,229 of 2,441
noinimod
 
Curra is right...taking the nuts off the FET bolts is a PITA but the method I devised helped considerably with this tedious job. With the unit on it's side (after you have unsoldered the back panel connections) I took masking tape and laid a strip of it across the bolt recesses. I then laid the DAC back down on it's feet and got out my socket set. A 3/8ths socket (I think) on an extender (5 inch) makes for easy loosening of the nuts. Sometimes the bolt will spin if the tape isn't making good contact with the bolt head. In that case simply put the DAC  on it's side once again locate the correct bolt and hold it in place with a phillips screwdriver while you rotate the socket to loosen the corresponding nut...once it's loose you can lift off the nuts one by one with needle nose pliers...along with the washers. The bolts remain in place because of the masking tape which makes reassembly a lot easier once you are done with the modifications. Just reverse the process when the board is ready to drop back in.
 
Oh ...the PSU connections (wiring from the bridge rec pcb above the traffo) should be unsoldered before you pull the board (before removing the pcb standoff bolts). Make note of what wires goe to where and mark them accordingly. If I forgot anything I'll add to this post as I think of them....
 
It's pretty straight forward to do but time consuming. I must have spent more time in disassembly and then reassembly then I did pulling caps and replacing them. Make sure you clean any flux or stray solder off the pcb's...flux will eat the pcb and through hole material. Rubbing alcohol and an old clean toothbrush work great for this job.
 
Peete.
 
 
PS:You might be able to use the old board as a PSU for the modded Hi Face but I would suggest just building a linear PSU of a smaller size instead. Ask Kingwa what can be done with it. What do you want from it, regulated +5V line and 3.3V DC or just 5V DC ?
 
Jul 8, 2010 at 6:13 PM Post #2,230 of 2,441

Quote:
 
1 - ....I'm curious, why did you swap out the BB IC OPA's ? They are only used (IIRC) to control DC offset (servo duty) and keep that offset at 0.
 
2 - How do you like the Legenberg wire ?
 
3 - I think since the BG and Mundorf caps are very similar in construction the burn in period for my mod may actually be closer to 700 hours than the usual 4-500. I can wait but hopefully this number is closer to 500 for my electricity bill's sake.
 
4 - Oh what was the issue that had your R channel MIA ? Good to hear you have it fixed.
 
Peete.



Hello !
 
1 - You all guys are right. Kingwa told me the same thing. OPA in this position (DC servo) are not important.
But I had the SUN V2 in hands and I'm a curious guy.
So I wondered to myself "why not to try them". The only risk is to loose 10 minutes.
And you know what ?....... I will never remove them.
Subjectively, the benefits are more resolution and more tensed (?) bass.
I would like to put the REF1 on a mesures bench before and after each tweak, but I don't have one. So I only talk about subjectives sensations.
The only thing I would recommend to REF1 owners is to try the SUN v2 and make your own opinion.
 
2 - I have choosen the legenburg wire because of is geometry,which is supposed to have advantages again round wire. But I have not notice any significant change. But I will continue to use it for the good conductivity of pure copper.
 
3 - No burn in for me now, as the replacement of the caps have given to me what I wanted. Details are easier to ear (it's was not bad before) but without being fatiguing. And my god, what an impact
eek.gif

On a disk I know well (Hadouk trio - air hadouk: http://www.amazon.fr/Air-Hadouk-Trio/dp/B003162RFQ/ref=sr_1_1?ie=UTF8&s=music&qid=1278625970&sr=1-1) I was surprised when a blow of percussion drum came.
I don't know if a streaming listening session will give to you an idea of what I want to say.
 
4 - I have redone the solder points of the new caps and the sound came back after I replug the dac two or three times. So I suppose it was the problem, even if they was no sound after the first replug.
 
Djilbe
 
Jul 8, 2010 at 7:56 PM Post #2,231 of 2,441
Thanks Djibe,
 
I can't be bothered to replace the BB OPA's...but I do understand exactly what you mean by dynamic speed and weight. My setup is lightning fast and very dynamic. It was before but now that aspect has been improved in terms of control and resolution level even further. other changes have been noted already but I will hold off on the final assessment until enough time has passed.
 
On another note.....
 
I've committed a silly mistake the past few days...I was listening last evening and my ears were telling me one thing while the brain was telling me something was off but my ears must be full of crap. I decided it was my imagination until I double checked to see if the digital room correction processor was functioning as it should ( in the dual sub pathway and switched into the line). It had been put in bypass mode...how I don't know, I guess I had bumped it when trouble shooting my tube amp (which is still out of action). Feel kind of dumb for not double checking it earlier this week...I usually look at it at least once a week to make sure it's on and doing it's job...anyway long story short I took it out of bypass mode and slapped in a ref disc...WOW what a difference. The bass slam, extension, texture, body...everything is jaw dropping good. Now I feel better about the 100hz and under region. Lesson here ? Even the most experienced audiophile should double check things (gear settings,xover settings, proper function/routing etc) from time to time if your ears are telling you something is slightly wrong...I had chalked it up to needing burn in (which it does need BTW) but obviously a critical part of the system had been bypassed. I really do appreciate that processor (even though it was a major PITA to setup, nothing auto about this unit) and the result of being bypassed really drove home how important it is to have the room interaction with the speakers brought under as much control as possible ( or for as much as one can safely afford to do). A little bit of room treatment makes or breaks the result in my room that is for sure.
 
Cold solder joints can be tricky to diagnose which is why any joint I have my doubts on I immediately reflow or redo until it's perfect. Trouble shooting intermittent faults is the most difficult task of all, best to make absolutely sure the solder joints are correct and sound. I use my continuity meter to double check anything I have any lingering doubts about. It might take more time but it's well worth that time IMO to rule out areas if the unit has issues at power up.
 
Peete.
 
Jul 9, 2010 at 10:36 PM Post #2,232 of 2,441
can someone recommend me a XLR cable stock or DIY with good quality. I am expecting my blackcat velose coming monday.
 
Jul 10, 2010 at 8:27 AM Post #2,233 of 2,441
http://www.amazon.com/XLR-male-female-Microphone-Cable/dp/B000165DSM/ref=sr_1_fkmr0_1?ie=UTF8&qid=1278763751&sr=8-1-fkmr0
 
Two of these and shorten to the length you need, and you'll have lots of spare wire :).
 
Aug 11, 2010 at 2:02 AM Post #2,235 of 2,441
I finally finished off all of the additional tweaks originally envisioned for the MAX mod WRT to the RE1. All 9 Black Gate caps ( 8 x 100uf 16V and 1 x 10uf 50V) are now bypassed with Roderstein .01uf 160V MKP-1837 and WIMA .1uf 63V film caps. The orientation and location of the caps are on purpose since clearance issues with parts underneath necessitated some trial and error fitting before going ahead with the final layout. The outboard trio of Black Gates (not in the pic but on the main pcb adjacent to the DSP! board) will be bypassed by MKP-1837's once my order for the CD7 MAX mod arrives from PCX (along with the tighter tolerance clock from Audio-gd) The trip to the cottage (with the Ref system in tow) was very good but these mods are still continuing to show improvement even after 600 hours...I would guess the BG caps should be nearing maturity (within the nest 100 hours or so) while the fresh MKP bypass caps installed today might take 50-100 hours before settling in. Anyway the results are spectacular to say the least (this past 10 days at the cottage and hours after finishing the DSP1 bypass mods). I must comment on the Mundorf M'Lytic's, they are truly outstanding main filter caps.......I'm very impressed with the cumulative overall improvements these mods have produced overall but I think a good part of the improvement could be attributed to the Mundorf caps themselves.I will be using the M'Lytics from here on on with futrure mods that require large value snap in types. Sure they cost a little more than the already good Nichicon KG Gold Tune but they cost considerably less $$ and are better IMO than the highly regarded Black Gate Power Tank WK's  It's not surprising, (to me at least) since the formulation Mundorf used for the L'ytic line up is an improved recipe of the venerable Black Gates. Anyway that is only IMO of course YMMV...  Some short SQ comments ....My system via speakers (and cans) sounds. well..nothing short of amazing. The wife this evening was very impressed with a track she knows inside and out and said she'd never heard it sound this good. The 3D imaging is spooky as is the detail retrieval, dynamics are scary fast with zero bass bloat, overhang etc. I'm a happy camper. A couple of further tweaks around the edges could be performed if I can find film caps (PIO) that will fit underneath the DSP1 board to replace the .1uf 630V Solen MKP's but only if I have some that fit (I'm not going to go with anything expensive here, if the .1uf 160V K42 Russian caps fit these will do the job nicely).
 
Here are the pics....
 

 
6 x Blue Box caps are the MKP-1837's and the Lone red cap in the BLH corner is 1 of the 3 WIMA .1uf 63V caps
 
Here is the underside pcb shot. Note the heat shrink tubing (on the leads to insulate them) and slight mod of the lead spacing on the one pair of MKP's Each pair of MKP's required a little different installation method (as is plainly seen). Follow this orientation and you'll have no problems.
 

 
Final pic.......
 

 
I would have to say that if you swap out the stock NoVer caps on the DSP1 module for something better (like Nichicon KZ MUSE or Black Gate) it is pretty much essential to bypass those (all 9) with HQ film caps. I really wanted to use some T-1 Teflons (820 pF 200V) I have on hand for this job but the space restrictions of the DSP1 (and it's relationship to the main board pcb) made that choice impossible. That being said the Roderstein MKP-1837's are something special in their own right and are well suited for just this use ( the WIMA's are pretty good too but definitely not my first choice).
 
Peete.
 

Users who are viewing this thread

Back
Top