|
Tweakers' Asylum Tweaks for systems, rooms and Do It Yourself (DIY) help. FAQ. |
|
|
|
Tweakers' Asylum Tweaks for systems, rooms and Do It Yourself (DIY) help. FAQ. |
|
In Reply to: Marantz cd-94 posted by bcherry on January 14, 2002 at 18:18:37:
It is quite a classic, much like a little CD12/DA12. So don't
change too much. If you search the archives you would find something
like:Posted by Thorsten (i) on August 01, 2001 at 03:08:16
Hi all,
"Hot town, summer in the city...."In fact, it's so hot I have trouble sleeping. So I decided that a long tweaking session on the LHH-1000 DAC (same as the legendary Marantz DA-12 but wearing a Philips Badge) was on order. This DAC is amazing. It uses a masisve cast chassis, all copperplated, has three major torroidal transformers (encapsulated) for the Powersupplies, uses my favourite TDA1541 DAC, has transformercoupled balanced and normal unbalanced outputs and is just great. In stock condition (12 Years old) it sounded already pretty good, great potential I felt. Hence I decided only on a basic "conservationist" Modification package, replacing old capacitors, fitting better op-amps and a few other little tricks.
I already had most of the stuff for modding, the only things I did not have was the small Emporers ransom worth of Elna Silmic or Black Gate capacitors to replace all the 12 Year old Elna Cerafines ('lytic caps age and unlike wine not well).
So I needed to go inside to get a list of capacitors to order and I felt I might as well do one or two mod sessions. As some have asked, here is what I did....
I started rebuilding the Digital supplies (+/-12V, -6V D, +5V A, +5VD), including replacement of all electrolytic caps with brandnew, usually higher value units. I used what was left over from other projects, including Nichicon PL, Panasonic FC and Shamwa "Audio Grade". All these are low Z, low ESL, low ESR types, mostly rated long live at 105 Degrees.
I also replaced all rectifier Diodes with Telefunken soft & fastwitching ones (I bough a silly load of them when they where for sale for a penny each or so) for the 12V and 5V A supplies and used Motorola MUR410 for the high current +5V Digital supply. All reservoir capacitors also received bypass capacitor combos, Foil & FIlm Polypropylen//Polystyren//Mica - whatever I had around in the junk box of odds and ends.
2) I rebuild the analog stage for the unbalanced output. I changed the I/V converter op-amp to a Burr Brown OPA637, the lowpass filter op-amp to a Burr Brown OPA627 and the output buffer to a Burr Brown BUF634, all kindly supplied FOC by TI/BB.
In order to make the OPA637 stable with a 1k5 Feedback resistor and a 2n4 feedback capacitor I changed the RC combo hanging from the +15V rail to the DAC's output, used to inject an offset current to zero the op-amp output. I used a 300 Ohm resistor to the DAC, a 220uF Elna Silmic Capacitor as filter and 7k5 Resistor to +15V, giving pretty accuratly +2mA offset current to compensate for the 2mA quiescent output current of the TDA1541 at "digital silence". I also fited a 10nF Polystyren Cap from the OPA637 negative input to ground.
This way the op-amp operates under stable conditions and still allows us to use superior speed (100V/uS slewrate instead of 40V/uS for OPA627) of the OPA637 where it is most needed. I also placed a 100pF Silver Mica cap directly from the output to the inverting input with minimal loop length. This serves to offer a low impedance noisepath above frequencies where the 2n4 Polystyren integrator capaictor becomes inductive. As the OPA637 still has gain at around 80Mhz this is not an unneccesary precaution.
The OPA637 was biased into class A with a 2SK61 J-Fet operated as constant current source connected between output and negative rail. I found two that gave 7mA, a bit high but if Class A the go DEEP class A. The lowpass filters OPA627 was a straight swap, again a 2SK61 CCS was used to bias this unit, this time I had two 6mA units.
The BUF 634 was a little trickier to implement. I selected it for wide bandwidth mode (180MHz - short pin 1 to pin 4), this can get silly if the PSU decoupling is not attended to. I also removed all the components connected to where the negative input on an op-amp would be. A 1uF Siemens stacked film capacitor was connected direcly between pin 4 and pin 7 with minimum leadlength, this has allways guranteed stability for me without resorting to nasty sounding ceramics. The Siemens stacked film cap's have a residual inductance of around 6nH, this is usually less than the op-amp pins, frame etc together.
I also replaced all signal path resistors with my favourite metal films and bypassed the main PSU electrolytics with a Film & Foil Polypropylen//Polystyren//Silver Mica combo. The input damping resistor on the BUF634 was chosen to be 51 Ohm, the same was used for the output resistor. As the back to back output coupling capacitors are rated only 6.3V I could not simply bias them with a resistor to the negative rail, so I only bypassed them using (again) a PPFF//PS//SM combo, this time of the best parts I had at hand.
After that I had to listen what this had done - the sound had opened up a lot, more detail, transparency and depth to the soundscape. Instrument placement was well outside the stereobasis, very close to goosebump territory. The warm tone and good bass I had liked was retained. Great so far. Now for optimising the DAC.
For a start I replaced the small leaded ceramic bypass cap's near the DAC with 10uF Sanyo Os-Con's. On the -15V supply (analog) I further used a 10nF WIMA FKC cap in parallel with a 15nF Polystyren and a 5nF Silver Mica. On the +/-5V Supplies I used my classic 0805 SMD decoupling combo of 10nF X7R and 100nF Z/Y5U. These where tricky to fit so that current loops where minimised without disturbing current loops, but I got it. Then I stared on the fourteen decoupling cap's on the DAC. Original they where 100nF "Blue", no name, unknown, but looking like metalised mylar.
In the MSB and and nearest to MSB pin (12,13,18,19) I replaced these using Sanyo Os-Con's (10uF/25V - posiive side to ground!!!). This gives improved bass impact, dynamic and prat. On the LSB pins and the three nearest to them (21,22,23,24,7,8,9,10) I used WIMA 10nF FKC bypasses with a further Silver Mica bypass in the 100pF range across the originals cap's.
Back into the system - same music again (by now it's 4am), wow, this is now mighty close to what I remembered from my TDA 1541 Non Oversampling DAC. There is still the slight edge that calls for a Taddeo Digital Antidote (I copied his RLC Circuit from the patent - great stuff) or the removal of oversampling, but otherwise this is amazing. And I have not yet cleaned out all the old Cerafines. The sound has scale, imapct and killer dynamics, loads of space and detail and so on. Very musical and very audiophile at the same time.
Okay, now to order the Emporers ransom worth of BG's and Silmics and do a bit of work on the Digital section, also internal rewiring. I'll do it some time real soon, really. Well, real soon after I have finished listening through some new CD's, building a 45 and a 300B amp and finishing some other projects anyway....
Ciao T
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Marantz cd-94 - Werner 01/15/0222:47:46 01/15/02 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
