musical-fidelity-a3.24 mods

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have done some mods to the MF A3.24 which some of you with some electronic experience may wish to try. These mods greatly improve the transparency and bass tightness of the unit. Which truly makes it one of the best dac's around. The output stage of this dac is awesome, fully discrete and capable of driving a speaker direct, up to 6wrms!!!!!!!

1/ Direct coupling to the outside world, instead of going through a very average coupling electro. I tried a Blackgate instead of what they used, it was better but nowhere as good as direct coupling (removing the cap altogether). But you CANNOT just remove the cap as you'll get large amounts of DC offset at the output, and you will blow you pre or power amp.

This is how it's done

A/ Remove and short or just short out the output caps 470uf/63v C101 and C201.

B/ Find the 2X 2.2kohms resistors R108 and R208 (around 3inches from the back of the board) remove these resistors and replace them with 2x multiturn 5k cermet mini pots set at 2.2k to start with (Spectrol are good and avalible at most good electronic shops) You only need to use the centre and one of the outside legs the spare outside leg can be left in the air or attached to the center leg.

C/ put a digital multimeter on the output and earth and set the sensitivity to dcmv 200mv, powerup the unit and let it warm up and adjust the mulit turn pots for zero (0dcmv) offset. you can get it to stay at zero it will drift +1mv -1mv.

2/ For the very keen ones this will make it sound even better again.

A/ Find the i/o opamps 2x NE5532 after the dac and replace them with 4x AD825's these are smd singles and have to be mounted on 2x Brown dog boards. 2x ad825's for each Browndog board. They now replace the 2x NE5532's This jobs takes some patiences and care as the NE 5532's are through the hole plated on both sides.

Now you have a dac that is truly awesome to listen to.

Note: these mods will definatly void your warrenty

If you do a search i have posted the diagrams of both the circuit and power supply somewhere in this forum, should you need them.

Cheers George
 
I've been asked by a number of members by private email to post the circuits again as they cannot find them in previous posts that I've put them on, so here's the links. I cannot email them as they are A3 and I do'nt have a scanner that big. This is as sharp as I can make them. All the components are marked on the circuit board so you should'nt need the diagram to do the mods.

Front stage:
http://www.diyaudio.com/forums/attachment.php?s=&postid=585827&stamp=1109562245

Rear stage:
http://www.diyaudio.com/forums/attachment.php?s=&postid=584404&stamp=1109372290


Power supply: minus transformer
http://www.diyaudio.com/forums/attachment.php?s=&postid=585884&stamp=1109570483


Cheers George
 
Thanks George. The diagrams are very interesting. I used to own a A3 24 myself.

I was looking at the power supply pic and some of the wordings are not clear. Can you tell me what are the values for the following?:

i. Choke at CH1
ii. Capacitors at C110, C112, C210, C212


Thanks
 
I finished mods on MF A3.24 CD Player two weeks ago, so couple of hints might be helpful…

Both analog and digital board are done very nicely, i liked the ground topology a lot, but I wasn't impressed with the ribbon cables connecting two boards, nor with the length of the same type membrane cable connecting mechanism and digital PCB. Even worse, parts quality is very bad. Using a lot of 0.01uF disk-ceramics for the local decoupling brought-up the highs in this CD Player - but also completely removed the natural sound character. NE5532 as current to voltage just can not perform this duty properly. Buffer - maybe, but rather not.

I have installed a lot of Black Gates, used 2134, 275 and 826. The end configuration: first two IC's 2134, second two IC's 826. Silver ribbons were used as well, and 4.7uF V-Caps for sound coupling.
Both oscillators got their separate low noise power supply. The output pulse transformer was replaced.


The end result was very, very good.

Regards,
Extreme_Boky
 
Hi Georgehifi, I have just registered and also cannot send you a PM ^^. I'd like to ask a few questions about your Musical Fidelity A324 DAC mods.

1. How did the AD844 compare to the OPA627 or AD825 ? (I have some OPA627 that I can use and I don't fancy soldering expensive chips SMD). Will a PDIP-N package of the AD844 fit the Brown-Dog adapters to convert 2x single op-amp to dual? Is it worth doing the mod to use the OPA627?

2. Are you able to recommend a part number for the mini-pots?

Something like a Vishay Spectrol 064W502 (Digi-Key part number SP064W-100K-ND ) or a 63M type?

Thanks a lot for any advice you can give. I wasn't sure where to post to get your attention.
 
Not the correct forum for this so this is all she wrote, this was a while ago but the AD825's were used in place of the 4 x NE5532's for their great sound (for an opamp) and because of it's fet input it had close to zero dc offset in this circuit, just a little tweak with the instaled 5k Spectrol timpots in place of the 2.2k resistors, and it was adjusted down to .1mV offset at the rca outputs. This was important now that the whole shebang was now dc coupled from the dual dif outputs of the dac through the opamp I/V stage and then through the discrete output stage, and it could drive 6watts, pretty impressive for a dac, it could almost drive a speaker direct.

Cheers George
 

Attachments

  • brother scan2.gif
    brother scan2.gif
    77.5 KB · Views: 1,513
rogerlike said:
Hi Georgehifi, I have just registered and also cannot send you a PM ^^. I'd like to ask a few questions about your Musical Fidelity A324 DAC mods.

1. How did the AD844 compare to the OPA627 or AD825 ? (I have some OPA627 that I can use and I don't fancy soldering expensive chips SMD). Will a PDIP-N package of the AD844 fit the Brown-Dog adapters to convert 2x single op-amp to dual? Is it worth doing the mod to use the OPA627?

2. Are you able to recommend a part number for the mini-pots?

Something like a Vishay Spectrol 064W502 (Digi-Key part number SP064W-100K-ND ) or a 63M type?

Thanks a lot for any advice you can give. I wasn't sure where to post to get your attention.

3. Are the stock NE5532 op-amp AP, N or P type?
 
Thanks George. I know the OPA627 has FET inputs too. I think i'm sorted now with the mini-pots. I had read that you were going to try the AD844 current-feedback type and I'm really interested to know how well you think they worked. Would it possibly for you to perhaps email me so we can discuss it? Thanks :)
 
AP stands for all purpose, that's all.
Do not direct couple with the NE5532 in, as it has way too much dc offset to be trimmed out with the trimmer, only a fet input opamp (AD825) can be used in this position direct coupled as the trimmer will only have to be adjusted slightly off the original fixed resistor value of 2.2k.
If the NE5532 is in, the trimmer will be way off the 2.2k mark to zero out the dc offset and distortion will happen and maybe stress on surrounding components.

Cheers George
 
PierreG said:
Could we use LM4562 in place of the NE5532?

Couldn't say without seeing what happens to it on the cro, I have moved on from the A3.24 for some time now, and I would be speculating on your questions.
It's possible that the LM4562 would work if it is fet input as fet input opamps to my experience usually have low stable dc offset were as bipolar input opamps usually have too much dc offset for dc coupling the output, especially in this case.
Then there's the chance something may be oscillating all this needs to be viewed on the cro. I know the AD825 is fine just stick with it. It's sweet, dynamic, transparent, detailed and has excellent bass slam, and it cooks heavily into class A for an opamp. It's the best neutral balanced opamp I've heard.

Cheers George
 
PierreG said:
Could we use LM4562 in place of the NE5532?

The LM4562 is an especially fast op-amp, and as such I believe you would need to add some capacitors across the power-rails for each op-amp, which the A324 design doesn't include.

Regarding the NE5532A, I'm pretty sure the A is some kind of upgraded version of the NE5532, although the TI data-sheet indicated no difference? <shrug>. I understood the P suffix to indicate PDIP (Plastic DIP) package type.

I have completed the mods using AD825AR mounted on the little dual-SO8-single-DIP8 BrownDog boards and it works a treat.

To remove the original NE5532AP, I found it necessary to cut the legs so I could de-solder the legs one by one. That was probably the most tricky part of the upgrade.

Direct coupling with the NE5532 in place brought about a noticeable improvement to bass and detail but, the treble became more prominent and perhaps a little harsh. Certainly very up-front anyway. Putting in the AD825 fixed up the tonal balance very nicely, back to more like what the A324 originally sounded like, but with big improvements in bass, detail and transparency* All the usual suspects that people try to explain with words! Certainly worthwhile. I am very pleased.

*Disclaimer: Of course this is only with my ears/equipment/music/room etc. and the things that were most obvious to my after listening for about 5 hours. As always, YMMV!

Another note: My DAC board is marked revision 2.

Georgehifi, the schematic scans are pretty hard to read. By any chance, could you write out the part numbers for chips IC9 and IC10 please? Thanks.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.