MFB544 modification from /00 to /50 questions

Alles over onze geliefde MFB boxen in deze categorie.

Moderator: Moderators

Plaats reactie
pserci
Berichten: 37
Lid geworden op: 02 aug 2016, 09:17

31 dec 2018, 17:55

Hi all and happy new year!!
In these holidays, i've completed upgrade and transformation of my MFB544 set from version /00 to version /50.
MFB are equipped with ad0211/sq and ad0163 on the medium/high section of the speaker. All service bulletins and driver change advisories was applied on both box. Also have modified the subsonic filter, adding a pair of 47 nF caps in parallel to C514 and 515 and put a 1nF for C561 on the feedback loop of mid-high amp as suggested in many thread of this forum.
So, from a first listening, speakers are greatly improved in terms of quality and soundstage, but it seems that are a "hole in the mid band, because voices appears a little backward, especially female voices, and high tones seems a little increased, causing a sort of unnatural sound to my ears..

One question that i have is why Philips Engineering have shifted crossover frequencies from 500/4K of the /00 version to 650/3,5k of the /50 version, after all they have reduced reproduction band for the midrange. Could it be the reason for the different acoustical timbre of my modified 544 set?

Next step will be measure RTA with white noise, when i find some time to arrange my setup (Behringer ECM8000 and DEQ2496) to verify if frequency response is correct or in some way altered from linearity.
Any help or members experiences are appreciated...

Thanks in advance

Paolo
Salvador.
Berichten: 142
Lid geworden op: 24 jan 2017, 17:42

31 dec 2018, 23:11

Hi Paolo,

You need to change the following parts also:

C552 to 15nF
C553 to 3,3nF
C568 to 3,3uF
R706 to 240 Ohm

Salvador
pserci
Berichten: 37
Lid geworden op: 02 aug 2016, 09:17

01 jan 2019, 10:52

Hi Salvador, thanks for your reply and i wish a happy 2019!! :D
Sure, i've changed also these components, according to service bulletin regarding ad0163 introduction on 544.
I also have let c566 on the Mid section with a 3,3 uF (same value as c568). Could this be the reason? In an approximately analysis, a lesser value of c566 extend operating area of the Mid in the high side....
Have to try replacing c566 with a 4,7uF cap...

Regards, Paolo
Salvador.
Berichten: 142
Lid geworden op: 24 jan 2017, 17:42

01 jan 2019, 14:35

Hi Paolo,

Thank you very much, you also a happy new year 2019!!

Your thoughts about C566 are right, C566 has to be 4,7uF. Only C568 has to be 3,3uF.

It will make a huge difference to change all the resistors in the midhigh signalpath, from R600 till R708, to metalfilm resistors (I use 0,6 Watt types). I change also the foil capacitors to Wima MKP10, only the two capacitors rm15 150nF of C514 and C515 will stay (I place on both of them 47nF Wima MKP10 in parallel). I change the capacitors of the passive filter, C566 and C568 to JB capacitors axial MKP type. It makes a huge difference and costs not that much (you can buy most of the parts at TME.eu). I also change Ts421 and Ts422 to BC559B/BC560B, Ts441 to BC559A/BC560A.

Best regards,
Salvador
Oane
Berichten: 26
Lid geworden op: 09 dec 2018, 16:40
Locatie: Damwoude

01 jan 2019, 18:09

Is er echt zoveel verschil tussen de drivers van de /00 en de /50 serie?
STR-DA5800ES
pserci
Berichten: 37
Lid geworden op: 02 aug 2016, 09:17

01 jan 2019, 18:20

Oane schreef: Is er echt zoveel verschil tussen de drivers van de /00 en de /50 serie?
Yes, there are...and some other intersting little corrections through the evolution of the MFB544 project in electronics. I've also excluded the tone control present below /50 revision that in some way cause sound to be a little close on highest frequencies, even with the pot at minimum.

Anyway, main problem in my 544 set was that i've made a mistake mounting the midranges, reversing the phase. Now that drivers are correctly mounted (in fact AD0211 have to be connected in reverse respect to the tweeter) sound are greatly improved!! Only a little fastidious "s" prominence have to be corrected...have to measure with RTA in open field...more will follow..

Regards to all
Paolo
Plaats reactie