Melodious MX-U8 USB Interface 8 Core XMOS chip
Jun 2, 2015 at 1:15 PM Post #31 of 658
Hi Ginetto,
You asked this question before (previous page of this thread,,,,), and I uploaded a schematic (which is from another device) to explain to you how Melodious configured psu.
Again:
 

 
You can see the two diode-bridges each fed by one seperate winding.   As you can see there's only ONE positive signal, NOT TWO.   

 
Hi Alex ! thank you sincerely again.  I told you i am slow to catch up ? i was wrong ... i am slower 
redface.gif

Thanks a lot for your patience and extremely precious help.
But sorry ... do you mean that all the four big blue caps in the picture are in parallel ? with the same Voltage at their terminals ?  (around 7x1.4 = 10 V ?)  
strange arrangement indeed. 
 

 
So, there's only ONE MAIN regulator, LT1963, and several (8 or so) ADP150 regulators to create "separate" power stages.  

if i understand well all the caps are in parallel feeding the main regulator, and after this main other let's say secondary e local regulator are on the pcb ?
I was completely wrong ... when i see two diode bridges i think of a dual power supply or two separate supply channels ... so also the bridges are in parallel ?
strange design indeed.
 
 Obviously you really want to experiment with your lab psu, or another psu.   

redface.gif
 yes .. that would be the idea.  Of course i could try a kit for a very low noise single supply but the lab psu could be enough as a first try.
 
The easiest way is, like i told you before, disconnect your tranny and solder +wire from your external psu to one of the 4 +legs of the big caps,  and -wire from your external psu to one of the 4 -legs of the big caps.
In that case you will feed the LT1963 with your external psu, and also take use of the local psu buffer caps (the 4 2200uF-25V BC's)
To be 100% sure, just measure if all the -legs are connected to eachother, and if all the +legs are connected to eachother.
Hope this clears up a few things.  
Cheers 
beerchug.gif

Alex

 
Thanks a lot again Alex !  yes now it is more clear.  
I did not catch that all 4 big caps were in parallel (in your schematic i saw just two of them).
Now i understand that everything is in parallel, diode bridges and caps ... wov ... a lot of buffer uF for a single regulator indeed.
Given that the secondary are 7V i guess that there will be about 10V (i.e. 7x1.4)  at caps terminals.  
So i should provide 10 clean Volts to one of caps
I have to take out the bridges maybe ? because one - leg of one caps seems connected to the + leg of another ?
I see written on the pcb  after the transformer  " AC 6-10V " ... so i guess that i could increase to around 14 VDC max.    
I have a desoldering gun and i could try to remove all the blue caps and see the traces below the caps
Thanks a lot again.  Your advice has been fundamental.
Kindest regards,  gino  
 
Jun 2, 2015 at 1:51 PM Post #32 of 658
   
Hi Alex ! thank you sincerely again.  I told you i am slow to catch up ? i was wrong ... i am slower 
redface.gif

Thanks a lot for your patience and extremely precious help.
But sorry ... do you mean that all the four big blue caps in the picture are in parallel ? with the same Voltage at their terminals ?  (around 7x1.4 = 10 V ?)  
strange arrangement indeed. 
 

 
if i understand well all the caps are in parallel feeding the main regulator, and after this main other let's say secondary e local regulator are on the pcb ?
I was completely wrong ... when i see two diode bridges i think of a dual power supply or two separate supply channels ... so also the bridges are in parallel ?
strange design indeed.
 
redface.gif
 yes .. that would be the idea.  Of course i could try a kit for a very low noise single supply but the lab psu could be enough as a first try.
 
 
Thanks a lot again Alex !  yes now it is more clear.  
I did not catch that all 4 big caps were in parallel (in your schematic i saw just two of them).
Now i understand that everything is in parallel, diode bridges and caps ... wov ... a lot of buffer uF for a single regulator indeed.
Given that the secondary are 7V i guess that there will be about 10V (i.e. 7x1.4)  at caps terminals.  
So i should provide 10 clean Volts to one of caps
I have to take out the bridges maybe ? because one - leg of one caps seems connected to the + leg of another ?
I see written on the pcb  after the transformer  " AC 6-10V " ... so i guess that i could increase to around 14 VDC max.    
I have a desoldering gun and i could try to remove all the blue caps and see the traces below the caps
Thanks a lot again.  Your advice has been fundamental.
Kindest regards,  gino  

 
I just measured, just to be sure you won't do anything which you would regret in the end,,,,,,,,
 
Voltage on caps is 12V in my case. All 4 caps are in parallel.
Stay away from 14V, 10V DC is good.
 
In my case Melodious changed transformer to better quality Noratel, but, they put in 15VA 2x9Volts, which is NOT as good as 15VA 2x6V or 2x7V,
simply because of the fact that if VA stays the same, and voltage increases, amperage decreases, thus a higher resistance which is always worse.
 
Just solder psu wires directly on one of the caps, and remove connector transformer.
 
8-10V DC preferable (the higher you go, the hotter your LT1963 gets,,,), 12V DC max.
 
 
Cheers,
 
Alex
 
Jun 2, 2015 at 2:13 PM Post #33 of 658
   I just measured, just to be sure you won't do anything which you would regret in the end,,,,,,,, 

 
Thank you Alex !
 
Voltage on caps is 12V in my case. All 4 caps are in parallel.   Stay away from 14V, 10V DC is good.
In my case Melodious changed transformer to better quality Noratel, but, they put in 15VA 2x9Volts, which is NOT as good as 15VA 2x6V or 2x7V,
simply because of the fact that if VA stays the same, and voltage increases, amperage decreases, thus a higher resistance which is always worse.
Just solder psu wires directly on one of the caps, and remove connector transformer.
8-10V DC preferable (the higher you go, the hotter your LT1963 gets,,,), 12V DC max.
Cheers,
Alex 

Thanks a lot again.  Now i can proceed with more confidence. I will not go over 10V then. No more. Do not want to stress the regulator. 
I think that in the end i will sacrifice one cap (desolder it) to use the holes to solder the wires. I have bought a desoldering pump so i have to use it now 
redface.gif

I will leave the pcb on a wooden board and listen ... if the sound will be convincing i think i will replace the blue caps with something same diameter and taller.  Like Nichicon ... good parts i mean. 
Increasing to 3300 uF the capacitance of each. There is space for taller caps. 25V type are quite compact.   
I am curious.  The stock unit was already quite ok.   
Thanks a lot again.  Regards, gino  
 
Jun 2, 2015 at 2:41 PM Post #34 of 658
I would NOT remove caps, just solder on the legs on the solderside, very easy, and wires from psu just run under pcb, no problem at all.
 
About size caps:
 
You have 30mm MAX in height.
 
 
 
Now, lets be smart and choose the Nichicon FG caps which are the prefered version, just read the U12 thread, FG series are the prefered series, better than HW series.
 
And, if we are REALLY smart, we use 16V caps instead of 25V caps! They are smaller in size, which means that in FG series you could mount 2200uF!!!
 
 
In HW series you even could mount 5600uF, but I would try the FG series.
 
 
I havent the slightest idea why Gustard and Melodious using 25V caps in their psu designs, since voltage never raises above 12V when using 9V Transformer.
And, if using the preferable voltage (6-7 V~) you would stay nicely around 9Volts DC,
 
 
Just my 2 cents,
 
Regards,
Alex
 
Jun 2, 2015 at 3:04 PM Post #35 of 658
I would NOT remove caps, just solder on the legs on the solderside, very easy, and wires from psu just run under pcb, no problem at all.
About size caps:  You have 30mm MAX in height.
Now, lets be smart and choose the Nichicon FG caps which are the prefered version, just read the U12 thread, FG series are the prefered series, better than HW series.
And, if we are REALLY smart, we use 16V caps instead of 25V caps! They are smaller in size, which means that in FG series you could mount 2200uF!!!
In HW series you even could mount 5600uF, but I would try the FG series.  

 
Thanks Alex !  bought the FG 2200uF/16V now on ebay ! thanks !
.. i hope they are not fake ... from my country .Italy 
rolleyes.gif

this is the datasheet attached by the seller ...  http://www.tubes.it/nichicon_FG.pdf
i will use the legs of one cap to solder the wires feeding the dc.  
 
I havent the slightest idea why Gustard and Melodious using 25V caps in their psu designs, since voltage never raises above 12V when using 9V Transformer.
And, if using the preferable voltage (6-7 V~) you would stay nicely around 9Volts DC,
Just my 2 cents,
Regards,
Alex

i think that they do not want have problems.  25V are more robust than 16V i think.   Like oversizing. 
I will set 9VDC as you recommend.  
Do you have any idea of the consumption of this unit ? more or less ?  1A ?
 
now what about adding another regulation stage inside ... i mean providing let's say 15-16VDC with the regulated lab power supply doing the 1st cleaning job
another regulation stage from 16 to 9VDC could give any additional benefit ?  
in this way i would have 3 regulation stages.  
Zero noise ?
there is a lot of space free if i take out the transformer. 
 
On the 3 quartzs there is a transparent adhesive tape ... do i have to leave it or i can remove it ? do the quartzs get got ? could the hot melt the tape ?
I will post some pics of the pcb with the new caps as soon as they will arrive.
Thanks again, gino  
 
Jun 2, 2015 at 4:00 PM Post #36 of 658
That are the good ones, 2200uF-16V 16x25mm


That much oversizing in voltage is useless, unless, and that disturbs me the most, they would use a 12V~ or even higher transformer in case they wouldn't had the proper version at hand, and, it's an awfully thought, but, if that's the case, wouldn't they also be willing to sacrifice in SQ and use other components (like smd parts) with the wrong values if they weren't at hand, so they could produce anyway?

Maybe i'm wrong, but that's what often happens with cheap audio equipment from china.

It's sad to say, but I had a few chinese diy dacs at hand which had versions with lots of different component values. Finally I got those working because of the presence of audio forums like head-fi.



The yellow tape, you can remove it, don't think it will help to absorb vibrations. But, why would you want to remove it?

1A at 230V?? Or??

Why would you regulate an extra time? Each regulation provides you with artifacts like noise, they just will add,,,,, like your lab psu,,,,,,

Cheers,

Alex
 
Jun 3, 2015 at 2:06 AM Post #37 of 658
 
Quote:
That are the good ones, 2200uF-16V 16x25mm

 
Good morning Alex !
thanks again for the very helpful suggestion.   Do you think that small value bypass film caps could be beneficial with the FGs ?
they should clean up some HF noise, the origin of all problems with digital ?
 
That much oversizing in voltage is useless, unless, and that disturbs me the most, they would use a 12V~ or even higher transformer in case they wouldn't had the proper version at hand,
and, it's an awfully thought, but, if that's the case, wouldn't they also be willing to sacrifice in SQ and use other components (like smd parts) with the wrong values if they weren't at hand, so they could produce anyway?
Maybe i'm wrong, but that's what often happens with cheap audio equipment from china.
It's sad to say, but I had a few chinese diy dacs at hand which had versions with lots of different component values.
Finally I got those working because of the presence of audio forums like head-fi.

 
i said that because i have seen someting similar on some Mc Intosh amps. But PS caps in power amps are much more stressed, i think, that in this little unit.
This units look to me successive generations of prototypes.  Many versions one after another.
But if this chips need low voltage to work i also do not understand why go up to much with the output voltage in the transformer ... it does not make sense.
Instead as also cheap different type transformers are available i would experiment with them ... like the Hammond type one.  Just a small shield can block there higher EMI emission compared to toroids.
 
http://sigma.octopart.com/37503537/image/Hammond-229C16.jpg
 
as i said i see these transformers used in the spectacular products by Berkeley Audio .... this immediately makes them approved by me ... without even thinking.
 
Anyway --- i dont even know which is the actual power consumption of this unit.   If it draws 0.1 A max, for instance, maybe even batteries become an option ?
The very nice thing with batteries is that there is no connection completely with the electrical grid.   This is very good.
I understand that PS can be made with even lower noise than batteries ... but still batteries are something very intriguing in their concept.
 
The yellow tape, you can remove it, don't think it will help to absorb vibrations. But, why would you want to remove it?

 
Ok. No problem. I will leave it.  I was just worried of possible problems with heat that could melt it down. Now i understand it is a non issue at all.
 
1A at 230V?? Or??

 
no. I wonder which is the current draw of the pcb circuit at 10VDC .... how many A in order to calculate the power consumption of the all circuit.
I guess it is not that much. But how much ?
 
Why would you regulate an extra time? Each regulation provides you with artifacts like noise, they just will add,,,,, like your lab psu,,,,,,
Cheers,
Alex

 
Ok. I think i see your point.
For instance i like much better the PS in the Tanly converter. Same transformer but i can spot a choke and caps on the mains input and those fantastic diodes you mentioned at the transformer output.
I am quite sure a similar PS could be extremely beneficial also with the Melodious.
Is it difficult to do something similar ?
That would really makes the Melodious pretty definitive.   At least in my situation.
However i am also very confident that the superior Nichicon FGs (thanks again for the very precious suggestion) will lower noise even more.
What about adding small value bypass film caps ? something like 0.1uF ?  and in case, which ones ?
 
Nice PS in the Tanly ... and it should not cost that much to built also ?
Thanks a lot again Alex !
Have a nice day !
Regards,  gino
 
P.S.  Alex ... i bought another transformer ...
redface.gif
 ... PCB mount ... 40VA/2x9V (i did not find 7V ... there was one 2x6V but i thought it was too low V)
So in the end i will change the toroid, keep the diode bridges (they look poor to me ... ) and replace the PS caps with the FGs and stop.
The new x-former seems of the good type with primary and secondaries separated (this should avoid the need of chokes at the input)
The FGs i am sure will be a decisive step-up. 
Great brand Nichicon ... i read better than Panasonic.  Another good one seems to be Siemens.
I will put some pictures of what comes out. I will have to drill some holes to fix the x-former to the chassis ... not big deal.
 
Jun 3, 2015 at 4:48 AM Post #38 of 658
Hi Gino,
 
 
It is always good to bypass psu caps with small caps, I use Rifa 0.1uF (MKP), they work very well.
 
The Tanly has Cree diodes in psu, they are very good, probably they need bypassing cap 0.01uF.
I ordered a bunch of them to test with.
 
The Tanly also has a choke with caps in psu, thats definitely a positive thing to have!
 
Only thing i would want to have is a transformer with earth-shield between the two secondary outputs,
or, even better, two seperate transformers, but those wouln't fit in enclosure.
 
I have no idea what the current draw is of the MX-U8, but I suppose it's less than 1A at 10V DC
 
 
Regards,
Alex
 
Jun 3, 2015 at 5:08 AM Post #39 of 658
Hi Gino,
It is always good to bypass psu caps with small caps, I use Rifa 0.1uF (MKP), they work very well.

Hi Alex, i am buying some to use.  I see many times this caps bypassing ... i guess it could be beneficial for HF ripple suppression.
The Tanly has Cree diodes in psu, they are very good, probably they need bypassing cap 0.01uF.
I ordered a bunch of them to test with.

 
Do you know maybe if there are integrated bridges made with these special diodes ?
http://chinaimportexport.wikispaces.com/file/view/Bridge-Rectifiers-KBPC5A-QL-SQL-.jpg/33608079/Bridge-Rectifiers-KBPC5A-QL-SQL-.jpg
(pic is only to explain what i am referring at)
I would be willing to try them in place of the existing ones that look quite poor.
The Tanly also has a choke with caps in psu, thats definitely a positive thing to have!  Only thing i would want to have is a transformer with earth-shield between the two secondary outputs,  or, even better, two seperate transformers, but those wouln't fit in enclosure.

Yes. The Tanly usb looks more carefully designed even using the same identical transformer than the Melodious.
This is the last i bought ... i have some idea on how to mount it to the chassis (i think i will have to lifet the lid ...it is quite tall ... and big ... and powerful.
 
http://www.ebay.com/itm/180858536621?_trksid=p2057872.m2749.l2649&var=480098140751&ssPageName=STRK%3AMEBIDX%3AIT
 
http://images.mcmanager.co.uk/images/1382537526985.jpg
 
No datasheet attached.   But i guess it should have separation bwteen primary and secondaries ... my obsession.
A very interesting interview to audio designer  John Curl  ...
 
http://www.parasound.com/pdfs/JCinterview.pdf
Q: What can you tell us about the transformers? Are they very important?
JC: Absolutely, but when we talk about transformers we have to separate  power amps and preamps. While they are not perfect, Toroidal transformers are the logical choice for power amplifiers because they are very efficient, they tend to have a fairly low hum field, and they’re readily available in large power ratings.
For preamplifiers and other line-level components, the old type EI transformers or what’s called a D-core or split C- core transformer is actually better than a toroid. First of all, they tend to be more compact, and second, and perhaps more importantly, they have very low capacitance between the windings.
This can be a problem when Toroids are used in low signal level applications; the windings are on top of each other so they talk to each other. It used to not be so bad but today the AC power is so dirty.

I have no idea what the current draw is of the MX-U8, but I suppose it's less than 1A at 10V DC
Regards,
Alex

 
Ok .. no problem.  The one i bought now it is a 40VA thing ... a beast for this application.
If only i could find some integrated bridge made with those fantastic diodes you mention ... that would be very nice.
I guess the project will take me two weeks to receive parts.
I am optimistic at the point that i could even buy a very nice and bigger case to put everything in.
This will be my corner stone of my audio systems for the years to come ... i am sure of this.
Thanks again for you very kind and exceptionally valuable support.
Kind regards,  gino 
 
Jun 3, 2015 at 5:44 AM Post #40 of 658
Hi Gino,
 
There are no rectifier-bridges with those Cree diodes, you have to build them yourself, and no, it is not an easy job to accomplish
in XM-U8 because there's not enough room to do so,,, But I will try at least,,,,
 
I still prefer R-Core transformers, and off course, the copper shielded ones with electrostatic shield between primary and secondary,
they aren't that expensive nowadays, but be aware, there are many cheap versions which are not of high quality.
 
Cheers,
 
 
Alex
 
Jun 3, 2015 at 6:07 AM Post #41 of 658
Originally Posted by abartels /img/forum/go_quote.gif
Hi Gino,
There are no rectifier-bridges with those Cree diodes, you have to build them yourself, and no, it is not an easy job to accomplish in XM-U8 because there's not enough room to do so,,, But I will try at least,,,,

Hi and thanks and i see.  Neither bridges with Schottky diodes ? i understand they are a step-up from normal diodes as well.
These stock ones are miserable ... really ...  Bought these to try out ...
http://www.ebay.com/itm/100Pcs-1N5819-Low-Drop-Power-Schottky-Rectifier-Diode-25A-40V-Package-DO-41-/391070733361?pt=LH_DefaultDomain_0&hash=item5b0da1d431
 
to be arranged this way i understand ...
 
http://4.bp.blogspot.com/-vzZlhdJqiN8/Tx0QvrJPvoI/AAAAAAAAAw8/385bvTfrwcc/s1600/BRIDGE.png
 
i even remember to have read that using single diodes to make the bridges is better than to use integrated one (more handy by the way)
 
maybe just one bridge would be more than enough instead of two. I do not like parallel in general.  I think just using one secondary of the transformer and one bridge made out of these diodes should be enough ( maybe i could even use just two caps ... i do not want to exaggerate with PS).  The circuit total consumption will be less than 10VA i guess.  My new x-former will have 20VA per secondary.
I do not know what to do with the secondary not used ... can i leave it open ? no damage to the transformer ?
I still prefer R-Core transformers, and off course, the copper shielded ones with electrostatic shield between primary and secondary, they aren't that expensive nowadays, but be aware, there are many cheap versions which are not of high quality.
Cheers,
Alex

Yes.  I saw them used in very high level dacs and they are very nice.  But as you say poor quality ones are everywhere ... sometimes i wonder how items can be so cheap.
There must be something poor in their execution.  Quality always comes at a price ... an hard lesson i have learnt.
Thanks a lot again,  gino
 
Jun 3, 2015 at 8:08 AM Post #42 of 658
Hi Gino,
 
If you really want to start modding your psu section, do it the right way the first time.
Pcb's don't like it when components will be removed several times,,,,,
 
We are talking about a multilayer, which means there are most likely through hole metallisation within every whole for through hole components.
If desoldering too hot, or too long, it's most likely you will remove the through hole metallisation which will end in one or more "defective" layers. 
confused_face.gif

 
So, please be careful. 
wink_face.gif

 
I would suggest you try the best diodes available, those ones used in Tanly, http://www.digikey.nl/product-detail/en/0/C3D02060F-ND
They are a little bigger, so you have to work harder to get them in 
blink.gif

 
It's up to you if you want to replace only one bridge rectifier, but if I were you I would do it the right way and leave original schematics intact.
You don't like parallel in general??? What are you talking about? I can assure you Melodious, and Gustard, and many others use this concept
because it has some advantages.
 
And, removing caps, and using a lower capacitance, I wouldn't prefer that either.
 
 
Note:
 
Keep it as original as possible, replace components with better ones, and leave the "altering schematics" to the people who studied electronics.
Off course you can do it the way you think it's best, but to be honest, I don't think you should.
 
So, using both secondary outputs of your transformer, replacing both rectifier bridges, replacing 4 caps, that should do the trick.
Maybe shield transformer with copper shield for best results.
 
Regards,
Alex
 
Jun 3, 2015 at 8:30 AM Post #43 of 658
Hi Gino,
If you really want to start modding your psu section, do it the right way the first time.
Pcb's don't like it when components will be removed several times,,,,,
We are talking about a multilayer, which means there are most likely through hole metallisation within every whole for through hole components.
If desoldering too hot, or too long, it's most likely you will remove the through hole metallisation which will end in one or more "defective" layers. 
confused_face.gif

 
So, please be careful. 
wink_face.gif

 
Hi Alex ! thanks again for the advice .. i have already noticed this thing.  I have to be careful indeed. i could ruin very easily the tiny contacts.
I have this desoldering pump but it has a chinese plug ... and i do not have an adapter to schuko.  I have still to change the plug ... this weekend.
I would suggest you try the best diodes available, those ones used in Tanly, http://www.digikey.nl/product-detail/en/0/C3D02060F-ND
They are a little bigger, so you have to work harder to get them in
blink.gif

I trust you completely that these are indeed the very best parts for this application but it is much more tricky to work with them
With the other i have just to twist their legs and it is done.  And i am more than sure that they are much better than the stock ones.  Much better indeed.
Not like yours i agree .. .but i just need a bit of improvement.
 
It's up to you if you want to replace only one bridge rectifier, but if I were you I would do it the right way and leave original schematics intact.
You don't like parallel in general??? What are you talking about? I can assure you Melodious, and Gustard, and many others use this concept
because it has some advantages.

the idea is to use only one bridge and one secondary of the 2 of the new transformer, a 40VA unit
20VA should be more than enough to power effectively the board.
I do not know if leaving one of the secondaries unused and open will damage the transformer. 
I do not think so but i would like to be sure anyway.
 
Quote:
And, removing caps, and using a lower capacitance, I wouldn't prefer that either.

 
i could leave all the caps but maybe 4400uF of capacitance in the PS should be enough ... it is a lot indeed.  And of FG quality.
 
Quote:
Note: Keep it as original as possible, replace components with better ones, and leave the "altering schematics" to the people who studied electronics.
Off course you can do it the way you think it's best, but to be honest, I don't think you should.
So, using both secondary outputs of your transformer, replacing both rectifier bridges, replacing 4 caps, that should do the trick.
Maybe shield transformer with copper shield for best results.
Regards,
Alex

 
The diodes you propose are much more difficult to mount ...
when i see the Tanly PS  i have immediately the feeling of a better PS and that i would like to see in the Melodious as well.
It seems to me that they design and build spectacular boards but then they skip on power supplies, that are also very important for the overall outcome.
Maybe there is an agreement with manufacturers of much more expensive units do build good cheap things ... but not too good to put in danger the market of the boutique highend ?
biggrin.gif

I am sure this Melodious with the Tanly power supply, that would mean in terms of cost some more USD not more, would be my definitive converter.
But the Tanly imported here with customs and shipping is more than 3 times the Melodious.
Better to play a little with the PS in the Melodious.
If i will damage it i will buy another ... and i will not touch it
wink.gif

Ok ... maybe only the FGs ...
rolleyes.gif

Sorry but your diodes are too difficult for me to work with and this is sad because i am completely sure that are very very good
like i trust all your opinions and advice, let's be clear and thanks for giving me directions
but i am an extremely low skilled guy
redface.gif

However ...  Why they do not provide integrated bridges is beyond my ability to understand
Bridges are so handy to use ... they are just fantastic. There are even marks helping to wire them correctly ... what more one can ask ?
Really i do not understand
To end i have really to be more careful with the board ... i have seen that is delicate
Why it must be so delicate i do not know ...
Thanks again a lot,   gino
 
Jun 4, 2015 at 4:08 AM Post #44 of 658
Hi just to say that i have decided to pull the trigger and i am buying things on ebay to build and extremely basic not regulated DC supply.
And leaving the regulation task to the regulator on board.
The idea is just to put together better parts that the ones in the Melodious box.
I have bought:
  1. a better transformer,
  2. better diodes and
  3. good Sprague caps ...
and that will be all.
 
Then the unregulated DC voltage (about 12,6V because the transformer is 9V)  will be sent to the Melodious board
In this way the mod will be completely reversible and i will be able to compare the two solutions: external and internal PS.
 
However I have a big concern about the regulator on board.
I wonder if it will get too hot with 12.6V at the input. This is my main concern.
What do you think ? will 12.6V be too much for it ?
 
Unfortunately i cannot find 7v or 8V unit on ebay.   They go from 6 to 9 usually.
If you have any suggestion of any supplier  ...
it is a midterm project.
Thanks a lot again.
have a nice day
Gino
 

Users who are viewing this thread

Back
Top