Jump to content

Cavalli Audio Liquid Lightning


spritzer
 Share

Recommended Posts

I started the tear down of the amp.  Here is the back of the front panel:

 

813B2SIh.jpg

 

I'm impressed, this is very, very nicely done.  Here is the front panel stripped:

 

hKGPDE8h.jpg

 

Here is a closeup of the sockets.  The drawback of these sockets is how sensitive they are to the metal pins. 

 

WkHxZlih.jpg

 

Here is a picture of the power switch and the paint flaking off:

 

WxXMhvHh.jpg

 

Here is the knob or rather the back of it.  Nice paint spray effect, why this isn't powder coated I don't know.  Far more durable...

 

JWRgFSlh.jpg

 

The sockets, front and back.  Pretty nice design.

 

WH6qdhmh.jpg

 

Taken apart.  Headamp Teflon socket for comparison

 

0XkZQ8zh.jpg

 

...and here is why they are almost impossible to use.  When the Headamp sockets were made extreme care was taken to get the clamping force just right.  The pins just need to be moved a bit to loosen the clamping force.  Almost impossible to get them out of the chassis though... 

 

DAPkJeOh.jpg

 

Bottom of the amp PCB.  Let's play spot the cold solder joints... 

 

9XmViBgh.jpg

 

The PSU board. 

Link to comment
Share on other sites

that really is the way it arrived. and in a few places on the top, hennyo like balls of solder on

the parts. seriously the solder job on this is miserable. also notice NO CONNECTORS.

the whole mess is soldered together board to board from the top. Also notice that getting

any of the power transistors off the heatsinks is virtually impossible. especially the ones

on the triple of small heatsinks. This thing is virtually unrepairable if a part blows.

 

updated schematic with more stuff.

Link to comment
Share on other sites

Yeah... dbel talking about anything Kevin has designed is going to be fair and honest. 

 

Been busy stripping the amp PCB and holy fuck this is cheap.  I'm sure they had to force Imagineering to make something so thin. 

 

The mounting of the transistors is something truly special.  Stay tuned for pictures... 

Link to comment
Share on other sites

Time to share my joy with the amp board.  To start off with, there is no way in hell this could ever be serviced... by anybody.  If something fails in the amp it needs a new amp board as the CCS is utterly impossible to get to and the output devices are little better.  The CCS has to be the most stupid design I've ever come across, 12 small sinks in a row, all of them with the overly long screws mounted one way so if KSA1156 nr. 11 or 12 needs to be replaced all of them have to be removed.  Alternating the screws and using much smaller ones would have made it possible to service this thing.  Anyway, time to crack on...

 

L3n3VhOh.jpg

 

This is how all of the mosfets are mounted.  That red bit is similar to mica pads but plastic and flexible.  Similar to mica these need thermal paste on both sides to conduct heat properly...  :palm:

 

TpOLPrlh.jpg

 

How the transistors are mounted.  Screw,  shoulder washer, transistor, isolating film, heatsink, another shoulder washer, locking washer and the nut.  So the screw is floating but to accomplish that they needed to do this:

 

UbkDzBmh.jpg

 

...drill out the center holes in all the heatsinks.  The middle one is a stock unit out of a new box from Aavid.  What a colossal waste of time and to make matters worse, this just increases the output capacitance of the amp.  That's not something it could afford... 

 

rUEIvFRh.jpg

 

The main culprit are these.  Data sheet can be found here, look for Coss and remember that two of these drive each side of the headphones and the capacitance of the main Stax lineup is 110pf.  The SR-003 is 44pf and it sounds utterly horrid. 

 

TWALOR0h.jpg

 

Here is what the amp board looks like now.  Need to remove a bunch of parts and start building up the new CCS.  I'll also add terminals for the input, redo the whole input section as the soldering is quite bad.  Perhaps I should get some JPS cable for its magical properties?  Either that or the 50+ year old milspec silver wire cable Kevin gave me. 

Edited by spritzer
Link to comment
Share on other sites

Well there is a story about those holes.  When the Exstata was under development and I was still a part of that, I insisted that the board had holes around the heatsinks to suck up air from underneath.  Alex dismissed this but finally gave in when he saw it was a good idea.  Here this is taken to the extreme though as it's compromised the structural integrity of the board. 

 

One funny thing are the ground traces on top of the board.  Almost no other traces are there so why on earth isn't there a solid ground plane?  It lowers the noise and there are no drawbacks.  Doesn't make any sense...

Link to comment
Share on other sites

That is certainly possible.  Sent the pics to two people I know who assemble PCB's and they are appalled.  Not just the soldering but also how utterly impossible any rework is. 

 

I've stripped all unnecessary parts off the amp board so I can start building it up again.  Already installed are terminal blocks for the input , new power LED (hate blue led's) and safety output resistors.  Next up are new BJT's for the output and 2SA1968's for the CCS.  Then the amp will be ready for testing again.

 

After that I have to turn my focus to the PSU.  The bias supply is utterly fucked and not safe to use.  Then I'll add loopouts to the back and rewire the input section. 

Link to comment
Share on other sites

What is the cause of a poor soldering?

a - Worker without skill experience;

b - Thin PCB substrate and thin traces that cannot handle the temperature; or,

c - Soldering station without temperature control?

All of the Above?

 

That is certainly possible.  Sent the pics to two people I know who assemble PCB's and they are appalled.  Not just the soldering but also how utterly impossible any rework is. 

 

I've stripped all unnecessary parts off the amp board so I can start building it up again.  Already installed are terminal blocks for the input , new power LED (hate blue led's) and safety output resistors.  Next up are new BJT's for the output and 2SA1968's for the CCS.  Then the amp will be ready for testing again.

 

After that I have to turn my focus to the PSU.  The bias supply is utterly fucked and not safe to use.  Then I'll add loopouts to the back and rewire the input section. 

 

So we're talking a total rebuild, which would yield more or less a totally different amp?

Link to comment
Share on other sites

Some progress:

 

C5APUHsh.jpg

 

The entire CCS removed and replaced by a single 2SA1968.  All transistors are now 2SC4686A's due to their super low capacitance and general awesomeness.  The FJP2145 would also have been a good choice that is in current production. 

 

So we're talking a total rebuild, which would yield more or less a totally different amp?

 

The circuit is completely identical.  Only thing I've done is remove those bloody mosfets which should never have been there and fix the CCS.  So now the output capacitance is cut down to less than 10pf and the huge bump in distortion from the old CCS is gone. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...