Stacking BUF634Us in Mint
May 29, 2003 at 10:15 AM Post #2 of 14
it should have the same basic effect, as it is the same as stacking the usual buffers, you're just parallelling the output buffers... it would however not be overly easy with SMD devices...

g
 
May 29, 2003 at 11:22 AM Post #3 of 14
hi all... just wonder is there any theory behind stacking the buffer? as far as i know, buf634 is more than sufficient to drive one (or two
tongue.gif
) cans... can anyone clear me up about this? Thanks!
 
May 29, 2003 at 4:46 PM Post #4 of 14
Stacking buffers improves things for two main reasons.

First, each buffer has less work to do, so distortion goes down. See KurtW's buffer distortion posts here in the archives, and also the benchmark tests on my PPA site.

Second, the output impedance inside the feedback loop goes down. If your buffers have an 8 ohm output impedance, stacking 4 of them brings the output impedance down to 2 ohms. This is one of the reasons we think the PPA prototypes sound so good, incidentally: the HA-5002 has a 0 ohm output impedance. We're still paralleling them, though, for the distortion reason, and also for safety reasons.

A feedback-controlled amp can be thought of as having 0 ohm output impedance, since the output voltage doesn't drop as the load goes up. (To a point, anyway.) But, if you've ever experimented with putting a resistor inside the feedback loop at the output (e.g. R5 in the CMoy pocket amp design) you know how it can affect the sound of the amp. What is the theory behind this fact? I don't know for certain, but try it and see if you don't find it to be true.
 
May 30, 2003 at 2:01 PM Post #6 of 14
Eh?

I read the spec, it says ha-5002-5 have an output resistance of 3 to 10 ohm?

Or mine is an outdated spec?
 
May 30, 2003 at 4:08 PM Post #7 of 14
Sorry, I was confused. I was thinking of its lack of an output protection circuit. Where I was going with the thought is that this chip can destroy itself with very low output voltages, if current is allowed to go high enough. Adding more output resistance would fix this, but it would destroy this buffer's fine qualities.
 
Jun 3, 2003 at 3:30 AM Post #9 of 14
Hehehe... I finally got 2 buf634s thats i stacked on the exhisting ones on the Mint. Amazing! I love the Mint even more now. Just wonderful, controlled, deep bass. It sure was hard to solder them on, thats for sure. It makes soldering the smd opamps/buffers on the Mint board itself seem like nothing.
 
Jun 4, 2003 at 5:18 AM Post #13 of 14
Quote:

I finally got 2 buf634s thats i stacked on the exhisting ones on the Mint.


Most impressive, WIL.

Instead of pictures, what I'd be most interested to know are the basic techniques you used. For example, did you bend the leads straight on the second set of buffers? Did you flux the pins? Did you just solder wildly then use desoldering wick to clean out the gaps between the pins? That sort of thing.
 
Jun 4, 2003 at 6:44 AM Post #14 of 14
Hi guys. Directions:
1. straighten out legs and make them flat - use pliers to do so
2. put them in a 90 degree angle
3. pretin the legs of the 2nd buffer and the 1st buffer.(1st already on Mint board)
4. place the 2nd buffer on top of the 1st and attempt to connect the two legs with the solder. I suggest using your finger to hold down the 2nd buffer on the 1st one securely.
*Use desoldering braid if the solder produces bridges between legs.
*Make sure the iron is pretinned also.
*If any leg is crooked, simply use a knife or something small to straighten it out. Makes life easier.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top