Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › New Developments for me (DP)
New Posts  All Forums:Forum Nav:

New Developments for me (DP)

post #1 of 9
Thread Starter 
Well, What is the worst that can happen? It must be close to this.
1% of 302K, Waiting for reply! On an ancient MAC nonetheless. My good old laptop is is the shop and I just couldn't stand it anymore. Fired up the old POS! Is that a model number? Like the FORD POS from "Men In black"

Anyway, the laptop will be back soon, maybe tonight. Yes, I knew what was wrong, but it required endless dis-assembly. God, I hate these big keys.
I finaly built my 627- Buf 634 amp and it smokes. Talk about current and power, Why would anyone need three buffers at full +- 15V power? One kicks the old Grados so much that they rattle endlessly. These old Grados are just lazy it seems after being used only to test the low current amps I've built lately.
I built this one on breadboard so I could change components for fine tuning etc.
But try this with the AKG 501, A match made in Heaven! Maybe some more buffers for this phone, but others like the ER4 are in danger zone.
I built this while waiting, waiting, waiting for parts from PPL for two other boards.
Anyway, the other development is with several mods of the Sony FM/AM digital portable. These mods really make a big difference. They used too many caps. Caps everywhere killing the sound, plus I am trying to find room for a line out to bypass the low powered opamp for headphone drive. A few mods make this little portable very much improved. I will continue this later as I can't stand this keyboard.
Dan
post #2 of 9
wait...I don't get it. you do know what POS stands for, right?

I like the sound of that amp, though! I take it this beats the hansen board amp you were using before?
post #3 of 9
Dan,

You mentioned that you built the amp in several topologies (buffer inside and outside a feedback loop).

Which did you settle on as the best sounding/performing?

I built the AD823/EL2001 amp with buffers inside and outside the feedback loop. I'm still tweaking the amps, but subjectively, a multi-loop topology sounds best to me.

I still want to setup an A/B test for the amps and make a few quantitative measurements.

I'll loan you my Mac Plus with an astounding 4 MB of memory.
post #4 of 9
Thread Starter 
POS= (Piece Of S___!) It's just a nickname. Actually I like Macintosh but this one is ancient. Not qoute as bad as the Mac Plus though. Somehow I believe those old Macs like the classic and original 512 will become collector's items some day. I don't know why I believe this, but I do.

Anyway, The new amp is SO MUCH BETTER than anything I've heard that I am at a loss to figure out how to accomplish this with a battery amp. The 627 and Buf 634 is a great match no matter the loop type you use. I will also try the 634 with a 132 and that could be battery powered. First I tried just the 627, then added some extra power supply caps, then added the 634 and some more caps. The gain is set at around 4.5 10K divided by 2.26K. A gain of ten was way too much.
The power supply is an older Burr Brown Model 550 plastic encased +-15 at 25ma which is not a lot of power but is very adequate with extra capacitors. The chips are bypassed with WIMA .022 at the pins followed by two Nichicon MUSE 470uf as well as two DME 10uf metal films. Not a lot of capacitors but good ones. and I plan to add some 1uf WIMA at the supply pins when I make a board for the final amp.
One thing to note is the stability of the 627 even on the plug-in breadboard.
The amp is very smooth with much less distortion in fast transients than the dual 9V Hansen built to the extreme. It makes the 501 come alive as the battery amps just can't do and also sounds incredible with the ER4-S.
Obviously there are some expensive parts here. The 627 is the most expensive, but well worth every penny. I have not tried the 132 yet as I will next. I just don't have any single 132's except in the SOIC size which is impossible to breadboard without some fancy solder work.
This amp changes everything.
Dan
post #5 of 9
Daniel,

Did you build the amp per the schematic in the tech article from Texas Instrments ("Boosting Opamp Current with Buffers" or something like that)?

ok,
erix
post #6 of 9
Thread Starter 
No I drew my own schematic from PPL's suggestions, and from the data from Burr Brown on the buf 634. Just don't use a gain of 21, it's too much. Adding a buffer to an opamp is a simple thing, but the supply must be right to keep enough juice for the buffers.
Dan
post #7 of 9
Yey I am back On Line Again after upgrading to Windows XP my Old and i might add rather Great Creative labs Modem Blaster Flash 56 would not work. After about 100 Hours of Hassels i got another Modem that was on Microsofts List and what do ya Know it works.

Anyway Regarding the OPA-627 + BUF634 on batteries. Three 9 volts will power these Chips just fine. Two 9 volts will work ok But Three will sound Better. At least it dose wih the MDR-7506/MDR-V6. Operate the BUF-634 with a Resistor between The bandwidth Pin and the Neg supply. Do use a multiloop Topology so as to get a Constant amount of Negitive Feedback Throughout the Audio range. In most cases a constant amount of feedback is possible up to 200KHz. with a 16 MHZ. Opamp like the OPA-627. The main advantage of the Multiloop feedback is alowing this to be possible with opamps. Normaly only a Discreet component design allowed the Designer to Control the Openloop gain and hence the Openloop bandwidth. I have tried some alternate ways of reducing an opamps openloop Gain in the past and thay are a real big hassel to impliment. One method is to put matching resistors from one offset Adj. pin and the Output and a matching resistor from the remaining offset pin to ground. This works with alot of opamps but some it will not, it depends where the Offset pins are connected within the opamp. On opamps that the technique works on the DC offset is quite large and has to be trimed out by intentionaly unbalancing the resistors connected to the offset pins. this is because the Resistor's connected to the Offset pins actualy reduces the real openloop gain of the OPamp. Most Opamps use the large openloop gain to controll the DC offset. With walt's multiloop Method The Full real gain is still available but is Localy Reduced via the Local feedback loop around the Opamp. I have found thhat the DC offset using Multiloop is the same as using a Conventional 100% feedback method as discribed in the BUF-634 Data sheet.
post #8 of 9
Daniel,
Any news yet on the new schematic to drive the 501s?
I'd like to try a new project. I am currently using a CMOY with 501s and I'd like to take it to another level.
nick
post #9 of 9

how much volts ?

Hello,

I also aggree that the BUF634 and OPA627 will make a superp headamp. At the moment I have a pcb that I made my self that would have had two single opamps in double output current arrangment.

I use apper47 design and modified it by changing the second single opamp with a BUF634 and removing the two 47 ohm resistors.

The end result is a OPA27 and BUF634 per channel with a single 12 V 7 Ah lead acid battery that is rail splitted with a BUF634.

I am not using a volume pot because I test it with a PC sound card and use the volume control on PC.

To me even with approx 6.4 V per supply (+V and -V) there is more that enough power to drive headphone with low impedance with a gain of only 6.67.

Does anyone aggree?

I plan to try the 627 and 637 in both double output current and buf arrangement soon.

Gavin
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › New Developments for me (DP)