MPP

inverse RIAA

Here is the inverse RIAA box, two fully differential channels. The four connectors on the left side are the inputs (differential through the cables, single-ended via the connectors), right side is outputs XLR/RCA/BNC all in parallel. Now I want to do a few measurements to see if it is precise.....

Did connect a CD player to the input RCAs and the Paradise to the output, it worked fine, tonal balance appeared to be correct (but it was only a very brief listening). What was interesting though is that the sound character of the Paradise (muscular, well-structured bass, and "forward" character in the mids and highs) was still there...
 

Attachments

  • inverse RIAA box r2.JPG
    inverse RIAA box r2.JPG
    239.4 KB · Views: 471
It is a little quiet here so i decided you should have some fun and look at the first power amp i build in 40 years. Actually my power amp career came to an abrupt halt after i had build my first kit at the age of 13. I had saved the little money i had for several month to build a power amp. I soldered it together and fired it up. With a big bang the electrolytic´s went up in smoke. I had connected them the wrong way around. That frustrated me so much that i decided to take the road to speaker building and making low level stages, like preamps, filters and phono stages.
Since then i have repaired and updated several power amps and have studied numerous schematics, books and threads. So theoretical i know about most of what is going on in power amp design. I even had designed a SUSY amp based on power Opas. I have given that circuit to Brianco but again karma struck, preventing Brianco from finishing it. Still there was this spell on me that one friend of mine had formulated : "You are not allowed to design with more then plus-minus 24V and 1A."
I thought that after 40years in no power amp prison i break free and try something.
Actually what i wanted to build is not a complete power amp but a power buffer.
There are several good designs by Macura, Ciuffoli, Tubecad and others. Mostly they are single ended with N-Channels that work on a constant current source. I chose another circuit that i found one day on the web. I drew it by hand and put it away. Recently i discovered it again in a pile of papers and i must say that i like it. It uses only 3 transistors per channel, One N-Channel, one P-Channel and one BJT driver. I will draw the schematic over the weekend. It is a push-pull Class A amp. I run it on plus-minus 33V with 1,65A idle. That draws around 100W per channel so the amp gets rather hot. I think the Fets can stand that because they are robust but i may reduce the voltage somewhat and-or use bigger cooling finns. In 8 Ohm the amp produces around 20W in Class A. That is enough to drive my MPLs without leaving Class A much. So far i could not check
clipping because my generator produces only 20V. I will do that tomorrow when i have made the voltage amplification section. The buffer is fast with -3dB at 7MHz. It has a small overshot so i filtered the input at 1,5MHZ -3dB. I have not measured or simulated distortion but i will measure it ones in a while. If somebody want to do that, please put it in the simulator. It worked right away and does not produce any nasty on-off transient.
There is a small hum, because PSRR is not that great. It has potential for improvement, so when you have any idea, i am open for suggestions.
Here are some pictures.
Schematic comes later.
 

Attachments

  • MosFet Power Buffer 1.jpg
    MosFet Power Buffer 1.jpg
    134.6 KB · Views: 391
  • Mosfet Power Buffer 2.jpg
    Mosfet Power Buffer 2.jpg
    132 KB · Views: 374
  • Mosfet Power Buffer 3.jpg
    Mosfet Power Buffer 3.jpg
    139.1 KB · Views: 377