Hifisonix kx-Amplifier

Hi, I bought ksa1220 and ksc2690 on ebay. I don't know if they are fake.
I measured hfe:
20 pcs PNP between 140 and 180 hfe, pinout ECB.
12 pcs NPN hfe 140 approx, and 8 pcs NPN hfe 440 approx, pinout ECB.
Can you give me the answer?
Thanks.
 

Attachments

  • ksa ksc.jpg
    ksa ksc.jpg
    545.7 KB · Views: 387
Yes - make sure that both Q7's (ie left and right channel) hFE is within 5% of each other. This will ensure the standing current between the two channels is the same. if you have big differences (say 200 and 600) you can get up to 60 or 70mA difference between the channels.
 
Davide, I am sorry to hear that, but cannot help feeling part of your problem was to do with buying parts off eBay. You really have to get components from reputable suppliers (Mouser, RS and Digikey) and stick to the design and selected components.

The direct coupled feedback nature of most CFA amps can also present someone new to DIY audio with some real challenges if your build has problems (which happens more often than not).

A good way to get going is to build something really simple and then progress from there - have you tried Nelson’s Amp Camp design ( Amp Camp Amp - ACA ) or perhaps a JLH?

Anyway, good luck in your quest!

🙂
 
Last edited:
KSA1220AY and KSC2690AY

ebay was the last place to find them.

Hi

If that is the only issue you have then I have more than 10 parts of each bought at Digi-Key.
I can measure closest hfe and then spare 2 pairs of each for your project.
You would only have to pay for shipping.

Fab
 

Attachments

  • 9E12F9E7-47F4-4F92-A940-A04C8FCD2D19.jpg
    9E12F9E7-47F4-4F92-A940-A04C8FCD2D19.jpg
    772.1 KB · Views: 270
Last edited:
ebay was the last place to find them.

Both transistors are available from RS Components or Farnell

I had problems with the original design as previously noted on this forum (mainly solved by the addition of Q1/Q2 collector resistors). However, most of the problems appeared to have been running 2 ripple eaters, one per channel, from a single transformer.

Do you have access to an oscilloscope and signal generator available? If so, it makes it easy to check the waveforms and proceed from there if any problems are found.

I'm very satisfied with my build since I ran both channels from a single ripple eater. Waveforms (sine and square) are 'spot on', minimal noise (< 1mV), and the sound is very good.