I am using two 680R in series for R14 to change the bias range. I am able to get very very small bias current now and max is about 90mA. I am currently running at 80mA. The changes to R4 and R6 as described above for LED plus 1N4148 seems to work well. LED intensity seems to have increased a bit. DC offset is -5mV - an excellent value for not having a trimpot.
Listening to some test tracks now. A very neutral sounding smooth amp. Bass at higher volumes (>5v rms output) has a problem though. On accentuated notes like a kick drum, there is a popping sound or scratchy sound - almost like it is clipping. Kick drums will have a 'pop pop pop' sound. Never heard anything like it.
Any ideas what this is? Could it be too little or too much quiescent bias?
Edit: bad news, listening to a few tracks and amp suddenly craps out, nothing hot or smoking, no smells or burnt appearance. Checked fuses and still good. So something went bad. I saw it happen as the red LED died and sound went away. I immediately removed power.
Not so easy, this AX11. FH9 was an easy dream to operate by comparison...
Last edited:
All my parts are dubious then (but all amps seem to work well up to now), all resistors, caps, transistors, etc. with exception of early on I got latFETs and hexFETs from US sources but since get those from Ali also (genuine). I have only one case of bad fake transistors so far. Let me start probing around to see what burned out.
Edit: both BD139/140's are dead. Although this may have resulted in an earlier pre-damage when I had a bad silicone heatsink insulator pad conduct and caused collectors on both BD's to be tied together and fuse went.
This is probably why it sounded funny and then died after playing for a few minutes. Will replace and see what happens.
Last edited:
It's all good now. Replaced the BD139 and BD140 and BC546 temp compensator. Singing now - bass if fine and amp sounds really good.
Finally, success I can say. Currently, max bias is only about 90mA.
Thanks Prasi for the layout and thanks Bimo and Apex for the design.
Edit: Sounds great but Q5 and Q6 TO92's are very hot - can't touch them more than 1 second! Something off on resistors and currents then with the settings above. Maybe go back to double 1n4148's and original design?
Finally, success I can say. Currently, max bias is only about 90mA.
Thanks Prasi for the layout and thanks Bimo and Apex for the design.
Edit: Sounds great but Q5 and Q6 TO92's are very hot - can't touch them more than 1 second! Something off on resistors and currents then with the settings above. Maybe go back to double 1n4148's and original design?
Last edited:
It's all good now. Replaced the BD139 and BD140 and BC546 temp compensator. Singing now - bass if fine and amp sounds really good.
Finally, success I can say. Currently, max bias is only about 90mA.
Thanks Prasi for the layout and thanks Bimo and Apex for the design.
Edit: Sounds great but Q5 and Q6 TO92's are very hot - can't touch them more than 1 second! Something off on resistors and currents then with the settings above. Maybe go back to double 1n4148's and original design?
You can not use TO92 for VAS transistors, use TO126 with heatsinks.
That was the Bimo mod though - putting in TO126 now will mean twisted legs as 2N5xxx are EBC and typical MJE350/340 are ECB and holes will have to be drilled out.
The original AX11 design seems to use TO92 for VAS as well, the TEF version has MJE TO126.
Last edited:
That was the Bimo mod though - putting in TO126 now will mean twisted legs as 2N5xxx are EBC and typical MJE350/340 are ECB and holes will have to be drilled out.
The original AX11 design seems to use TO92 for VAS as well, the TEF version has MJE TO126.
http://www.diyaudio.com/forums/solid-state/164093-100w-ultimate-fidelity-amplifier-554.html#post4522738
Bimo,
When I go to the link you post, it's not at all obvious what I am looking at/for? Are you saying that the TO92 transistors are fine as is?
You can increase R12 and R16 to 68R like I did to lower the temperature of VAS transisitors. In original Sansui amps from the late seventies these resisitors were 82R (Sansui AU217) and 150R (Sansui AU317). This is push-pull VAS topology. Even with higher value resistors (and hence lower VAS current) you will have plenty of drive from the VAS stage because with push pull VAS current is doubled compared to single ended VAS.
Nice simpler is better design... is there a schematic that goes with this?
my version of layout in post # 7297 http://www.diyaudio.com/forums/soli...imate-fidelity-amplifier-730.html#post4697718
reg
Prasi
I swapped out the 47R for 100R and the Q5 and Q6 are smoking hot - they would burn your finger if touch more than 1 second. It plays music and sounds nice but don't know how long that will last. I have it set for 100mA bias current at present. Based on how hot those two VAS transistors are, I would imagine 680R is what is needed to get temps down to reasonable level.
Or go with TO126 like MJE340/350
There is an occasional crackling or popping sound right now - perhaps imminent failure of the 2N5401/2N5551?
Have you seen this post by Mr. Mile?
Use resistor value to get 6mA VAS current for TO92 transistors.