The KA-9100 restoration.

Wanna bet it's a 2SC1775? :biggrin: That's okay, I'm doing a 9800 in which I've replaced everything! ... and the L CH is MIA. It's something small and small takes longer to find, but it can't hide from me! ...or you. Time for a cigar break. :thumbsup:

Nope not the 1775s. I wish it were that'd be simple. It's rushing noise in the preamp and at the very least one of the jfets Q2 2SK68A(L) is responsible.
 
Update

Upon completion of the KA-9100 and detailed listening I was getting rushing noise from the right channel. I had hoped that maybe it was coming from the amp board since it's fairly accessible, but switching the amp to separate mode and reversing interconnects R to L across the pre-ins / pre-outs made the noise switch channels. Of course this eliminated the amp as the source of trouble. The noise was affected by volume, but not by the switching of tape modes or tone defeat. That lead me to consider Q6 2SA850 might be suspect. After swapping Q6 (and Q5) for a 2SA1220AYS the problem remained. Leaving only Q2 or Q4 as possibilities. A swap of Q3 and Q4 resulted in no change, but a swap of Q2 and Q1 resulted in the noise moving channels. Q2 is the noisy offender!! a 2SK68A(L) jfet.

A used, old stock 2SK68A(L) jfet was shipped from my favorite parts source, el gorto sesenta nueve, perhaps you've heard of him. (Muchos Gracias!) Priority shipped, I received the part today. The 9100 was awaiting its arrival prepped on the bench for install. After removing the part from the cleverly labeled anti-static bag I dropped it in and gave it a test. No more rushing noise. The 9100 is now healthy. I'm putting this PITA on the shelf for a long listen. I win!
 
Buen trabajo senor Juan. (de nada) Cervezas.

You definitely won the early rounds, but it made a strong comeback.

If I were rjsalvi I'd suggest you name this one Murphy (after the law).

Enjoy!
 
Here are my personal build notes and resistor list. I hope it will be of use to those attempting to repair or rebuild their KA-9100.


DISCLAIMER: If I begin seeing efforts to monetize from information provided in these lists they'll be removed. These are intended to help individuals rebuild or repair their own personal KA-9100 amplifiers, nothing more. If anyone suspects or witnesses a misuse of the information provided here as a means to profit, please report it to me immediately via PM, thanks.
 

Attachments

  • KA-9100 resistors by board.pdf
    109.8 KB · Views: 133
  • KA-9100 Build Sheet.pdf
    164.8 KB · Views: 183
On the other hand, anyone trying to make a PROFIT from rebuilding KA-9100's is gonna find they'll have to give away their time for free and start taking anti-stress meds. They are an enormous, involved undertaking.
 
Thanks for posting the notes, mine would be un-intelligible LOL.

My wife dresses me like this when I'm away from the electronics.

View attachment 103 0845[/ATTACH]
 

Attachments

  • perturbed.jpg
    perturbed.jpg
    67.9 KB · Views: 31
Hi Hopjohn,
I am very impressed. Nice work. You have given me some inspiration to perhaps re-cap a 7100 that I just got working.

Again. Very impressive.
 
Murphy has struck back. During functionality testing I did only a quick listen to ensure the phono section was working and missed something. The output balance is way louder in the right channel than the left. The front end guts are spilled out AGAIN. I've narrowed it down to the eq amp. The input is good, the output is not. Seeing as how I've pretty much replaced everything but the transistors, I have to suspect another 2SK68A(L).

I've been meaning to put a simple signal probe together and this particular troubleshoot begs for one to help find the mischievous devil. I made a quick trip to Five Below and picked up this little USB powered Panda Bear speaker to use. His other animal friends begged me to take them home also, but no, just Panda. I tested proof of concept with some alligator leads and a 0.22uf 400v axial film cap and it works well. Once I've finished building the Panda probe he is going to tell me where my problem lies.

2862498_critter-aux-speaker-ast_ecom1727_ecom1728-4.jpg
 
7u7M09s.jpg


The Panda probe suggests that Qi29 2SA899 is faulty. When checking the collector of Qi29 there is a discernible difference in gain compared to probing the collector Qi30 in the right channel. Working back toward the input the levels remain equal in both channels. I'll swap out both Q29 and Q31 (Qi30 and Qi32) for 2SA1220AY and KSC2690AY respectively and may also replace the four 2SC1328 at Qi25-28 with KSC1845.
 
The Panda probe suggests that Qi29 2SA899 is faulty. When checking the collector of Qi29 there is a discernible difference in gain compared to probing the collector Qi30 in the right channel. Working back toward the input the levels remain equal in both channels. I'll swap out both Q29 and Q31 (Qi30 and Qi32) for 2SA1220AY and KSC2690AY respectively and may also replace the four 2SC1328 at Qi25-28 with KSC1845.

Do you hFE-match complementary pairs like the 1220A/2690A?
 
Update

As much as I like Panda probe he wasn't pointing me in the right direction. I replaced the left channel transistors mentioned in the last post, but it made no difference. There was a thought that was sticking in the back of my head from the time I discovered the unbalanced levels...although the left channel was considerably lower in level it wasn't at all distorted. Typically a failing transistor will exhibit noise along with the signal if there still is one. My focus at that point shifted to an errant resistor. I recall being up really late while swapping in the metal films on this board, and as a much as I tried to prevent making an error by going through redundancy checks it happened. Ri115 was replaced with a 1.6K when it should have been a 160 Ohm. It took a good while to find it, but that was where the little devil was hiding. With a 160 ohm installed, the channels are level again. It won't beat me, I refuse to quit. Now to reassemble it without breaking off another wire.

Do you hFE-match complementary pairs like the 1220A/2690A?
No, I gain match differential pairs in a main amp circuit when needed to lower offset.

Have you tried this?
Ahh yes, the magic 8-Ball. 35 years ago I used to ask it things about little Sally and never liked the answer so I used it as baseball. Good riddance.
 
Now there ought to be a law against numbering a resistor board slot with the same numbers as the resistor value!
It's done intentionally actually. I've seen it too many times for it to be a coincidence. Funny we both had issues with a 1.6K involved, what are the odds?
 
Back
Top Bottom