KA-8100 Restoration & repair of several build defects

Status
Not open for further replies.
MODE & Loudness (Frequency & Level) switches

These are the three sealed or semi-sealed switches on the right side of the control board. Normally would probably leave these alone, but the MODE switch was acting up so something had to be done.

In the video linked below, the guy has melted holes in the clear plastic shields around these switches to provide entry points for Deoxit. This starts at 14:14. The whole video is informative and entertaining, well worth watching. Wonder if the videographer is an AK member ??
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwij6a3puMTVAhXm64MKHTbbAmMQtwIIKDAA&url=https://www.youtube.com/watch?v=WonSXuOnTtU&usg=AFQjCNHjqC46ZczkAo_p6I2JTfOQKE_65w



Anyway, I opened up the control board on my KA-7100, which has two of these switches, and practiced my technique on it prior to working on the KA-8100 switches. I used a punch from an inexpensive set of four from Walmart. The tool needs to be the right diameter, not too thin or thick. Heated it up in the flame of my gas kitchen stove and carefully melted a hole in the plastic a little at a time.

The construction of these switches is such that there is a wafer inside the switch about in the middle. On one side of the wafer, closer to the PCB, are the switch contacts. On the other side of the wafer is an air space. In the video, the holes were melted into the inner portion of the switch where the contacts are. But when I attempted that on my KA-7100 switches, the punch was pushing melted plastic inward, towards the contacts. So I tried putting the hole on the other side of the wafer, away from the contacts. That worked well as the Deoxit D5 spray went everywhere inside the switch regardless of where the hole was situated.

After a couple sprays, the D5 could be seen through the clear plastic partially filling up the insides of the switches. After working the switches through their range of motion about 50 times, I then sprayed compressed air (canned) into the same hole. The air flushed out most of the liquid nicely. I cleaned up the mess with IPA and repeated the process with Deoxit Gold and another air flush and cleaning, then waited overnight before powering up.

I never got any of the holes large enough to insert the little red Deoxit tube inside the switches. But I was able to press the tube sorta-part-way into the outer lip of the hole, enough to get the fluid (or air) inside. I did a total of five switches and all went well! The one switch on the KA-8100 that had been misbehaving worked fine after this treatment.




EDIT: Two things I would like to add here...

First, in regards to the video linked above, when I say that it is informative, entertaining and worth watching, to me it was all of those things. But I am not suggesting or implying that every method or action in that video should be followed exactly. Information there is presented in a coherent and understandable way, which I find to be a rare thing on youtube. And it is about KA-8100, so I thought I would add the link to this thread for whatever it may be worth to readers. Each person should, hopefully, assess what is presented and decide for themselves if anything might be worth adding to their own processes. Same thing with any information presented in this thread.

And the second thing that I would like to add to this post is that melting holes in the plastic cover of switches, as I have done here, is a risky maneuver and there is definite risk of a negative outcome. I am documenting one possible solution to a problem. I am not advocating that anyone else attempt this, especially if you feel uncertain about doing so.

In my case, I actually have past experience with doing something similar. Admittedly, it is highly unusual experience most likely. When I was a kid my Father and I were into slot car racing. I was young and he did most of the work on the cars. There was a certain Revell plastic slot car chassis that we "modified" so that a popular motor type would fit into that chassis. This modification procedure involved open flames (stove), kitchen knives, and melting/cutting that plastic chassis in just a certain way. My dad came up with this of course, but I got to be quite good at it.

Thus, the process as suggested in that video did not intimidate or put me off, indeed I had confidence going in that this method would work for me. As with anything, YMMV


MODE+Loudnessx2 DSC09271rs800.JPG
 
Last edited:
.
.The same switch and wiring arrangement was used on KA-7100 as well (not sure about KA-9100 or any other Kenwood models, but it is possible).

Really enjoying this thread Roger - thanks for posting it. The same power switch was used in KA-6100,7100, 8100 & 9100 and probably in the KT-7500, and all used only 1of the 4 sets of contacts. The only difference is that the higher powered models possibly used stiffer springs under the little nubbons (as is the case on KR-9600 vs 6600/7600), likely to limit bouncing.

I usually slide a piece of wire insulation over the safety cap leads prior to soldering.
 
Last edited:
Blackened transistor leads on main board

This black, furry looking corrosion was on the legs of all TO-92 transistors on the main board. I cleaned it off using 91% isopropyl alcohol and small brushes. Most of the transistor legs were accessible. I did have to remove a resistor or a cap to get to one or two of them. The alcohol did a good job of removing the black fuzz. I was careful to also clean the area on the bottoms of transistors between the legs, where they enter transistor body, as this is the point where legs are closest together (whether that matters I do not know, but it was done to be on the safe side).

Forgot to take "after pics" of the cleaned legs. Pics attached show transistor legs before cleaning.
The black stuff is dielectric paint, probably lacquer, that dried up and is flaking. They probably used it to prevent arching between legs of transistors at high voltages, later on they just used ptfe on the center leg in high voltage applications. I don't know why anyone thought it was corrosion since it would corrode green if it was tin plated copper or red if it was copper/zinc plated steel.
 
The black stuff is dielectric paint, probably lacquer, that dried up and is flaking. They probably used it to prevent arching between legs of transistors at high voltages, later on they just used ptfe on the center leg in high voltage applications. I don't know why anyone thought it was corrosion since it would corrode green if it was tin plated copper or red if it was copper/zinc plated steel.

Not buying that, sorry. :rolleyes:

For every issue I encounter when doing this type of project I do thorough searches. On the topic of "black leg transistors" I found several past threads and/or posts, both here and at diyaudio.com. No one in those past threads mentioned dielectric paint. As for what was said, anyone can search with google advanced search, as I did. Below are a couple of the references I found.

There is not 100% consensus within the community on whether the black corrosion on transistor legs is harmful. Some say maybe yes, some say no.

This thread has 4 posts by EchoWars. No mention anywhere of dielectric paint.
http://www.audiokarma.org/forums/index.php?threads/transistor-leads-that-turned-black.180725/

In this thread, another user states that an exceptionally bad case of black legs caused noise in his amp. Starting at post #11
http://www.audiokarma.org/forums/index.php?threads/au-9900-fun.565642/#post-7427451
 
Last edited:
  • Like
Reactions: mbz
....

I usually slide a piece of wire insulation over the safety cap leads prior to soldering.

That would certainly give a neater appearance. The power switch is not visible, even with the amp's top cover off, because there is a metal shield covering it. But yeah, good idea gort69 :thumbsup:
 
Really enjoying this thread Roger - thanks for posting it. The same power switch was used in KA-6100,7100, 8100 & 9100 and probably in the KT-7500, and all used only 1of the 4 sets of contacts. The only difference is that the higher powered models possibly used stiffer springs under the little nubbons (as is the case on KR-9600 vs 6600/7600), likely to limit bouncing.

I usually slide a piece of wire insulation over the safety cap leads prior to soldering.
You can still buy an alp switch that has similiar internals and rebuild your old switch. It's called the alps tv-5 and it's available on ebay. I bought ten and used one to rebuild the KR-9600.
 
Last edited:
Not buying that, sorry. :rolleyes:

For every issue I encounter when doing this type of project I do thorough searches. On the topic of "black leg transistors" I found several past threads and/or posts, both here and at diyaudio.com. No one in those past threads mentioned dielectric paint. As for what was said, anyone can search with google advanced search, as I did. Below are a couple of the references I found.

There is not 100% consensus within the community on whether the black corrosion on transistor legs is harmful. Some say maybe yes, some say no.

This thread has 4 posts by EchoWars. No mention anywhere of dielectric paint.
http://www.audiokarma.org/forums/index.php?threads/transistor-leads-that-turned-black.180725/

In this thread, another user states that an exceptionally bad case of black legs caused noise in his amp. Starting at post #11
http://www.audiokarma.org/forums/index.php?threads/au-9900-fun.565642/#post-7427451
It's flaking paint that has dried. Sorry I can't convince you. I replace all transistor in vintage gear when performing a recap (and zeners, and diodes, and anything else that is currently available except resistors, carbon comps last, Germans replace resistors with metal film) anyways and therefore say, go ahead and replace them. It's still flaking paint and not corrosion.
 
SUBSONIC-LOW-HIGH filter switches

Before winding up this thread I wanted to post my observations about the Sub-Low-High filter switch.

IMO this switch assembly is the Achilles' Heel of the KA-8100. Signal passes through all sections of this switch even when in the off position(s), which not unusual. But the three sections of this switch are the smallest, least robust mechanically, and have the smallest contact area compared to all of the others in this amp. Not surprising that out of 25-30 threads I read researching KA-8100 prior to doing this project, the most frequently mentioned problem area was "switches", and several times the Sub-Low-High was named specifically.

Yet, although I believe in disassembly for thorough cleaning of any known troublesome switch, in the future I personally would only disassemble this particular switch as a final resort. Getting it back together properly is tedious and risky, the middle section in particular.

And the way this switch is situated on the control board, it is possible to get the D100L needle point applicator (or D5 red tube), close to viable entry points on both ends of all three switches. The only significant obstruction is Ci36, a large mylar cap directly beneath the Subsonic switch. If I do another KA-8100, I would gladly remove that one cap to gain easy access for Deoxit.

I forgot to take any "after" pics of this these switches. Here is a before pic showing the middle section disassembled.

DSC08823  rs800.JPG
 
Last edited:
Roger - I'm going through my KA-8150 and following along. Ci8 in this photo of your project appears to be a bi-polar cap. Mine has a polar 330uf 10v cap as is called for in the service manual. Thoughts? Thanks!

And here's the photo I forgot to attach

191DSC09180.JPG
 
Last edited:
@gort69

Yes, and the new Nichicon Muse ES caps at Ci7 & Ci8 are 330/10, same ratings as the originals. They are also the same diameter and have the same lead spacing although, IIRC, they may be a couple mm taller.

Ci7 & Ci8 had 0.1 VDC across them when I measured. In my readings I have come across some who recommend bi-polar for positions with very little bias voltage, in particular ilimzn. He mentions bi-polar caps in this post, I believe the 6th paragraph from the bottom. http://www.audiokarma.org/forums/in...to-better-than-new-sound.656640/#post-8745545

ilimzn is not as active as he may have been in the past (here and diyaudio). But I consider him one of my top reference sources, along side of EW, ConradH and a select few others. It is interesting to google and read his past postings which include (positive) interactions with and references by EW.

I am planning on replacing those two similar cap positions in my daily driver KA-7100 with bi-polar next time I have her on the bench.
 
Last edited:
As for ilimzn's recommendation, to me it makes sense that a cap designed to work with either positive or negative bias, should also work well when the voltage is near zero as (I would assume) it could possibly be called on to function as voltage swings through zero. As opposed to a polar cap which works optimally at some percentage of its nominal working voltage, (don't recall the specific numbers cited but it is something like 75% give or take), and has also been said to be less efficient and have reduced life when its bias voltage is far below that optimum value. 0.1V is nowhere near the nominal (or optimal) working voltage of those 10V caps.
 
@roger2
I just acquired a ka-8100 and although I am a not an EE or anything close to it I've been following your posts on rebuilding the ka-8100 and thought I would give it a try.
The 8100 sounded good but there was no difference when using the tone controls so here I go! I have found your posts very informative and am pretty sure I can follow
along with out screwing up anything. However I have two questions: Is it normal to have to get different components from different suppliers? And: Would you have a
'BOM' list that you would be willing to share? Thanks
 
Status
Not open for further replies.
Back
Top Bottom