The KA-9100 restoration.

I should point out that the brackets on either side of the power supply board, as can be seen in the photo of post #63, are shown mounted incorrectly. This is how it came to me, but the metal tabs where the plastic rivet goes through should be at the back of the board, not the front. When positioned correctly the board will be perfectly vertical with the brackets screwed to the chassis.

Interesting note on Ck16 - I didn't remember seeing that. Looked at the photo I took and found it mounted on the top side. Must have been a board change at some point or due to mine having switchable voltage, who knows.

Thanks for the information. I hadn't considered they'd make a change to the board for one additional cap. Replacement of Qk4 on these type boards should be hassle free.

LOOKING GOOD!
Thanks!, much appreciated.
 
Some basic housekeeping is in order before moving away from the back panel. There's probably nothing here anyone restoring hasn't done a hundred times, but I always find it interesting nonetheless just to see what methods people use. I'm hoping that some of you do too.


First, I need to get to the normal/separated switch. It resides behind a shield on the back panel. With the shield off, it's just a matter of removing the two screws from the panel to free it. The sides of the switch allow access to the sliders and contacts for a good cleaning with D5 or D100.
EayQBn8.jpg



Since the rear panel is still opened up, now is also a good time to clean the speaker selector rotary switch. A good spray down of CRC followed by some D100 or D5 will do nicely here. The contacts are on either side of both wafers. I follow the cleaning with a lube of the selector mechanism where the two steel balls are.
siLoC3G.jpg



The RCA jacks on the vintage stuff always need a good cleaning up. I used DeOxit Gold G100L applied with a Q-Tip. In the photo you can see the before and after difference where the AUX, TUNER, and Tape B jacks have been cleaned and the rest have not.
DhLS4jY.jpg



For the insides of the jacks a mascara brush is useful for getting out all the dust and cleaning up any oxidation with some D100L or G100L applied.
VyAv6ID.jpg



The speaker terminals are always dusty and have nasty dead skin in their grooves. A soapy tooth brush does wonders at getting rid of the gunk. I carefully follow the grooves with the bristles of the brush while slowly turning them.
vhj4gmr.jpg


The mascara brush is also useful for getting into and cleaning the wire slots of the terminals.
l9iFhuP.jpg
 
The RCA jacks on the vintage stuff always need a good cleaning up. I used DeOxit Gold G100L applied with a Q-Tip.

Ok, here's a great learning opportunity for me: you're saying DeOxit Gold is good for non-gold standard RCA jacks? I'd bought a tube thinking it would be useful, but thought it's directions said "for gold contacts only".
 
Ok, here's a great learning opportunity for me: you're saying DeOxit Gold is good for non-gold standard RCA jacks? I'd bought a tube thinking it would be useful, but thought it's directions said "for gold contacts only".

Ideally you should use some D series on the jacks to remove any oxidation first (I did a bit of no-no) and then use the G series to preserve what you've cleaned. These jacks weren't too badly oxidized, more dirty than anything. The G series works fine on base or plated metals. The gold thing is more of a way of distinguishing it's not a cleaner it's a preserver.
 
Last edited:
Is there any downsides to disconnecting the rec/play DIN connector on the back?
 
An additional note about cleaning the speaker terminals. After receiving a reminder from gort69 that the terminal screws can be removed on the 9100 (they are reverse threaded type), I took advantage of this and cleaned the insides of the posts and all the thread throughs on the terminal block with D100L. I followed that up with some G100L for preservation. Thanks gort69 for the timely PM.

jBMvc46.jpg


Is there any downsides to disconnecting the rec/play DIN connector on the back?
Just functionality of the jack and potential shorts if the wires aren't managed. I don't like disabling things personally.

the thread is drifting in the wrong direction
We're back on track Johnny boy. Up next is the Triac install. :thumbsup:
 
the terminal screws can be removed on the 9100 (they are reverse threaded type)

The same is true for the two GROUND screws. The backsides are soldered to the panel, but once you overcome that, turning the Phillips screw of the GROUND posts clockwise will back them out of the panel (if you need to eliminate rust spots under them or some such).
 
  • Like
Reactions: Rep
Adding a thryistor (triac) to the KA-9100 Part 1

Among many models, the power switch is a Kenwood weak point. It's important to protect and one of the best ways to do that is to add a Triac inline with it. The Triac suppresses inrush current which otherwise causes arching. The arching damages the contacts which eventually destroys the conductivity of the rocker interrupting electron flow. This renders the switch inoperable and power to the amplifier is lost. When power to the amplifier is lost, it no worky. I make a point to describe this with exactness in jest having had this subject be a previous point of contention. A contemporary was being "nit-picky" about my use of semantics in describing the cause of switch failure. This time, I hope to have left no room for interpretation.

When it comes to wiring the triac I've come to use the annotation provided by Glenn McDonald. For the KA-9100 I'll be using the part number Q6040J7TP along with a safety rated X2 0.1uf film cap and a couple of 100ohm 1/2W metal film resistors. (Thanks and credit to Glenn for the diagram photo below)
zseaXY5.jpg


The first thing to do is to identify which of the wires leading to the switch is coming directly from the 120V AC source this will be the "Hot" as noted in the diagram. In my unit the white wire is the "Hot". This white wire is connected at a terminal of the unswitched convenience outlet to the incoming brown power cord. The yellow wire connects to a red wire from the transformer(s) or "Load" in the diagram.
N1lIbKz.jpg


Next, I want to document what the switch appears like in stock form. In my unit the hot wire (white) is on the toggle side and the load (yellow) is at the middle position. There is a .01uf safety cap across the switch that will be discarded. There is a second set of contacts that is unused on the other side of the switch which will be implemented later in the install.
lMkXVYS.jpg


The wires and safety cap are snipped from the switch. All of the remaining wire and solder is cleared from the tabs of the switch in preparation for the new wiring.
G1JlMCk.jpg


The switch is then disassembled for cleaning. Notice the damage arching has caused to the upper right contact and also one side of the top rocker. This is what we are trying to avoid from happening further. The below photo is prior to any cleaning which should be done with D100L followed by an application of G100L. When reassembling I turn the damaged rocker 180 degrees toward the undamaged contact.
lNlvtdW.jpg


Once the switch is reassembled it is wired with the "hot" to the top toggle side, the gate to middle tab, and the right and left sides bridged together as shown below. Wiring both sides will divide the current, though it's not entirely necessary with the Triac in place. The Triac will dramatically lower the current that the switch sees, though in practice, bridging here is still a positive thing for the life of the switch.
iygfDKr.jpg
 
Last edited:
Adding a thryistor (triac) to the KA-9100 Part 2


The pinout for the Q6040J7TP part is MT1, MT2, G left to right as shown below.
OCvi0bA.jpg


The gate is given a 100 ohm 0.5W resistor. I recommend using a longer length of wire between the gate and resistor legs then what is shown here. It needs to be flexible so use stranded wire for this connection.
pV8s7lp.jpg


A 100ohm resistor is added to a leg of the .1uf safety cap to create a snubber. This will go across MT1 and MT2.
qiLncyI.jpg


Heat shrink tubing of adequate diameter needs to be put in place prior to soldering in the snubber.
TMN6OMy.jpg


The white wire from the switch is put through the heat shrink tubing and soldered at the same time as the resistor side of the snubber to MT2. You have to be careful not to shrink the tubing too much with the heat of the iron while you solder here. You only get one shot at this or you'll be undoing some things, don't screw up. No pressure....we're all counting on you!
3TXy5lT.jpg


Some thin black Insulation is placed on the capacitor lead (hidden below, but can be seen above). The yellow wire from the transformer is placed through the heat shrink tubing and soldered in place at the same time as the capacitor side of the snubber to MT1. Again, you get one shot to not shrink the tubing too much so it can be put in place. (Rob Schenider voice form Waterboy) "You can do it!"
QpX9ogz.jpg


Success!, no early shrinkage. The tubing is put over the leg of MT1 and insulated.
faT73ib.jpg


The final connection is made from the black wire coming from the switch to the 100ohm resistor that connects to the Triac's gate. Additional insulation is provided on each side of this resistor. Zip ties provide strain relief to the rather fragile leads of the resistor. The Triac is mounted with a split washer with one of the existing self tapping screws (albeit a little longer one ) that mounts the left side amp heatsink to the main chassis. Some thermal compound can be applied to the back of the triac body before mounting it to help keep things cooler. I recommend some thread lock to the back side of the screw to keep it in place, but not until after work on the amp boards has been completed.
3hm7mw0.jpg
 
Last edited:
Looking good John, I'm sure people will bookmark this thread. Very good reference.
I take it you are going to keep this one.

Thanks John, I appreciate it. This one is filling a gaping hole in my personal collection.

I agree, John. I don't know if it's possible, but it might be cool to have a "sticky" section for threads with best practices/most informative for each model. And similar to the Marantz group, maybe a sticky for schematic errors? Again, I do agree, this thread is a keeper! Nice work, Mr. Hopjohn!

I think a sticky makes sense, but it should be a comprehensive new thread that lists and links all the best restorations for each Kenwood/Trio/ Accuphase model. If anyone is up for it, give it a go. I've got my hands full at the moment.

HUMOR AHEAD, I TEASE...dlucy is busy right now cleaning rust off his 9150 and hijacking a thread. Sicman is microwaving his Stoufer entree for a parts tray and tripping over his 9100 and 9600s. gort69 is at work polishing his Camaro and deleting my nutty memes before anyone else sees them. I'm taking another photo for this ridiculous thread (think how useless this thing will be if/when the images go offline.) I nominate rjsalvi. Robert's white-out 5700 hasn't sold so he's just twiddling his thumbs waiting for a payday.

Totally kidding with all of you! I'm no comedian, but I hope you laughed a little. And seriously if anyone wants to make that resto link thread, that'd be awesome.
 
The left and right amp boards are up next. (X07-1500-11 A/2 and B/2) or (X07-1500-00 A/2 and B/2)

Access to the amp boards require removal of the large heatsinks. This is a pretty straight forward procedure that involves taking out around a dozen screws. With the heastsink out, all of the thermal paste from both the heatsink and the amp module is removed to prevent making a massive mess on the bench. The amount of paste Kenwood used seems way too much. Perhaps that was the recommendation back in the 70's with the type they had available, but using today's modern thermal compounds, the recommendation is instead for a very thin coat.

The heatsink freshly removed from the amp module. The stock application of the compound is curiously thick.
cIeyQhF.jpg


The left amp board in stock form. (X07-1500-00(A/2)

96rZE7r.jpg


Planned changes:

Ce3/4 100uf 35v 10mm > 100uf 35V PX 10mm
Ce13/14/15/16 47uf 16v 8mm > 47uf 35v KL 8mm

De1/2 EQA01-24R Zener > 1N5252B 24V 0.5W Zener

Qe5/6/7/8 2SA912(R) or (S) > KSA1220AY (TO-126 package)
Qe9/10 2SC1400(U) or (E) > KSC1845F
Qe11/12 2SC1885(R) or (S) > KSC2690AY (TO-126 package)

All resistors to metal film 1% 0.5W, except Re13/14 3.3K 1W > 3.3K 2W metal film

Vre1/2 470ohm > 500ohm Multi-turn
 
Qe1-4? 2SC1775 >>> 2690, or 1845?

The 2SC1775 is a low noise, low current amplifier so the low noise, low current amplifier KSC1845 would be a better match than the not-specifically-for-low-noise, higher current KSC2690A, right?
 
The left amp board is completed with the components as planned in my previous post. It should be noted that the value of R23/24 may or may not match what is shown in the service manual, but should be kept as it is found on the board. In my case the stock mounted part was 100ohm, the S/M has it listed as 1K. I replaced it with the 100ohm (seen dead center of the board) value that was there.

iMr8PMA.jpg


The TA-100WA gets a fresh coat of thermal paste before reinstalling the heatsink.
J1sMayK.jpg



Qe1-4? 2SC1775 >>> 2690, or 1845?
The 2SC1775 is a low noise, low current amplifier so the low noise, low current amplifier KSC1845 would be a better match than the not-specifically-for-low-noise, higher current KSC2690A, right?

The Qe1-4 2SC1775 aren't problematic so I don't intend to change them, but if you were going to the KSC1845F is what to use.

Edit: FWIW Qe5 and Qe7 on the left amp board had visibly stressed solder joints, so I'd say replacing those, at minimum, is a healthy idea.
 
Last edited:
Hey John:
You don't need much thermal paste, that's just to fill in the unevenness of metal to metal contact.
I've found many with hardly any grease or way too much. That pic you showed looked quite excessive.
More is not better here.
If redoing the amp boards I would put all new components, why leave an old one or two that may fail. :dunno:

Nice work on that Triac fix too.

John
 
Here is a view of the solder side of the right amp board highlighting evidence of stressed joints at Qe6 and Qe8. These are the corresponding transistors to Qe5 and Qe7 that I mentioned had poor joints on left amp board. One of these joints is holding on by a thread, nearly desoldering itself from the board. The others all show thinning and ringing from thermal cycling. When mounting the replacements, be sure to leave some breathing room between the transistor body and the PCB.

jeMcth0.jpg
 
Last edited:
Back
Top Bottom