A KA-5500 repair, plus a bit of restoration.

[Tiberio]

Hello.
I bought a Kenwood KA-5500 with a dead channel. Once open, it was clear it stopped working when a cup of thick, sugar dense coffee was spilled on the power amplifier board.


Alcohol does not dissolve sugar so my only option was to wash the board with water, which believe me was the last thing I wanted to do...

Then I noticed someone had previously worked on this unit, and changed just one of the transistors in the left channel differential pair with a different type altogether (the pair at the first stage of the power amp). So all 4 transistors (2 per channel) had to go :-(
Here are the differential "pairs" I have removed.




FYI: the service manual says those should be 2SA763 but I am sure the 2SA841 were stock, meaning that substitution was blessed by Kenwood. I have replaced them with KSA992FBU which works just fine.

There was also a BC172 as a replacement for a Japanese 2SA733 in the protection circuit. The BCs have a different pin layout and a shiny metal cap which made it stand out like a sore thumb, I had to get rid of it . It got replaced with a ksa733C and the C matters.

Not sure how to test this protection circuit, I'll post the question separately.



The next problem was that I could not set the bias on this channel. With the Qe8 current source working (I could see 2mA through the resistor Re20 ) , I tried changing Qe10 and YESSSS ! bias started working again. Qe10 was indeed leaking and got replaced by a KSA992F

Since I had the power amp board out of the frame for a shower, I decided to recap it, except for a "nonpolarized cap Ce27" which I would not know how to replace ( I am thinking to use a Nichicon UES Series, Bi-Polar, Radial Leaded Aluminum Electrolytic Capacitors - assuming I can get one of those over here. I will post also this question separately to get some feedback).

As indicated by hopjohn in an old post, the foam on the back of the VU-meters was crumbling in small particles, falling on the tone defeat switch.


... so I also opted for some craft felt as replacement.


Here is the family picture of all that was replaced from the amp board.


My plan is to compare this unit with my other KA-5700 and see which one wins the top shelf spot ;-)

I hope this post will help someone out there !!! Most of what I know about vintage gears, I have learned it here.

 

[hopjohn]

I'm guessing you are in Belgium? For Ce27 100uf 16V I typically use a 100uf 25V Nichicon (Muse) UES often referred to as simply ES. Other Bi-Polar series electrolytics I've used are Nichicon UVP and UEP or Panasonic SU. Here's Mousers Belgium page with the Nichicon ES 100uf 25V cap.

 

[Watthour]

Well done. I like the clever repair of the meter padding, too.

As for the protection circuit, the two 22KΩ resistors (Re51 and Re52) from the Left and Right channel outputs create a summing junction for output voltage. Since the voltages are, by the nature of audio signals, both positive and negative with respect to ground, the capacitor installed there as a buffer needs to be non-polarized.

If/when the summed voltage between the two 22KΩ resistors exceeds about 1.0V, the voltage divider created by Re53 and Re54 will apply 0.6V to the base of Qe27. That would reduce the voltage at the collector of Qe27 through Re55, and subsequently reduce the base voltage of Qe28. When Qe28 is biased off, the relay RLe1 will be de-energized and disconnect the speakers.

As the amplifier output voltages are reduced, and the voltage between the two 22KΩ resistors decreases, Qe27 will turn off. This allows capacitor Ce28 to charge slowly through Re55 (about 3-6 seconds) and will again produce a base voltage at Qe28 sufficient to forward bias the transistor and energize the relay. In this way, Ce28 becomes the timing device for turning on the relay when the unit is first powered on, or when the output stages recover from an over-voltage condition.

 

[dstariha]

Thanks, Watthour, for the information on Tiberio's problem. I can apply this to my own KA-7002 that seems to trip early. The caps on the KA-7002 are polarized elec's; there is a diode bridge bridge in the circuit. I will spend some of this dreary day trying to figure this circuit out after I look up the schematic for Tiberio's unit.
Cheers,
Dave

 

[Tiberio]

And once again, thank you guys. Great help as always !
@hopjohn : yes I'm currently based in Belgium. Thank for the link to the capacitor. It is going to be part of my next order.

 

[Tiberio]

Well done. I like the clever repair of the meter padding, too.

As for the protection circuit, the two 22KΩ resistors (Re51 and Re52) from the Left and Right channel outputs create a summing junction for output voltage. Since the voltages are, by the nature of audio signals, both positive and negative with respect to ground, the capacitor installed there as a buffer needs to be non-polarized.

If/when the summed voltage between the two 22KΩ resistors exceeds about 1.0V, the voltage divider created by Re53 and Re54 will apply 0.6V to the base of Qe27. That would reduce the voltage at the collector of Qe27 through Re55, and subsequently reduce the base voltage of Qe28. When Qe28 is biased off, the relay RLe1 will be de-energized and disconnect the speakers.

As the amplifier output voltages are reduced, and the voltage between the two 22KΩ resistors decreases, Qe27 will turn off. This allows capacitor Ce28 to charge slowly through Re55 (about 3-6 seconds) and will again produce a base voltage at Qe28 sufficient to forward bias the transistor and energize the relay. In this way, Ce28 becomes the timing device for turning on the relay when the unit is first powered on, or when the output stages recover from an over-voltage condition.

Thank you !!!
So I guess that if I play carefully with Re51 and Re52 (e.g. with a 1MΩ trimmer in parallel to one of those resistors) I could trigger the base of Qe27 and see if the relay de-activates. That could be the test procedure I was looking for, no ?

Regards
Tiberio

 

[Tiberio]

I'm a little lost...the only reference I could find regarding the protection circuit was this: "Not sure how to test this protection circuit, I'll post the question separately." Is there an issue with it? What DC voltage do you have at Pins 15 & 16 on the amp board? If you have high DC on either of those pins, it's one reason why the relay won't close.

Hi. The amplifier operates normally, with sound from both channels. My question was more the following: since I have replaced a transistor in the protection circuit, how do I test that - in case of need - it will operate correctly and save my speakers?
I think I got the answer, or at least an idea on which I can work: I will try to trigger a DC voltage on the non-polarized capacitor, and see if the protection activates. I will try t do so by temporarily lower the resistors that charge it (Ce27).

Thank you for asking
Tiberio