KA-8006 staying in protection

[JoshHendi]

Hello all,

Got a KA-8006 from a friend that he was gifted. The unit powers up but doesn't come out of protection. All I have done so far (besides a visual check for bad components) is check the output transistors using the diode tester on my DMM and they all seem to check out. I am curious and looking for suggestions on what to do/check next?

Keep in mind: while I know the basics on repair, I am far from being considered an expert, or even average, at repairs. I'd really like to get this fixed myself with the help of AKers. I have been waiting to come across a nice unit that needs some attention, and hoping this one will help expand my knowledge and troubleshooting.

What should be the next logic step(s) to take to help find the problem?

 

[petehall347]

check the speaker outputs dc voltages before the relay

 

[EchoWars]

Possibilities are:
One or both channels blown and feeding enough DC to the protection circuit that the circuit will not allow the relay to close, OR:
Amp channels are fine, and there's an issue with the protection circuit that prevents the relay from closing.

Note that the power amp has a couple of 'known bad actors' in the 2SC1451's.

Yes, Step #1 is to see what kind of DC is at the output of each channel.

 

[JoshHendi]

10mv and 15mv before hitting the relay, and 0mv coming out...

Thanks for the info... Should i check/replace the relay? Or could it be another problem on the power/relay board?

 

[EchoWars]

10mv and 15mv before hitting the relay, and 0mv coming out...

Then the protection circuitry is the issue.

Relays are nearly indestructible. Contacts get worn over time, but these devices rarely fail (but because of worn contacts, it's nice to replace them). The most likely cause is a bad Qk1, Qk2, or Ck4. Or Rk9 (or Rk12) has desoldered itself or gone high.

Measure the voltage on the cathode of Dk12. Should be 50V & change. If so, then check the voltage on the base of Qk1. Should be well under 500mV. The voltage on the base of Qk2 should be somewhere just under 1V. If so, then Qk2 should be ON, and the relay closing. If not, then check that the coil on the relay has a resistance reading of about 650 ohms. If that is true, then I'd just replace Qk1 and Qk2 and be done with it.

Replacement relay is on Omron MY4-02-DC24. Qk1 -> KSC1845, Qk2 -> KSC2690AY

 

[JoshHendi]

Thank you EchoWars!

I wont be able to get back to it for a couple of days now, but I will sure check those and reply back

 

[JoshHendi]

DK12 cathode voltage came in at 53v. Tried checking the voltage on Qk1 base and only got 3mv? Base of Qk2 was 420mv.

Verified Ck4, Rk9 and Rk12 are all within spec.

Should I proceed to replace QK1? Only QK1? Or should I get both transistors replaced? What would be suitable replacement(s)?

Thanks!

 

[EchoWars]

The low voltage on the base of Qk1 simply means that the protection circuitry sees nothing wrong.

That being the case, there's nothing to prevent Qk2 from turning on. Power gets applied, and the +50V is fed through Rk9 (check the value of this one if you like). After Ck9 charges, the voltage on the base of Qk2 will rise to where it will conduct, closing the relay.

Dk5, Rk5, and Ck3 work together to turn off Qk2 as soon as power is removed to prevent any weird transients after the unit is turned off.

So, in order to CYA, I'd replace Qk2, Ck4, Dk5, and Ck3. But if you're digging in that deep, I'd replace all the electrolytics on that board, and probably both protection transistors (Qk1 and Qk2), and install a new relay. Might note that Dk6, Dk7, Rk1, Rk2, Rk3, Rk4, and Rk5 all get very hot, and I'd clean off all the old solder from these connections and reflow them.

Parts List (assuming that the parts on the board match what is in the unit - where there's a difference, the part chosen should match the one in the unit unless it is obvious that it has been changed out at some past date):

Omron 4PDT 24V relay - 653-MY4-02DC24
Fairchild TO-92 transistor - 512-KSC1845FTA (Qk1)
Fairchild TO-126 transistor - 512-KSC2690AYS (Qk2)
Fairchild Switching diode - 512-1N4148 (Dk5)
Nichicon PW 10uf 50V cap - 493-1890-ND (Ck3)
Nichicon EP 10uf 16V bipolar cap - 647-UEP1C101MPD (Ck4)
Nichicon HE cap - 493-1612-ND (Ck5 & Ck6) (330uf 50V)
Nichion HE cap - 493-1581-ND (Ck7 & Ck8) (330uf 35V)
Nichicon HE cap - 493-1548-ND (Ck9)
Solder Flux - 590-8341-10ML (A tiny dab of this on every solder joint before you hit it with the iron and solder will make for near-perfect joints...trust me . Cleans up nicely with a Q-Tip and alcohol.)

If any of the resistors mentioned (Rk1, Rk2, Rk3, Rk4, and Rk5) are badly discolored, they should be replaced. But I'd bet good money that the value of the parts on the board do not match the schematic, except perhaps Rk5. Panasonic has an excellent line of metal oxide resistors in 1, 2, and 3W sizes (ERG-2S). The 2W parts are about the size of the vintage 1W parts, and the 3W parts about the size of the vintage 2W parts. If you think there are some that need replacing, chime in with the board designation and the resistance value and I'll find you a part number.

Care must be taken installing the transistors. Study the PC board and the data sheets for the transistors till you are certain of the lead placement.

 

[JoshHendi]

Update on this... After pulling and rebuilding the protection board using the parts that Echowars shared, it is now working correctly! I didn't replace the relay, but did switch out all of the resistors, capacitors and the two transistors

 

[JoshHendi]

After adjusting the bias and unit idling for a few minutes, I noticed quite a bit of heat coming off of the protection board, specifically the top two center resistors, R1 and R2... Is this to be expected? The heat generated feels like all of it is coming off of those two resistors, which are 680 and 820 ohm 2W

 

[EchoWars]

Got a voltmeter, right? Measure the voltage at the output of the board (the +14V at pin 14 and the -14V at pin 12). If that looks OK, measure the voltage across R1 through R4, and calculate the power each is dealing with (V²/R).

The schematic says they are all 1W, but they may well be 2W, and perhaps even a different value than what the schematic that I have says, which is:
Rk1 - 680 ohm
Rk2 - 820 ohm
Rk3 - 470 ohm
Rk4 - 680 ohms

If the output voltages from this simple zener regulator are OK, and the power dissipated in those resistors is within their rating, then you're good to go.

 

[JoshHendi]

Got 14.9V on pin 14 and -14V on pin 12

When you say measure the voltage across R1 threw R4... Are you saying to measure the voltage on each side of each resistor, and then do the power calculation? Just want to make sure I understand and am doing so correctly... Example: voltage on R1 on one leg is 53.3V and 30.9V on the other... 53.3-30.9=22.4... 22.4x22.4= 501.76... 501.76/680=0.73

Is any of the above done correctly? (never measured and calculated voltage on resistors...)

 

[EchoWars]

The power dissipated in any component is a function of the voltage across it multiplied by the current through it. But for a resistor, an easier way to calculate power is to directly measure the voltage across it (connect one meter lead on one side of the resistor and the other meter lead on the other side of the resistor and simply read the voltage with the meter), square that value, and divide by the resistance value...the answer is in watts.

What you will find is that dissipated power heats things up fast. 1 or 2 watts in a component like a resistor makes them uncomfortably hot. Rule of thumb is that the dissipated power in a resistor should only be about 1/2 of the resistor's rating, so if you calculate that a resistor is dissipating, say, 0.9W, then a 2W resistor is appropriate. Also, power resistors that get appreciably hot should not be mounted flush against the PC board...they need 1/2" or so in order not to scorch the board.

 

[JoshHendi]

Right on, thanks for the info!

Looks like I am good to go then? Both of the resistors are 2w... voltage calculations land within the desired values. Both are mounted high off of the PCB

Thank you so much Echowars... because of your time, knowledge, and willing to share both, I got this thing fixed when initially I thought I'd have no chance! You're the best!