For a friend I am repairing a Harman Kardon HK3490 receiver. One of the channels was dead. After checking the output stage both output transistors, the bias transistor, a safety circuit resistor R362 and one half of the current limiting resistor R356 are broken. I have found and installed replacements but are still waiting for R356.
Does anyone have an idea what could have caused such a pretty big failure? The power supply is ok and the other channel is working. All the other transistors measure ok.
Does anyone have an idea what could have caused such a pretty big failure? The power supply is ok and the other channel is working. All the other transistors measure ok.
I can't imagine what could have caused but you need to replace the drivers Q314,316 THE BIAS TRIMMER VR32,the Q318,Q320,322,R536.
Use a protection bulb tester after repair.
Connect two 10-22ohm /10w resistors in series with each rail of power supply and amplifier.
Connect a voltmeter across resistor.
Push amplifier switch on.
Read the voltage and look at the bulb.
If you will see something about 2v across each of 22ohm power resistors (depending from VR32 adjustment)and the bulb is dark everything sould be ok.
Connect a voltmeter between speaker out+&speaker out-,you are looking for D.C voltage present.Anything about 0-100mV is ok.
Push amplifier switch off.
Remove the bulb tester but not the resistors.
Connect a voltmeter across R356(CN32),you are looking for BIAS current.
You can expect something 0-0.02(20mV) here,if everything is ok.
Now check if this will vary when you carefully adjust the VR32 bias trimmer.
If yes,adjust the trimmer for the minimum voltage there.
Do not remove the voltmeter.
Power off.
Remove the resistors.
Power on.
Readjust the bias trimmer CAREFULLY for about 0.015v.
See service manual recommendations,if available.
During test nothing is connected to amplifier,No speaker,No input.
Be careful when work with the main ⚡⚡⚡
You need an isolation transformer .
Keep one of your hand in your pocket.
Make sure that you have a circuit breaker installed.
Good luck!
Use a protection bulb tester after repair.
Connect two 10-22ohm /10w resistors in series with each rail of power supply and amplifier.
Connect a voltmeter across resistor.
Push amplifier switch on.
Read the voltage and look at the bulb.
If you will see something about 2v across each of 22ohm power resistors (depending from VR32 adjustment)and the bulb is dark everything sould be ok.
Connect a voltmeter between speaker out+&speaker out-,you are looking for D.C voltage present.Anything about 0-100mV is ok.
Push amplifier switch off.
Remove the bulb tester but not the resistors.
Connect a voltmeter across R356(CN32),you are looking for BIAS current.
You can expect something 0-0.02(20mV) here,if everything is ok.
Now check if this will vary when you carefully adjust the VR32 bias trimmer.
If yes,adjust the trimmer for the minimum voltage there.
Do not remove the voltmeter.
Power off.
Remove the resistors.
Power on.
Readjust the bias trimmer CAREFULLY for about 0.015v.
See service manual recommendations,if available.
During test nothing is connected to amplifier,No speaker,No input.
Be careful when work with the main ⚡⚡⚡
You need an isolation transformer .
Keep one of your hand in your pocket.
Make sure that you have a circuit breaker installed.
Good luck!
Last edited:
I have already replaced Q318,Q320,Q322 and R362. I also replaced C316 and C318 just in case. I then powered up the amp and no smoke, no DC on output. I then tried to adjust the bias and discovered that R356 was blown. As this is a safety resistor they are a bit hard to find so still waiting on the arrival.
Regarding Q314 and Q316. These are impossible to find. Any suggestions for replacements? KTA1360Y and KTC3423Y?
Regarding Q314 and Q316. These are impossible to find. Any suggestions for replacements? KTA1360Y and KTC3423Y?
Ksa1381,ksc3503.
For each of 0.22ohm resistors you can use 4x1ohm metal film resistors in parallel.You need 8 pieces.
You need to replace latter when you will receive your order.
As R536 was broken no current flow and the dc check no sense.
For each of 0.22ohm resistors you can use 4x1ohm metal film resistors in parallel.You need 8 pieces.
You need to replace latter when you will receive your order.
As R536 was broken no current flow and the dc check no sense.
Last edited:
DVM diode scale check of driver transistors, VI limiter transistors, VAS transistor, and diodes, can find shorted or open ones. This test is done at 1.5 to 2 volts. It cannot predict damaged transistors that will collapse to 0 volts as soon as 20-30 volts is put collector to emitter. Instead of replacing everything that might have been damaged by rail voltage coming out the emitter of the output transistors, I perform an Iceo check of driver transistors & previous at 12 or 24 v. Put a power supply (wall transformer okay) series a 47k resistor series a milliammeter, series the collector to emitter terminals of the suspect part. Plus to collector npn, minus to collector pnp. Leave the base open. If current is microamps the transistor or diode is most likely fine. If current is supply voltage/47k, the transistor or diode has failed.
Once all good parts are found, working forwards, you can generally stop replacing parts.
This output stage could have been damaged by either shorted speaker wire, or by an internal short in the speaker itself. Drivers with bad suspension can scrape the windings against each other, eventually removing the insulation and causing a shorted turn. Such shorted turns in a driver can cause overcurrent of the output stage and blown transistors emitter resistors etc. The most common short in speaker wiring is a 1/4 phone plug to speaker pulled out partially so the tip shorts both jack terminals. Also dual banana plugs for a speaker can be installed incorretly with one side touching the amp case.
ksa1381 is out of stock everywhere. Any alternatives? DK lists MJE340/350 as options. TTA004B/TTC004B maybe?
Update: I replaced Q314 and Q316 with TTA and TTC. Also replaced the trimmer and R356.
Turned on the amp with lightbulb inline. No smoke but also no bias-voltage.
Turned out R348 was also open. Replaced it and voila! I had bias-voltage, no smoke!
Removed the lightbulb, turned it on and let it sit for a while, adjusted the bias for both channels and played some music.
Amp is alive again!
Thanks.
Turned on the amp with lightbulb inline. No smoke but also no bias-voltage.
Turned out R348 was also open. Replaced it and voila! I had bias-voltage, no smoke!
Removed the lightbulb, turned it on and let it sit for a while, adjusted the bias for both channels and played some music.
Amp is alive again!
Thanks.
I can’t relate how happy I am to have found some experts on the HK 3490. 🙂
I have one that is in good shape except for an elevated hiss in the left channel. The Pre-out signal is clean, and so is the right channel output, so the problem is somewhere at or after Q301.
Any ideas on what I might be looking for?
I’m in the process of going through each stage with an LM386 signal tracer, but I’m listening for relatively subtle noise.
Any suggestions will be much appreciated.
I have one that is in good shape except for an elevated hiss in the left channel. The Pre-out signal is clean, and so is the right channel output, so the problem is somewhere at or after Q301.
Any ideas on what I might be looking for?
I’m in the process of going through each stage with an LM386 signal tracer, but I’m listening for relatively subtle noise.
Any suggestions will be much appreciated.
I don't know if the signal tracer is the ideal tool for this, an oscilloscope would be better, but you can throw a sinewave through the amp and trace the signal all through the endstage and see where it picks up the hiss.
Could be anything form a bad transistor/resistor/capacitor to a bad solder joint.
Could be anything form a bad transistor/resistor/capacitor to a bad solder joint.
The signal tracer has been useful. I am A-B-ing the right and left channels. Noise is present at Q313 and Q315 so the problem is somewhere before that. I was probing Q309 and Q311 when I managed to probe the power supply and smoke my LM386 amp. Further testing is delayed until a new one arrives from Amazon.
Which brings up another issue. How in the world do you do live testing on the main board? The location of the input and digital boards on the rear chassis make it extremely difficult to access components on the main board. If I disconnect those boards I immediately get a “voltage check” error and the receiver shuts down. I suppose HK had a test setup and longer connecting cables so those smaller boards could be of the way.
On another forum a number of people claim to have successfully eliminated noise by replacing the KTC3200s in the offending channel. I must have researched this before because I have a number of replacement transistors that I ordered from eBay a while ago. They are labeled K 708 C3200GR.
Any comments on using them as replacements? One poster opted for brute force replacement of all four based on the difficulty of removing the main board.
Which brings up another issue. How in the world do you do live testing on the main board? The location of the input and digital boards on the rear chassis make it extremely difficult to access components on the main board. If I disconnect those boards I immediately get a “voltage check” error and the receiver shuts down. I suppose HK had a test setup and longer connecting cables so those smaller boards could be of the way.
On another forum a number of people claim to have successfully eliminated noise by replacing the KTC3200s in the offending channel. I must have researched this before because I have a number of replacement transistors that I ordered from eBay a while ago. They are labeled K 708 C3200GR.
Any comments on using them as replacements? One poster opted for brute force replacement of all four based on the difficulty of removing the main board.
Dumb question.
I’m about to replace transistors.
I’ve got the Main Board removed, mostly, except for BN73. What is it? It’s unlike any of the other connectors. Is it a header connector? It is resisting any attempt to lift it off any header pins. It’s in a difficult spot for applying leverage and since I’m not exactly sure what I’m doing I’m nervous about forcing It.
I can work on the back side of the board the way it is but I’d really like to disconnect that cable. The other end goes to the Phones Board and I can easily disconnect that end, but it looks like it routes through a toroid and I’d prefer not to mess with It.
I’m about to replace transistors.
I’ve got the Main Board removed, mostly, except for BN73. What is it? It’s unlike any of the other connectors. Is it a header connector? It is resisting any attempt to lift it off any header pins. It’s in a difficult spot for applying leverage and since I’m not exactly sure what I’m doing I’m nervous about forcing It.
I can work on the back side of the board the way it is but I’d really like to disconnect that cable. The other end goes to the Phones Board and I can easily disconnect that end, but it looks like it routes through a toroid and I’d prefer not to mess with It.
The traces under the corrosion Q319 were intact.
The left channel noise fix involved replacing the KTC3200 Q301, Q303, Q305 and Q311. The left channel is so quiet now that it revealed a similar but much less noticeable problem in the right channel, but that’s a fix for another time.
The KTC3200 failures are a known issue with this receiver. The spec sheet states a hFE range of 200~700. The replacement transistors tested a hFE of 431. The removed Q301, Q303 and Q305 tested hFE between 60 and 86. Q311 tested 214.
I don’t look forward to taking this receiver apart again, but the process has helped me appreciate the quality of HK design and engineering.
Thanks for sharing your expertise.
The left channel noise fix involved replacing the KTC3200 Q301, Q303, Q305 and Q311. The left channel is so quiet now that it revealed a similar but much less noticeable problem in the right channel, but that’s a fix for another time.
The KTC3200 failures are a known issue with this receiver. The spec sheet states a hFE range of 200~700. The replacement transistors tested a hFE of 431. The removed Q301, Q303 and Q305 tested hFE between 60 and 86. Q311 tested 214.
I don’t look forward to taking this receiver apart again, but the process has helped me appreciate the quality of HK design and engineering.
Thanks for sharing your expertise.