Super sophisticated is what it looks like. When the wire at the left sees 5 volts the LED glows. The LED is instant response so you can see short duration pulses.
That could be the pin going high briefly and that would release the relay and mute the signal.
The devil is in the detail in all of this, it really is 🙂
When you say these two things they seem to contradict each other 🙂 Do you see. We have to be absolutely clear what is happening.
------------------------------------
That sounds OK. It rises slowly because the resistance feeding the cap is very high (R13).
Look at IC1 and try and understand the flow of voltage.
Pin 1 is shown at 4 volts which is a High. Pin 2 and pin 3 should be Low. Pin 4 should be High. Those levels should be a constant and only change if the voltage on C19 drops. So these pins are worth checking.
Also look at pins 8, 9 10 and 11. There looks to be an error in the voltages here on the diagram. Zero volts on pin 2 (that looks correct) is then shown as being 5 volts on pin 11, the same point. So pin 11 should be zero. Pin 10 and 9 should be High and pin 8 Low.
The takeaway point from that is pin 8 should be the same as pin 11 at all times. Connecting gates in series like this is common practice to 'use up' spare gates in a package.
This could be a worthwhile test
If you remove R33 and then connect a resistor (any value from 1k to 100k will do) from pin 5 (or the collector of Q10) to pin 14 we 100% remove all the mains detection stuff and that should help eliminate that area of the circuit. The resistor is just used as a collector load or 'pull up' resistor to take pin 5 high yet allow Q10 to remain in control.
The 4069 is a common chip that cost buttons but its not the first suspect. Do you have any freezer spray? If not a can of cheapo air duster which is butane based can be used. Hold the can upside down and drip the stuff on the chip and other suspect parts to see if it changes the fault.
I think I've got all the levels right in describing all that 🙂
One more question 🙂
When it works normally for about how long does it mute for when you change program. This will be the duration Pin 6 goes high for or the time the voltage across the relay coil disappears for.
When it works normally for about how long does it mute for when you change program. This will be the duration Pin 6 goes high for or the time the voltage across the relay coil disappears for.
Many thanks to both of you! I admire your patience with me, my english and my knowledge is probably not the best when it comes to electronics, but I learn something new every day.
Hmm, pin 6 only goes high when i force it by shorting C and E on Q10. Otherwise it´s 0,067, except when i change program, then the DMM briefly shows O.L. as i said earlier. No contradiction?
I think i have a LED and a resitor somewere and will give it a try.
And i will check all pins on the IC.
Hmm, pin 6 only goes high when i force it by shorting C and E on Q10. Otherwise it´s 0,067, except when i change program, then the DMM briefly shows O.L. as i said earlier. No contradiction?
I think i have a LED and a resitor somewere and will give it a try.
And i will check all pins on the IC.
This test should be fun!
There is no particular delay that I noticed, the programs are changed immediately.
(The only delay for mute is at switch (or remote) power off, the times the relay stays open, when it takes a couple of seconds before the relay mutes, and this is the problem).
Ok, here are the pins on IC1, #1 4.40 V, #2 0 V, #3 0 V, #4 4,26 V, #5 4,26 V, #6 0,064 V, #7 0 V, #8 0 V, #9 4,25 V, #10 4,25 V, #11 0 V, #12 0 V, #13 4,25 V and #14 4,25 V.
Lets just clear that up first.
Although the programs seem to change immediately I thought the relay muted each time you did change programs. I based that on this where you mention it clicking each time you change program:
----------------------------------------------------------------------------------------------------
So lets change our attention to this. The non muting at power off.
Can you measure this voltage on Q10 base and see if it behaves differently when in the working and faulty state at the instant you power off. If this voltage goes high it will instantly remove power to the relay and mute the signal.
If you can recreate the fault when putting the unit in standby then definitely test it in that mode of operation. We need to see if this control voltage behaves differently when working and when faulty.
I'll look at the all those reading on the IC in a little while 🙂 It is quite an interesting fault.
Those show no problems at this point. Each input and output is in a valid state. That means 1 in gives 0 out and vice versa.
When i power on, the voltage at Q10 base is 0, then after 9 seconds the relay opens base of Q10 is 0, then switch power off and relay closes, still 0 volts at the base of Q10.
Power on again, now waiting for no muting by the relay when i change program before power off when the relay mute is delayed a couple of seconds for the so far unknown reason.
This isn't making sense to me.
Where is this 9 second delay being generated? The only place that makes sense is from the LC7818 chips but you would see that voltage on Q10 base change.
The input to Q10 is the only way that the relay driver levels can be flipped. We have measured the other input which is the AC detect voltage which is pin 1 of IC1 and we are disregarding that having tested that area of the circuit.
You have to find 'what changes' in terms of voltage inputs to IC1 to prove to yourself how the relay is driven. Q10 has to be key to all this as this is what turns the relay driver on or off.
Where is this 9 second delay being generated? The only place that makes sense is from the LC7818 chips but you would see that voltage on Q10 base change.
Do you see the voltage on pin 21 of each LC7818 generate this 9 second delay, You will have to measure on both IC's in turn and monitor continuously during this period and see if the voltage flips level in time to this 9 second delay.
The input to Q10 is the only way that the relay driver levels can be flipped. We have measured the other input which is the AC detect voltage which is pin 1 of IC1 and we are disregarding that having tested that area of the circuit.
You have to find 'what changes' in terms of voltage inputs to IC1 to prove to yourself how the relay is driven. Q10 has to be key to all this as this is what turns the relay driver on or off.
I measured a bit this morning and found pin 21 on IC755 is 4.6V directly at power up. Changing program was 4.7V with clicking RL1.
Pin 21 on IC758 is 0.1V directly at power up. Changing program was 0.2V with clicking RL1.
I also measured the base of the Q10 and at a more sensitive measuring range on my Fluke, I saw 1.2V that came right at power up.
When changing programs I saw 2.3V at the base of Q10. I don't know, but the delay of RL1 on startup doesn't seem to depend on any of the above, instead it seems to be something after Q10?
That there is a delay of RL1 is probably due to the circuits having to stabilize before the signal is sent out on the outputs.
There is also a delay of the outputs on the Citation 22 output stage by a relay for the same reason, which is connected with a "remote control" of the switch-on via a cord between the devices.
The problem with this is, as previously described, that if RL1 does not close immediately when switching off/standby mode, it gives a signal to the output stage which has not yet reached standby mode and there will be noise in the speakers. Hope it is possible to understand some of this. Thanks for your time.
Pin 21 on IC758 is 0.1V directly at power up. Changing program was 0.2V with clicking RL1.
I also measured the base of the Q10 and at a more sensitive measuring range on my Fluke, I saw 1.2V that came right at power up.
When changing programs I saw 2.3V at the base of Q10. I don't know, but the delay of RL1 on startup doesn't seem to depend on any of the above, instead it seems to be something after Q10?
That there is a delay of RL1 is probably due to the circuits having to stabilize before the signal is sent out on the outputs.
There is also a delay of the outputs on the Citation 22 output stage by a relay for the same reason, which is connected with a "remote control" of the switch-on via a cord between the devices.
The problem with this is, as previously described, that if RL1 does not close immediately when switching off/standby mode, it gives a signal to the output stage which has not yet reached standby mode and there will be noise in the speakers. Hope it is possible to understand some of this. Thanks for your time.
This is proving a very difficult one to track isn't it 🙂
Can you answer this one... from a cold start when you turn the mains on, wait a couple of minutes and turn off does the relay behave correctly?
----------------------------------------------------------------
Pin 21 is labelled 'mute' and the two chips have their mute lines resistively mixed together by two resistors.
The manual seems to show a few different versions of the same things.
This is the same but slightly different. The chip numbers change as well.
There is nothing after Q10.... what happens on pin 5 of IC1 is the only control signal for the relay.
--------------------------------------------------------
Can you answer this one... from a cold start when you turn the mains on, wait a couple of minutes and turn off does the relay behave correctly?
----------------------------------------------------------------
Pin 21 is labelled 'mute' and the two chips have their mute lines resistively mixed together by two resistors.
The manual seems to show a few different versions of the same things.
This is the same but slightly different. The chip numbers change as well.
That sounds very plausible. It is 2.3 volts because of the action of those two resistors that mix the two mute signals together and the action of the resistive input of the so called 'self biasing' Q10 which has built in resistors.
There is nothing after Q10.... what happens on pin 5 of IC1 is the only control signal for the relay.
--------------------------------------------------------
This is very clear and needs no further explanation to most old fashioned troubleshooters I think as this used to be standard in about every Japanese built audio device. Maybe it helps to sum it up for you:
1. At power on a DC pulse occurs because of charging of caps, unequal timing of positive/negative rail coming up. The DC pulse causes plops at the outputs that are amplified. The outputs are therefor shorted ("muted") to GND and are opened with a delay as things settle within seconds. Sometimes the inferior method of series contacts is used so the outputs are then switched closed. Then the relay contacts are in the signal path and can influence the signal quality.
2. At power off it is again a DC pulse but now because of discharging of caps and also unequal dying of positive/negative rails. To prevent this the outputs should be shorted to GND immediately. Shorted output to GND = no possible DC pulse or whatever signal. The inferior method of series contacts is then to switch contacts "open" sometimes leading to slight hum.
3. Third category of output muting is when switching stuff or like in DACs when bit errors occur so to prevent amplification of undesired signals that are not music but crackle, noise, static, switches contact bouncing noise etc.
Last edited:
Ok, i just got back to the IC1 did measure pin 5 with the Fluke set to lower range.😳What i found was 0V until i changed program and just as the relay clicks i saw 4,5V!
At pin 6 i found 3,2V until i changed program then it fall to 0,3 or less!
Now that i found pin 5 and pin 6 Ok i think there has to a problem inside the relay, considering your latest reply🤔
Can't be as you replaced it.
All works OK but with intermittent error so it must likely be a part causing delayed signaling/switching (which you can hardly notice with regards to correct timing as it is probably less than a second too late). Not parts that have been replaced already 🙂
With delay one thinks of capacitance. You don't need much in a high ohmic and CMOS environment.... so I vote for IC1. BTW is the board clean? No debri?
All works OK but with intermittent error so it must likely be a part causing delayed signaling/switching (which you can hardly notice with regards to correct timing as it is probably less than a second too late). Not parts that have been replaced already 🙂
With delay one thinks of capacitance. You don't need much in a high ohmic and CMOS environment.... so I vote for IC1. BTW is the board clean? No debri?
Last edited:
This is why I think LED's would help. They show changes the meter might miss or misinterpret.
It may be worth changing the 4069 which are a common part and should be readily available. If only to prove eliminate it.
It would be worth trying the freezer spray (cheap air duster upside down) that I mentioned earlier.
It should be noted that the relay was replaced with a NOS 221D024 (w/o -P) a couple of years ago when the original relay failed. So it did it´s job until recently. Maybe these relays aren´t the best quality parts one would expect?
Anyways, since nothing after Q10 could cause this issue the only thing faulty would be the NOS replacement relay and since i had another NOS relay i replaced the replacement. Now the amp has been on 12 hour and the relay is still working, so i guess jobs done. Again, thanks to both of you giving me new insights in the electronics world, it´s been educational.
An anecdote: during the downtime of the Cit 25 i have listened to my HK930 and it sounds SO GOOD that i'm hesitant to put it back in the wardrobe!
This is where the LED across the coil might help. You would know with certainty whether the relay was hangining on at power off when it should not.
This was suggested very early in the thread in post #9 and 11 as it is most logical but no reply came (indicating it was replaced years ago). It was also not mentioned to have a relay fault years ago which would have been a dead giveaway to everyone involved. The 221D024 is known to be a crappy relay best replaced for a better type. There are even relay adapters for it too. So we lost time and you lost time. What was the cause for time loss?
Replacing it for again the same crappy NOS relay type will cost again time when that one fails. That words "NOS" and "original" were probably where was focused at. Well there exist enough NOS parts that are better forgotten and recycled. Shall we already reserve time for a new thread in 2028? 🙂
Last tip: never use any "open" relay type in audio, always choose fully sealed ones. Gold contacts for signal, silver for current (the famous Japanese error).
Replacing it for again the same crappy NOS relay type will cost again time when that one fails. That words "NOS" and "original" were probably where was focused at. Well there exist enough NOS parts that are better forgotten and recycled. Shall we already reserve time for a new thread in 2028? 🙂
Last tip: never use any "open" relay type in audio, always choose fully sealed ones. Gold contacts for signal, silver for current (the famous Japanese error).
Last edited: