G-9000 switches and such
- Thread starter JimBlackwood
- Start date
JimBlackwood
New Member
Thanks for the suggestion.
JB
JB
JimBlackwood
New Member
Back again. Looks like some more repair work in my future. The protection circuit has been acting up for some time, or perhaps just doing it's job and now it doesn't turn on any more. Am I right in thinking the place to start is with checking the DC offset? I've not done that, is it basically just a matter of measuring the voltage across the speaker terminals?
Jim
Jim
kevzep
Its all about the Music
Yes, the first thing to do is check the DC offset, but you have to measure this before the speaker relay.
You will need to take the bottom off the unit, and you will see the board with the relays on it.
I will see if I can dig up a picture and point out where you need to look....
After you get a reading from the output we can dig a little deeper into some diagnoses....
I had an interesting fault on mine recently (protection) which appeared right after I sold it....
JimBlackwood
New Member
Right, that makes sense. Obviously would have been easier if I'd done it while it would still kick on. Well I guess that gives ma a good excuse to go ahead and replace the light bulbs. Maybe there are some warm white LED's available now that can be made to work...
Jim
Jim
kevzep
Its all about the Music
Its not too difficult to find where the amplifier outputs go into the relay's, and once you do and get a measurement we will know where we are at....
I have a feeling it could be something fairly straight forward...
It doesnt have any blown fuses right? If not I doubt whether it has blown output transistors or anything serious like that....
JimBlackwood
New Member
I have the covers off but I have to work myself into the proper frame of mind to start on this, especially mucking around with the power on. But as luck would have it I also just recieved an assembly kit for a megashift transmission controller which should help cultivate the desired mental state. Really can't afford any stray motions in the neighborhood of large charged storage capacitors and such and my hands are not the steadiest. I will get it but it may take me another day or two.
Jim
Jim
JimBlackwood
New Member
Just put the meter on it and got some strange results. I connected to the relay leg of R61 and R62 and the common chassis ground lug. There is a fluctuating signal on the line, looks to be about 10 hz or so with an amplitude of about 5 mv. On power up the DC level jumps to about 120 mv then drops to less than 5 mv and after 10 seconds jumps up to about 45 mv on R62. After switch off it fluctuates, abruptly rises to 130 mv and then the voltage drops off.
On R61 it spikes to about 75 mv then drops to 40 mv and stays there. On power down it rises and then spikes to 100 mv before dropping off.
I haven't checked fuses yet but it seems maybe I should try to fix this first.
Jim
On R61 it spikes to about 75 mv then drops to 40 mv and stays there. On power down it rises and then spikes to 100 mv before dropping off.
I haven't checked fuses yet but it seems maybe I should try to fix this first.
Jim
kevzep
Its all about the Music
Are you sure you are on the actual output of the amp before the relay?
I wouldnt worry about the voltage fluctuating like that, as you power up the offset will take time to stabilize as the circuits become energized.
It takes about 2VDC to trip the DC protection, so we first need to establish you are actually connected to the right place, if you are then those readings are not the problem, there will be something else tripping the protection...
JimBlackwood
New Member
Hey guys, been away for awhile (lost connectivity in the shop, tried wireless repeaters and finally bit the bullet and laid a cable among other problems, let's just assume a more or less complete melt-down) Presently listening to Pandora through the laptop on small desktop amped speakers to give you a clue to how bad it's been. Didn't even have that until today.
SO. There sits the G9000 under a t-shirt to keep the dust off. Totally wrong frame of mind for tackling that problem and not sure when the pendulum will swing back. Therefore the question is, can I find anyone within a reasonable distance who can help me out here? Anybody got any ideas? Color me clueless I guess.
Jim
SO. There sits the G9000 under a t-shirt to keep the dust off. Totally wrong frame of mind for tackling that problem and not sure when the pendulum will swing back. Therefore the question is, can I find anyone within a reasonable distance who can help me out here? Anybody got any ideas? Color me clueless I guess.
Jim
JimBlackwood
New Member
I'm located in Florence, KY. Guess I should put that on my profile. I'd love to find someone in the Cincinnati area.
But I'm guessing I will have to bide my time until the mood strikes me to dig into it again. The good news is, that I did get one of those pocket storage scope/signal generators so once I have time to learn how to use it that should be a big help. It all seems to be menu based.
Jim
But I'm guessing I will have to bide my time until the mood strikes me to dig into it again. The good news is, that I did get one of those pocket storage scope/signal generators so once I have time to learn how to use it that should be a big help. It all seems to be menu based.
Jim
JimBlackwood
New Member
So it's been awhile but I've made some progress. I tested and reseated the final drivers with new thermal grease then popped the covers off the relays and tried cleaning them with a burnishing tool. Not too successful I'm afraid, too fine of an abrasive. and poor access. I finally figured out that the springs could come off and then the relay armatures could wiggle out and give good access to all of the points. So with some crocus cloth glued to a popsicle stick I got them all cleaned up nicely.
Next is the speaker output adjustment. The POST came up in 7 seconds but unfortunately I decided to power down and clean the 3 pots with fader cleaner and turning them to get the dead spots, which of course threw out the adjustment and undoubtedly put voltage on the outputs and now the POST won't come up, meaning the speaker relays won't pull in. I guess I need to pick up the outputs somewhere between the relay contacts and the drivers and adjust the voltage and balance that way. Guess I need to look at the schematic.
Anyway before I screwed up the adjustments I had just a bit of voltage on the left speakers, didn't check the right yet. When I went to adjust it, it went scratchy on me hence the cleaning job. But I did get it to go down to zero first so there's that. I suspect it'll be OK so get that done and then have a look at the balance control and see if there's still a problem.
What's the latest/greatest on light bulbs? Any proper colored LEDs yet?
Jim
Next is the speaker output adjustment. The POST came up in 7 seconds but unfortunately I decided to power down and clean the 3 pots with fader cleaner and turning them to get the dead spots, which of course threw out the adjustment and undoubtedly put voltage on the outputs and now the POST won't come up, meaning the speaker relays won't pull in. I guess I need to pick up the outputs somewhere between the relay contacts and the drivers and adjust the voltage and balance that way. Guess I need to look at the schematic.
Anyway before I screwed up the adjustments I had just a bit of voltage on the left speakers, didn't check the right yet. When I went to adjust it, it went scratchy on me hence the cleaning job. But I did get it to go down to zero first so there's that. I suspect it'll be OK so get that done and then have a look at the balance control and see if there's still a problem.
What's the latest/greatest on light bulbs? Any proper colored LEDs yet?
Jim
JimBlackwood
New Member
OK I'm a little confused I guess. First pot is coarse adjust, 2nd part is fine, and 3rd is bias voltage at TP-+ and - which as far as I've been able to tell is a terminal strip at the top right rear corner with two wires attached to it. Correct, or not?
So I adjusted the voltage on the left speaker terminals to zero. No issue there. But then the voltage on the right speaker terminal was about 0.35v which is outside the specified limit by quite a lot. Also those two pots don't change the voltage on the right side. What's up with that?
Then the voltage on TP + and- didn't show anything and that 3rd pot didn't change it.
Jim
So I adjusted the voltage on the left speaker terminals to zero. No issue there. But then the voltage on the right speaker terminal was about 0.35v which is outside the specified limit by quite a lot. Also those two pots don't change the voltage on the right side. What's up with that?
Then the voltage on TP + and- didn't show anything and that 3rd pot didn't change it.
Jim
kevzep
Its all about the Music
1: Don't bother wasting your time cleaning the contacts in relays, you need to replace them, the contact coating wears off over time and exposes the raw material which oxidises in no time. Drop in replacements are cheap and readily available.
2: What do you mean by "Post"? I have no idea what you are referring to.
3: Replace the trimmers, don't waste your time trying to clean 30-something year old trimmers, just slip some Bournes multi-turn in there. The course DC offset adjust needs a single turn trimmer.
4: You can buy drop in LED replacements djwojo has them. https://sites.google.com/site/dgwojocom/
5: The test point for Bias adjustment is the terminal strip you are referring to this is correct. You are measuring for 16mVDC (0.016VDC) at this point. You adjust VR03 to achieve this, anticlockwise INCREASES bias, Clockwise DECREASES bias. VR01 and 02 have nothing to do with this test point.
6: You need to connect to the speaker output to measure DC offset. If the protect relay is not engaging the speaker relay, you need to measure BEFORE the relay. You then use the coarse adjust to get as close as you can, then use the fine adjust to fine tune it.
7: If VR03 does not change the bias, then there is a problem. Usually this symptom is because fusible resistors R17 and R20 on the F2806 board have gone HI Resistance and are not allowing current to flow through the bias circuit.
8: If you have 350mV on the speaker output, then you have an issue in the front end of the driver circuit, it could be anything from a leaky capacitor to a failing Zener diode, so this will require a bit of troubleshooting.
Let us know how you get on...
2: What do you mean by "Post"? I have no idea what you are referring to.
3: Replace the trimmers, don't waste your time trying to clean 30-something year old trimmers, just slip some Bournes multi-turn in there. The course DC offset adjust needs a single turn trimmer.
4: You can buy drop in LED replacements djwojo has them. https://sites.google.com/site/dgwojocom/
5: The test point for Bias adjustment is the terminal strip you are referring to this is correct. You are measuring for 16mVDC (0.016VDC) at this point. You adjust VR03 to achieve this, anticlockwise INCREASES bias, Clockwise DECREASES bias. VR01 and 02 have nothing to do with this test point.
6: You need to connect to the speaker output to measure DC offset. If the protect relay is not engaging the speaker relay, you need to measure BEFORE the relay. You then use the coarse adjust to get as close as you can, then use the fine adjust to fine tune it.
7: If VR03 does not change the bias, then there is a problem. Usually this symptom is because fusible resistors R17 and R20 on the F2806 board have gone HI Resistance and are not allowing current to flow through the bias circuit.
8: If you have 350mV on the speaker output, then you have an issue in the front end of the driver circuit, it could be anything from a leaky capacitor to a failing Zener diode, so this will require a bit of troubleshooting.
Let us know how you get on...
JimBlackwood
New Member
Thanks for the tips. I will see what I can do with the cleaned relays and then next time they get dirty I'll replace them. That way I'll have time to acquire the replacements and can keep going with the troubleshooting. They have silver contacts, which grow an oxide layer. I can't match the factory micro-polish but they will be good for awhile anyway. You are right, there is no substitute for new relays. Maybe I can have relays in hand by the time I'm finishing up.
On the VR01/VR02 adjustment. Do you perhaps measure the voltage between the + side of the left and right channels? I was measuring to ground. The voltage on the left channel to ground can easily be adjusted into that (350mv) range with VR01. So it seems to me that if the measurement is from speaker + to ground then each side should have its own trimmer pots rather than just one set for one side. The only thing that makes sense to me in this scenario is that you would measure from left speaker + to right speaker + and adjust for zero. Maybe there is some leakage creating that voltage to ground. But I'll try it that way and my best guess is that I'll have about 350mv to ground on both sides (left+ and right+).
If that is not correct, do you have any further suggestions? (350mv may be above the permissible POST limits)
POST is Power On Self Test. I have it coming up in about 7 seconds now and pulling in the relays.
I sent an email to djwojo about the lamps.
I got the trimmers to adjust, gave them a shot of fader lube and that seemed fine. But, I'll keep your suggestion in mind.
I will look at fusible resistors R17 and R20 on the F2806 board as you suggested and report back. That's a very good tip. I'm just guessing this is the bias on the output drivers so it would affect the output level and distortion? I didn't expect to see an adjustment there. Guess a little more quality time with the schematic wouldn't hurt. Sounds like fun-to-find components.
I'm having a lot more fun with this than with sorting out my network problems I must say. Nowhere near the same level of frustration. It's good to be making progress, even if slowly.
Jim
On the VR01/VR02 adjustment. Do you perhaps measure the voltage between the + side of the left and right channels? I was measuring to ground. The voltage on the left channel to ground can easily be adjusted into that (350mv) range with VR01. So it seems to me that if the measurement is from speaker + to ground then each side should have its own trimmer pots rather than just one set for one side. The only thing that makes sense to me in this scenario is that you would measure from left speaker + to right speaker + and adjust for zero. Maybe there is some leakage creating that voltage to ground. But I'll try it that way and my best guess is that I'll have about 350mv to ground on both sides (left+ and right+).
If that is not correct, do you have any further suggestions? (350mv may be above the permissible POST limits)
POST is Power On Self Test. I have it coming up in about 7 seconds now and pulling in the relays.
I sent an email to djwojo about the lamps.
I got the trimmers to adjust, gave them a shot of fader lube and that seemed fine. But, I'll keep your suggestion in mind.
I will look at fusible resistors R17 and R20 on the F2806 board as you suggested and report back. That's a very good tip. I'm just guessing this is the bias on the output drivers so it would affect the output level and distortion? I didn't expect to see an adjustment there. Guess a little more quality time with the schematic wouldn't hurt. Sounds like fun-to-find components.
I'm having a lot more fun with this than with sorting out my network problems I must say. Nowhere near the same level of frustration. It's good to be making progress, even if slowly.
Jim
JimBlackwood
New Member
Well... I suppose that might have put paid to any hopes of claiming genius level brilliance then.
I just never stopped to think that the two channels might still be on opposite sides of the chassis at that point. Yeah, I don't know how either.
So... I have an entirely new appreciation for the nexus of corrosion and old trimmer pots but as it stands I do have the voltage down in the single digit millivolt range on both sides. It tends to jump around and was an absolute bear to adjust but last I looked it was holding at about 1 or 2mv. And the bias on the left side is 16.6mv, that from an extremely touchy trimmer. Sad to say, the right side is still sitting on zero. POST comes up at 7 seconds, consistently.
I'm just wondering if that might not be the origin of my longstanding issue with the balance knob on the front panel. Seems reasonable. If so, there might be very little else needed with this unit. Well, I mean other than jumping in and replacing all the caps and trimmers and whatever else age has destroyed.
Oddly the right side had a cover on the box, the left did not. Maybe they ran out. Doesn't look like any screws were ever in the holes. QC at it's best in the 1970's.
Again, I greatly appreciate the help. Now to track down those pesky fusible resistors (R17 & R20). Any identifying characteristics I could look for?
Jim
I just never stopped to think that the two channels might still be on opposite sides of the chassis at that point. Yeah, I don't know how either.
So... I have an entirely new appreciation for the nexus of corrosion and old trimmer pots but as it stands I do have the voltage down in the single digit millivolt range on both sides. It tends to jump around and was an absolute bear to adjust but last I looked it was holding at about 1 or 2mv. And the bias on the left side is 16.6mv, that from an extremely touchy trimmer. Sad to say, the right side is still sitting on zero. POST comes up at 7 seconds, consistently.
I'm just wondering if that might not be the origin of my longstanding issue with the balance knob on the front panel. Seems reasonable. If so, there might be very little else needed with this unit. Well, I mean other than jumping in and replacing all the caps and trimmers and whatever else age has destroyed.
Oddly the right side had a cover on the box, the left did not. Maybe they ran out. Doesn't look like any screws were ever in the holes. QC at it's best in the 1970's.
Again, I greatly appreciate the help. Now to track down those pesky fusible resistors (R17 & R20). Any identifying characteristics I could look for?
Jim
kevzep
Its all about the Music
Okay
1: There is no power up self test on this amplifier. The power up procedure has the amplifier in protection whilst the various voltages stabilise, then the protection relay engages which in turn then turns the speaker relay/s on.
It basically has a simple circuit with a Zener and a Cap, the cap charges up, and then turns a transistor on which triggers the relay.
You need to stop using this "post" term as it is confusing because the amplifier does not have this feature.
2: Are you using the service manual? Everything you are trying to do is clearly laid out in the service data, there are even clear diagrams.
3: R17 and R20 are clearly labeled in the screen printing on each driver board. They are 150Ω 250mW resistors, you should just measure them and check they are in spec. They are no at all difficult to find, just pull the driver board out and you'll see them.
3: There are two driver boards, it is a stereo amplifier. One driver board has one side missing on the cover, the other one has a full cover, Sansui must have decided the shielding was not necessary on one of them.
4: If you read the service manual, it clearly states in the protection circuit operation that the DC protection will kick in at 2~3 VDC.
5: It is completely normal for the DC offset to leap around, you'll never ever get it to sit at zero mV.
Its the nature of the thermal tracking in the dual JFET.
What else has age destroyed?
Anything and everything is suspect, I had a fault on my G9000 which was a part on the driver board you would never expect to fail....
You need to be thorough in diagnoses.
*******If you pull the driver boards out, you MUST and I repeat MUST be VERY VERY careful reinstalling them, it is very easy to plug them back in a whole row out on the pins, so you need to plug it in, and you can just see through the side of the chassis to check it is correct, using a torch*******
JimBlackwood
New Member
More relevant info. The right channel bias voltage is cycling on and off, 8 seconds on and 27 seconds off. I switched the two boards side to side with no significant differences so it is definitely in the board since it traveled with it. Guess I'll try the IR temp pistol on it next. I'm wondering if one of the bias diodes might have opened. I'll test those two resistors.
I have a copy of the manual. It isn't the best but is usable.
Jim
I have a copy of the manual. It isn't the best but is usable.
Jim
Last edited:
JimBlackwood
New Member
Is there any sort of a riser available for these driver boards? Sure would make them easier to work on. Or, if I knew wher to buy matching connectors I could make up a set of extender cables which would help immensely. I've been stuck because what I really need to do is compare voltages at various points but I can't get at them with the boards installed.
Jim
Jim
