The G22000 came back, but not to my collection.

kevzep

Its all about the Music
Well, I used to own this G22000, and its been away with another owner for 6 years. The guy who owned it, needed to sell and I simply couldn't come up with the funds quick enough.
However, a good friend of mine could, so for the sake of "keeping it in the circle", he bought it, I went and retrieved it and life is good.
I get to spend this week with it, as it needs a bit of work.

As most of you know these things are huge, it is almost too large for my bench, hahaha....
This far I have found and resolved these things.

1, The tuner was so far out of alignment, it gave me a good challenge getting it back, I have a love/hate relationship with Sansui's alignment instructions, sometimes you have to get creative and figure out what their whacky instructions actually mean. Notes are, its pretty much the same tuner as the G9000, same front end, same IF, Multiplex is a bit different and the meter drive for signal and tune are different. It uses the same IF board as the G9000 too.

2, When I had it, I replaced the driver transistors with 30mhz MJE15033/32 due to operator error took one channel out, at the time I never thought about it, but in actual fact the originals, 2SC2238 and 2SA968 are 100mhz devices, so the MJE's limit the performance of the amplifier.
I located some NOS devices and fitted those, dialled in the DC offset and Bias.
I ran it up both channels driven, it makes 244 watts per channel, THD is 0.001% up to just over 220watts and then it creeps up to 0.003%
Pretty impressive.

3, I didn't notice when I originally turned it on yesterday, but there are issues with the protection circuit, there is no time delay when you turn on, no Protector flashing, it just goes straight on, boom!!
Not good!!
So far I have found a dead Zener in the circuit which trips the protection, so its had no current or DC protection, it could never have worked in this condition. When I had it last it was working!!
So I am working through the rest of the issue now that the protection will actually trip which it could not have before.
So its not coming easy finding the issue, but I will find it. This morning I will trip the protection by disconnecting the input to the drive board, this loses its earth reference, and the DC offset slowly drifts up to rather large amounts of voltage, so that will do the trick, once we know its working and the vibrator is working for the protector indicator, and the protection is working and holding the speaker relays off, we can work on this time delay issue...

Af few pic's along the way....

These things are so huge, easy to forget.IMG_6155.JPG IMG_6154.JPG IMG_6151.JPG IMG_6143.JPG
 
So just finishing up here. What a journey that was.
So it turns out, this protection circuit is a bit temperamental, very very touchy. The voltages were off a little bit around the SCR and the Zener for no good reason, everything measured up as it should, but in this one area it was not quite right, and we are only talking 200mV, but it was enough to cause this issue of the protector only coming on for about 100milliseconds.
I could not see where I could possibly get this right at first, all components were correct and in tolerance...
Long story short, I had to modify the circuit a bit, the problem was the voltage was too high on the base of the protection relay driver. It started at 600-630mV which turns on that relay very very quickly, then after the delay cap charges up, the voltage goes up to about 950mV. So when there is a DC event on the output, the sensor diodes turn on the appropriate transistors, which is then supposed to reduce the voltage on the base of the relay driver, but because the voltage is a little high, it doesn't reduce below the 600mV mark which it needs to so it can turn the protection relay off.
So, I had to add another resistor to pull the voltage down a bit, tried to do it a few different ways but on the actual relay driver transistor was the best.
After I had replaced every diode and cap and measured every resistor, I have decided this protection circuit is really crap, and its like my ex wife and her hair trigger!!
I also changed the time delay cap to make the protector kick off after 6 sec, initially it was 12-13 sec which I reckon is a bit longer than it needs to be.

The DC protection kicks in right in the specified 2-3VDC zone at 2.8VDC.

Then!! There's the stupid current protection, that circuit is a complete joke!!
Yes it works, and I was even able to get it to kick in a little later than spec'd, but really, its a waste of time in my opinion.
One transient, which in all honesty is not going to take out your precious output transistors, and the damn thing shuts down, no warning, protector flashing, no audio.
So many amplifiers do not have this and survive for decades with no issues, okay so perhaps if you are a complete buffoon and don't know when you're running out of headroom it could save your ass, but personally, it annoys the hell out of me and I know its going to annoy the owner, so I disconnected the trigger diodes and left it at that.
I can't remember how many threads I've read about the hair trigger current protection on Sansui G's, I remember when Toby sold me this one he said the Current protection was trigger happy, theres a bit thread about it here somewhere.

I'm bench testing this G22000 now, I really like the sound of these amps, they are just great!!!

RIGHT, now where did I put my G9000 I was working on before the G22K turned up!!

Couple of pics......tricky to work on these amps at times, of course the protection relay, and the Vibrator, and associated relay drivers are in the pre-amp, the actual protection circuit is in the power-amp of course, so you have to set things up so you can get at everything you need to......

IMG_6157.JPG IMG_6159.JPG IMG_6160.JPG IMG_6165.JPG IMG_6167.JPG
 
I looked at the protection circuit of the G22/G33K a while ago and I thought it was the most complicated /comprehensive I'd ever seen. Maybe it was just the way the schematic was drawn, but it took me about 20 minutes of hard looking before I gained some understanding of what was going on in that section of the schematic.

Nice work Kev, I'm not above a bit of 'tweaking' myself now, thanks to people like yourself leading the way. ;)
 
I looked at the protection circuit of the G22/G33K a while ago and I thought it was the most complicated /comprehensive I'd ever seen. Maybe it was just the way the schematic was drawn, but it took me about 20 minutes of hard looking before I gained some understanding of what was going on in that section of the schematic.

Nice work Kev, I'm not above a bit of 'tweaking' myself now, thanks to people like yourself leading the way. ;)

Thanks John, well, its comprehensive, it does Thermal, DC, Current, and of course works to have the speaker relays closed at power up and power down.
It does a lot, and in fact the circuit that drives the Current protection is fairly complicated, just like the rest of it.
But its too sensitive, there is no real margin, usually on a circuit like this you'd make it adjustable so you can have it dialled in just right, but there is nowhere to adjust anything!! So a bunch of modern diodes, zeners, transistors in the circuit will quite likely throw it off....
Even down to the MJE150033/32's that you'll use in the regulated supplies that drive this circuit...

I think if you are going to have current protection, it needs to ne in the way of some sort of limiter, so at least you don't lose audio. the way this current protection works is fairly brutal!!

all good fun, 12hrs in the lab never hurt anyone!!
 
I had one of these on the bench with some odd problem in the protection circuit. I think it was the same thing, no delay. On the one I worked on it was missing the ground for the circuit. The wiring didn't match the schematic and I couldn't figure out where the ground was supposed to come from. I think I just jumpered it to a known good ground.
 
Maybe the protection circuit got rattled by the humongous tomo up near Rotorua. Just looking at that massive hole would have scared a few mv out of anyone.

Thanks for the detailed explanation. Being able to fix a beastie like this one that you know so well must put a smile on your face.

Well Done - - -
 
Kev; You Definitely NEED a larger bench. You can have my old dinner table (3' x a little over 6 ft) for shipping.:D:D THAT should handle the BIG "G".

Larry
 
That is quite a unit.
BTW, for drivers these days, I like 2SC4883A/2SA1859A. Mucho faster than MJE15033/32.
Only down side is that they are a full plastic package.
 
Instant on is a very common issue with these. Taking the small toggle power switch on the amp apart, and cleaning has fixed more than one of these with this problem, so another possible fix for others tacking one of these.

Another FYI, this little power switch gets cleaned on every one I have done now, and none of them have ever looked good. Very nasty actually.
 
That is quite a unit.
BTW, for drivers these days, I like 2SC4883A/2SA1859A. Mucho faster than MJE15033/32.
Only down side is that they are a full plastic package.

Thats a good alternative, they don't use the collector on the heatsink as a conductor anyway so all good there, I have noted that down though, in fact I'll just order some up for stock actually, the MJE's are good for power supplies but not really suitable for drivers in my opinion , the NOS 2238/968 's I got were not very expensive so I was happy with that.

I had one of these on the bench with some odd problem in the protection circuit. I think it was the same thing, no delay. On the one I worked on it was missing the ground for the circuit. The wiring didn't match the schematic and I couldn't figure out where the ground was supposed to come from. I think I just jumpered it to a known good ground.

Thats exactly what I thought it was at first. Yes same thing, the diagram didn't quite match, you are on your own in that circuit. After going around in circles for a hour or two I decided I could just make it work my own way...
If it was my own, I would have made it adjustable so you can set it precisely.

Kev; You Definitely NEED a larger bench. You can have my old dinner table (3' x a little over 6 ft) for shipping.:D:D THAT should handle the BIG "G".

Larry
No more room in my lab, what you don't see, is my race car that sits right behind me, so we are all full up!!

You really can't appreciate the size of these things until you see one in person. To give some perspective, here's an old picture of an 8080DB sitting on top of a G22.


View attachment 1182989

- Pete
Absolutely right Pete, even though I had owned this previously, I haven't seen it for 6 years, and it made me take a step back when I went to pick it up the the dude took the cover off....
They are enormous!!
 
I've been bit exactly once on "NOS" semiconductors. From then on, if they don't come from an authorized distributor, I don't buy 'em.
 
Where is the action shot of your load kettles steaming up the place as you coax 244wpc out of that beastie, (for the casual reader Kevzep has found a certain brand of electric kettle makes both a perfect dummy load and a fine cup of tea) :D
 
Where is the action shot of your load kettles steaming up the place as you coax 244wpc out of that beastie, (for the casual reader Kevzep has found a certain brand of electric kettle makes both a perfect dummy load and a fine cup of tea) :D
I really need to do that don't I!!
I've been bit exactly once on "NOS" semiconductors. From then on, if they don't come from an authorized distributor, I don't buy 'em.
Fair enough too, I was going to try it exactly once!! I used commonsense and I most definitely got the genuine example, so totally happy, but it is so easy to get bitten even you are smart and use commonsense...
 
Kev - we've discussed this before, did you do anything with the preamp/main amp slide switch on the side?
I should take mine apart and clean/polish it but that means I'd have to move it and my back isn't ready for that!
 
Kev - we've discussed this before, did you do anything with the preamp/main amp slide switch on the side?
I should take mine apart and clean/polish it but that means I'd have to move it and my back isn't ready for that!
Ah yes, we did talk about that didn't we.....No I didn't even need to touch this one, its mint. So I have left it. Since I have a G9000 now, I will have a look when I finish off the restore I am doing, I reckon we'll be able to replace the switch with something modern..
When I had this G22000 6 years ago I cannot recall ever having an issue with it either..
 
Back
Top Bottom