Sansui G-22000 Restoration/Upgrade Thread
- Thread starter mattsd
- Start date
-
- Tags
- g-22000 recap restoration sansui
Interesting question.
kevzep
Its all about the Music
My G22000 has MJ15024/25 in it anyway, so I'll add another pair of those....
kevzep
Its all about the Music
I might have argued to keep it original, but I hadn't owned one at the time and didn't realise what a pain that protection kicking in early was.
But having one now, I can't see the harm in making some discrete mods that improve the performance...
The G-33 uses 4 pairs of output transistors for each channel. The G-22 uses the same board with just 3. That's reason there is space for additional pair of output transistors. Sansui did this on other models as well. The 8080/9090 series comes to mind.
- Pete
stopkidding
Well-Known Member
Is this type of mod generic to all amplifier or requires a certain topology? I am assuming this won't be possible on a bridged amp like AU D11 mk2? And what do more outputs per channel do for the sound?
I was reading about a Dan D'Agostino power amp that had 24 output transistors per channel!!
I was reading about a Dan D'Agostino power amp that had 24 output transistors per channel!!
Much more difficult to accommodate the extra transistors because of the internally bridged design. The G-22000 uses 3 pairs of OP transistors per channel - so you can add another pair per channel with advantage quite easily, especially as the physical heat sinking allows it. But with the AU-D11 II, which has 2 pairs of OP transistors per channel (1 pair in each bridged amp) you would need to add another 2 pairs to each channel - thus doubling the total number of OP transistors. Obviously space to accommodate them is going to be a problem in this case.
Hard to say, what you get is the upgrade transistors sharing some of the load, the amplifier being able to deliver more current as a result. Obviously this requires more current from the power supply, so upgrades to main PSU filter capacitors will help up to a point, as the transformer will usually be overspec'd enough to deal with the extra demand. The individual OP transistors will be less stressed, but total heatsink capacity will begin to be a factor with prolonged high power output.
As I understand it the lower the 'OP transistor count' the better - I.O.W a single OP pair per channel would be ideal. More OP transistors satisfy the requirement for more output power and perhaps reliability for very high power amplifiers. However single devices with enough grunt to supply the required power for such amplifiers are either too expensive, are not practical, or don't exist.
Last edited:
Oh my word yes it does!! I had the same experience, I thought something was wrong with my Distortion analyser, I actually checked it as I could not believe what I was seeing.....Its an extraordinary tuner....easy to align.....I like the way the Zero adjust works on the tune meter, "move the stub".
Yeah agreed, the rest is pretty easy and it responds well to adjustments...
Awesome write up Matt.....
I am working on a G-22000 and came across the term "move the stub". Are they talking about the actual meter, making an adjustment to it? Actually the entire first section of the alignment is a bit confusing. Normally I would use the discriminator coils to center the tuning needle and get the lowest distortion. How did you do it?
kevzep
Its all about the Music
There is a short piece of wire with a piece of heatshrink over the end, from memory I think its a yellow piece of wire, you move it around until the Meter is Centred.I am working on a G-22000 and came across the term "move the stub". Are they talking about the actual meter, making an adjustment to it? Actually the entire first section of the alignment is a bit confusing. Normally I would use the discriminator coils to center the tuning needle and get the lowest distortion. How did you do it?
It shows you where it is on the location diagram of the coils etc etc......
When you find what you think it is, post a picture and I can confirm, but you'll know as you'll be able to centre the meter.
Thanks for the tip.
After looking at the service manual I can see the "stub". It's just a small piece of brown wire about 1 1/2 inches long with a fold of friction tape over the end of it.
At the moment I am just evaluating the unit. I plan on doing a tests on it before and after alignment. Then I will replace the caps and align it again and repeat the evaluation.
After looking at the service manual I can see the "stub". It's just a small piece of brown wire about 1 1/2 inches long with a fold of friction tape over the end of it.
At the moment I am just evaluating the unit. I plan on doing a tests on it before and after alignment. Then I will replace the caps and align it again and repeat the evaluation.
kevzep
Its all about the Music
Thats the one, interesting, the one in my unit is yellow, makes it easy to find.....they must have just used whatever off-cuts they had lying around maybeThanks for the tip.
After looking at the service manual I can see the "stub". It's just a small piece of brown wire about 1 1/2 inches long with a fold of friction tape over the end of it.
At the moment I am just evaluating the unit. I plan on doing a tests on it before and after alignment. Then I will replace the caps and align it again and repeat the evaluation.
All good though, these tuners are exceptionally good, incredibly low distortion....
It will be interesting to see your findings....
Sergimm
Active Member
I find this replacement very interesting, and it looks like an improvement on efficiency, given that it substitutes two components by one. For somebody without advanced knowledge of electronics, is there a way or “rule of thumb” to identify such double-cap setup to simulate a bipolar? Are they always replaceable by a single capacitor (size permitting).
When you see two electrolytic capacitors in series, connected '+ to +' this is simulating a single bipolar electrolytic capacitor. (rated at double the voltage and half the capacitance of the individual capacitors, so if you have a 10µF + 10µF combination, each rated at 25Volts to make up a bipolar electrolytic the resulting value is 5µF and voltage rating of 50Volts).
Now however, high quality 'single can' bipolar electrolytic capacitors are available in a range of values suitable for use in these situations. If the value of the combination is low enough, you can use a film capacitor to replace the electrolytic 'double' or 'single' bipolar, i.e. less than say, 5 or 10µF. However this should not be taken to extremes, size is a big consideration (pun not intended), and the replacement should fit comfortably in the space occupied by the original, not impede airflow, and not strain the tracks of the board due to excess weight. If the replacements are getting too big - to the point that their leads need to be sleeved to prevent short circuits, or they need to be supported while you solder them, this is the point to reconsider whether a single quality bipolar electrolytic is a better choice.
If the original bipolar electrolytic was also bypassed by a low value film capacitor, (0.047µF ?) this can usually be removed, thereby giving even more space for a suitable replacement for the entire combination. Removal (of the bypass), is recommended particularly (obviously?) if the bipolar replacement is a film capacitor.
Last edited:
Sergimm
Active Member
That makes sense, thanks for the clear explanation!
Why is the bypass there in the first place? What makes it "removable" once we have a single capacitor?
Why is the bypass there in the first place? What makes it "removable" once we have a single capacitor?
The bypass is used to even out the frequencies that are passed (helping with the higher frequencies), which has the effect of overriding the less desirable characteristics of the 'electrolytic pair' bipolar.
Last edited:
I can't help you with measurements, but I am sure in a restore thread somewhere the diameter and height are noted. Regarding value increase, there is no need to increase the capacitance value. However it might prove useful to upgrade either the value or the working voltage of the replacements slightly to increase the physical size of the replacements for a better fit. As you may know today's replacement capacitors tend to be a bit smaller than the originals - so a physical size increase can make for a 'better fit'. I think I have seen units with 22,000µF replacements, I am not sure what rectifiers are used, but an increase such as this should prompt a review of the rectifier capacity.
You might also want to increase the value slightly because new capacitors are tending to come out on the low side of the tolerance band capacitance value-wise. So where you might buy a 15,000µF +/- 20% capacitor, it will be nearer the lower tolerance limit rather than the higher tolerance limit. This might cost you as much as 3,000µF if the manufacturers are being as mean as they can be.
Last edited:
ukza
Active Member
Thanks for the answer. So far that is the only thread regarding the Sansui G-33000 I could find.
The thing is I would like to use 22000 because the 15000 hardly to get. But I dont want to break something, like the Power transistors.
I would like:
- colour black (I dont like the blue in this units)
- no cheap crap (chinese)
The thing is I would like to use 22000 because the 15000 hardly to get. But I dont want to break something, like the Power transistors.
I would like:
- colour black (I dont like the blue in this units)
- no cheap crap (chinese)