Hi,
These devices seem to be the only BJT in TO-3, and have fT of 4MHz, whereas originals are 25MHz. Whether this is relevant or not, they are seen by Sansui techs as a "temporary" solution, though.
Last edited:
Hpw about doing some research before spreading disinformation?
How you intend to pus 60V Vceo 1MHz fT transistors into an amp with a total 110V+ from + to - supply and original transistors for >40MHz fT I really do not understand.
So I would recommend that the original poster bill you for the damage that will happen to his amp once he follows your advice.
Originals are actually >40MHz fT typical and the design is still fairly overcompensated but will not work well with MJ21193/3 as good as they are. Actually it WILL work but will simply not sound right. Same thing with replacing the drivers with MJE15031/2. There are multiple reasons for this which I will not go into here. In fact, it is a good idea to replace the originals even if they are OK with higher Vceo parts.
Namely, these parts are actually 120V Vceo and the power supply, especially if the amp is set to 220/110V now that mains is 230/115V, is very close to that. In high mains situations it will even be over that. In fact, the filter caps are rated 63V and under high mains set to the old standard will work at 61V!!! I invariably find them with about 2/3 capacitance remaining because of this, when these amps come in for a rebuild - in fact I am working on one right now.
This situation results in SOA violations for 4 ohm nominal loads. The transistors are fairly robust (1st generation LAPT) but they will develop leakage.
Also with these amps it is IMPERATIVE to replace the 'black flag' silver mica capacitors. These are about 1x1cm black squares. The problem here is they tend to go open circuit, leaving the amp without compensation - and turning it into an oscillator. Result: dead output stage, due to cross-conduction. This is the most common reason of failures on AU919 and BA-F1 amplifiers, followed by output short circuit, followed by things like A+B low impedance speakers.
Unfortunately, the only transistors that could be used as direct replacements were TO-3 derivations of the Toshiba 2SC5200/2SA1943 parts, made by ST and then discontinued after a very short production run.
It IS possible to replace the original TO-3 transistors with 2SA1294/2SC3283 or even better 2SA2223A/2SC6145A in MT100/TO3P case but some work is needed. THis is from a Croatian forum but the pictures are self-explanatory - these were prepared specially for the AU-919.
diyaudio.com.hr: TO3P to TO3 adaptation
Namely, these parts are actually 120V Vceo and the power supply, especially if the amp is set to 220/110V now that mains is 230/115V, is very close to that. In high mains situations it will even be over that. In fact, the filter caps are rated 63V and under high mains set to the old standard will work at 61V!!! I invariably find them with about 2/3 capacitance remaining because of this, when these amps come in for a rebuild - in fact I am working on one right now.
This situation results in SOA violations for 4 ohm nominal loads. The transistors are fairly robust (1st generation LAPT) but they will develop leakage.
Also with these amps it is IMPERATIVE to replace the 'black flag' silver mica capacitors. These are about 1x1cm black squares. The problem here is they tend to go open circuit, leaving the amp without compensation - and turning it into an oscillator. Result: dead output stage, due to cross-conduction. This is the most common reason of failures on AU919 and BA-F1 amplifiers, followed by output short circuit, followed by things like A+B low impedance speakers.
Unfortunately, the only transistors that could be used as direct replacements were TO-3 derivations of the Toshiba 2SC5200/2SA1943 parts, made by ST and then discontinued after a very short production run.
It IS possible to replace the original TO-3 transistors with 2SA1294/2SC3283 or even better 2SA2223A/2SC6145A in MT100/TO3P case but some work is needed. THis is from a Croatian forum but the pictures are self-explanatory - these were prepared specially for the AU-919.
diyaudio.com.hr: TO3P to TO3 adaptation
Last edited:
Also have a look here - again on the same Croatian forum, but look at the pictures from the beginning of the page and you will get the idea. On this amp the sockets were rewired somewhat so that the transistor positions ended up different but this does not need to be done.
diyaudio.com.hr: Sansui AU-919
You may look through the whole thread for lots of pictures of a rather detailed restauration.
diyaudio.com.hr: Sansui AU-919
You may look through the whole thread for lots of pictures of a rather detailed restauration.
Very good!
Looking at DS, these are much better/safer than those from ON's MJ21193/4 to use in the AU-919 Do these perform OK?
AK's member Leestereo is also doing a (ongoing) complete and detailed resto in an AU-919, including info on the dangerous black flags caps.
Leestereo's Restoration and Upgrade Of A Sansui AU-919 | Audiokarma Home Audio Stereo Discussion Forums
Extra information
Leestereo's Restoration and Upgrade Of A Sansui AU-919 | Audiokarma Home Audio Stereo Discussion Forums
Extra information
@Karl,
The PNP part has a Ft of 40MHz min, 60MHz typ, the NPN is 25MHz min, 40MHz typ. I have been trying to find a (rather horrible) scan of the datasheet I had, there are 2SA/2SC equivalents for these, or rather, there WERE.
2ST2121 and 2ST5949 are ideal replacements, so it figures they made them obsolete it is now almost impossible to get them.
The PNP part has a Ft of 40MHz min, 60MHz typ, the NPN is 25MHz min, 40MHz typ. I have been trying to find a (rather horrible) scan of the datasheet I had, there are 2SA/2SC equivalents for these, or rather, there WERE.
2ST2121 and 2ST5949 are ideal replacements, so it figures they made them obsolete
it is now almost impossible to get them.
Hello guys!
Three days ago I entered in possession of an Au 919 and unfortunately for me, it looks painful inside. I'll put some pictures for the example.
The major problem with it is that out of the eight final transistors there is only one original one. The rest is a soup, a blend of 4 Thoshiba with 1 Sanken and 3 Motorola + 1 Sansui.
I read this forum about the restorations of this wonderful device, which I wanted very much, so I will appeal to yours goodness to help me with some ideas on what to do next. I tell you from the beginning that I do not know nothing about of electronics and I will appeal to one electronist who is a few hundred miles away from me. I would trust with one of those who posted here the Au 919 restoration but the distance is very big because I live in Romania.
I have been looking for some replacement days for the original transistors and from what I read here on the forum I understood that they successfully used the 2SA2121 / 2SC5949 pair to which I would have access but do not know if they are counterfeit by the Chinese. I searched the site: SANSUI-PARTS-SHOP where I found only 2SC2607, 2SA1116 have no stock, the latter being those recommended by the manufacturer.
Another problem I have found is that when applying tuna corrections, they feel hardly at all. So there's should be work there.
He is in working order and this looks very good. I really want to restore it and enjoy the sound of Sansui. Now with the transistor mix does not differ much from the old HK680 amplifier.
The speakers I'm going to pair with are HPM 110, which will also need to be restored. the parts in the filters are original, but I noticed that colloid was passed from the capacitors.
I apologize for English, I asked for Google translation help
Best regards!
Three days ago I entered in possession of an Au 919 and unfortunately for me, it looks painful inside. I'll put some pictures for the example.
The major problem with it is that out of the eight final transistors there is only one original one. The rest is a soup, a blend of 4 Thoshiba with 1 Sanken and 3 Motorola + 1 Sansui.
I read this forum about the restorations of this wonderful device, which I wanted very much, so I will appeal to yours goodness to help me with some ideas on what to do next. I tell you from the beginning that I do not know nothing about of electronics and I will appeal to one electronist who is a few hundred miles away from me. I would trust with one of those who posted here the Au 919 restoration but the distance is very big because I live in Romania.
I have been looking for some replacement days for the original transistors and from what I read here on the forum I understood that they successfully used the 2SA2121 / 2SC5949 pair to which I would have access but do not know if they are counterfeit by the Chinese. I searched the site: SANSUI-PARTS-SHOP where I found only 2SC2607, 2SA1116 have no stock, the latter being those recommended by the manufacturer.
Another problem I have found is that when applying tuna corrections, they feel hardly at all. So there's should be work there.
He is in working order and this looks very good. I really want to restore it and enjoy the sound of Sansui. Now with the transistor mix does not differ much from the old HK680 amplifier.
The speakers I'm going to pair with are HPM 110, which will also need to be restored. the parts in the filters are original, but I noticed that colloid was passed from the capacitors.
I apologize for English, I asked for Google translation help
Best regards!
Attachments
I have successfully tested 2SA2223A and 2SC6145A as outputs in both AU919 and BA-F1 (the latter is the separate power amp version which is a virtually identical circuit).
You can find these at Profusion Audio Semiconductors - Buy Online, guaranteed Sanken parts.
You can use 2SA1837 and 2SC4793 as drivers (on the small heatsinks on the power amp PCB) but you will need to modify them by grinding off the integrated heatsink insulation (these are completely plastic insulated) to expose the heatsink tab metal and use standard mica insulator, and using the middle lead (collector) instead of the heatsink tab as a contact. This MUST be done or the bias thermal tracking is not going to work properly. This is a bit of a complication but unfortunately anything else I could recommend as replacements is not guaranteed not to be fake.
The TO126 pre-divers etc you can replace with 2SB649 and 2SD669. This is not an ideal replacement but close enough not to require other changes.
All of the above are available at Profusion. The output stage will be a LOT more robust with the replacements.
You can find these at Profusion Audio Semiconductors - Buy Online, guaranteed Sanken parts.
You can use 2SA1837 and 2SC4793 as drivers (on the small heatsinks on the power amp PCB) but you will need to modify them by grinding off the integrated heatsink insulation (these are completely plastic insulated) to expose the heatsink tab metal and use standard mica insulator, and using the middle lead (collector) instead of the heatsink tab as a contact. This MUST be done or the bias thermal tracking is not going to work properly. This is a bit of a complication but unfortunately anything else I could recommend as replacements is not guaranteed not to be fake.
The TO126 pre-divers etc you can replace with 2SB649 and 2SD669. This is not an ideal replacement but close enough not to require other changes.
All of the above are available at Profusion. The output stage will be a LOT more robust with the replacements.
I am aware of what awaits me, from the purchase of all the components to the final restoration act. But I have a big question for you, those who have succeeded in restoring it have made a significant change in the sound of the amplifier? I asked this question because I did not have the opportunity to listen to an AU-919 with the original pieces(one restored) to make a difference between what I now hear, with all the transistor mix from mine!
How about the 2sc5171/2sa1930?
Dear Ilimzn,
The transistors you suggest have a maximum DC collector current of 1 amp, while the originals had 1,5 Amps. Would this be a limitation in the AU-919 or AU-X1?. Is there a reason not to use the 2SC5171/2SA1930 pair are rated at 2 amps of maximum collector current? These appear to be just as available as the 2SA1837/2SC4793.
Thank you,
Jose
Dear Ilimzn,
The transistors you suggest have a maximum DC collector current of 1 amp, while the originals had 1,5 Amps. Would this be a limitation in the AU-919 or AU-X1?. Is there a reason not to use the 2SC5171/2SA1930 pair are rated at 2 amps of maximum collector current? These appear to be just as available as the 2SA1837/2SC4793.
Thank you,
Jose
You can use either on the AU-919, on the X1 A1837/C4793 might be a bit strained since they have to drive 3 pairs of outputs, and also A1930/C5171 have not only a higher Icmax, but also a higher Ft, so might be a better choice for the X1. However, I cannot be absolutely sure as I have not tried them.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.