Hello Guys!
I am trying to restore an vintage Sony TA-1130 integrated amplifier.
The thing is in near mint condition, but it gives a hissing sound on the left channel.
It only becomes audible above 40% of volume
I already fully recapped with new electrolytes. I used Nichicon fine gold and Panasonic FC'a. But still it hiss/noise on the left channel.
For now it is narrowed down to the pre-amp section. Because when I use the power amp direct input it is quiet, no problem at all.
The problem doesn't go away after turning the balance knob to the other channel. My guess is that it has something to do with the smaller transistors or fets in the pre-amp.
Question, there are some 2sk23-35 and 2sk23-37 fets in the circuit. I cannot find the exact specs. Found some articles but there are inconclusive.
Does anyone know good replacements How to check these fets?
I did not replace the mylar and ceramic capacitors. Can these give the hissing problem?
I hope someone can help me out to locate the problem. I like this amp a lot!
Best Regards,
Kevin
I am trying to restore an vintage Sony TA-1130 integrated amplifier.
The thing is in near mint condition, but it gives a hissing sound on the left channel.
It only becomes audible above 40% of volume
I already fully recapped with new electrolytes. I used Nichicon fine gold and Panasonic FC'a. But still it hiss/noise on the left channel.
For now it is narrowed down to the pre-amp section. Because when I use the power amp direct input it is quiet, no problem at all.
The problem doesn't go away after turning the balance knob to the other channel. My guess is that it has something to do with the smaller transistors or fets in the pre-amp.
Question, there are some 2sk23-35 and 2sk23-37 fets in the circuit. I cannot find the exact specs. Found some articles but there are inconclusive.
Does anyone know good replacements How to check these fets?
I did not replace the mylar and ceramic capacitors. Can these give the hissing problem?
I hope someone can help me out to locate the problem. I like this amp a lot!
Best Regards,
Kevin
Try the following:
Select a high-level input.
Short said input on both channels.
Turn volume from min to max and observe noise levels.
If it goes away again at maximum volume, you've got excessive current noise there, indicating leakage. Also measure DC (mV range) across R303 (compare with R403 if needed). This should be barely measurable, if at all.
There are only two things that could possibly go leaky here, the FET or the 1n cap, C304. That is a mylar type, which tend to fail extremely rarely, though in equipment over 40 years old, you just never know. It should not be a problem to remove this bootstrap cap temporarily IMHO, if things are still hissy then you know it's the FET.
My guess is that the 2SK23 suffixes indicate Idss classes, -37 would be 3.7 mA. So basically finding a good sub for 2SK23, buying a whole bunch and screening them for Idss would be the way to go. I think 2SK23 isn't even that special, it was just the go-to JFET type back in the day. I get 2SK107 and 2SK192 as possible subs, also the trusty BF245 (A type I guess) but seemingly with pinout differences.
Select a high-level input.
Short said input on both channels.
Turn volume from min to max and observe noise levels.
If it goes away again at maximum volume, you've got excessive current noise there, indicating leakage. Also measure DC (mV range) across R303 (compare with R403 if needed). This should be barely measurable, if at all.
There are only two things that could possibly go leaky here, the FET or the 1n cap, C304. That is a mylar type, which tend to fail extremely rarely, though in equipment over 40 years old, you just never know. It should not be a problem to remove this bootstrap cap temporarily IMHO, if things are still hissy then you know it's the FET.
My guess is that the 2SK23 suffixes indicate Idss classes, -37 would be 3.7 mA. So basically finding a good sub for 2SK23, buying a whole bunch and screening them for Idss would be the way to go. I think 2SK23 isn't even that special, it was just the go-to JFET type back in the day. I get 2SK107 and 2SK192 as possible subs, also the trusty BF245 (A type I guess) but seemingly with pinout differences.
Thanx for your reply!
It seems you're right! I measure some mv's in one of the two channels. Also the noise disappears mostly when turning up the volume to the max (tried this only a few seconds, still 40 years old electronics
2SK107 seems hard to get. 2SK192 is possible, so is 2SK117. This is a slightly heavier part. But Idss seems close. How close should I be with the Idss?
Best Regards and many thanx!
It seems you're right! I measure some mv's in one of the two channels. Also the noise disappears mostly when turning up the volume to the max (tried this only a few seconds, still 40 years old electronics
2SK107 seems hard to get. 2SK192 is possible, so is 2SK117. This is a slightly heavier part. But Idss seems close. How close should I be with the Idss?
Best Regards and many thanx!
Good question, no idea. This one is better left to the FET gurus. I have attached the part of the schematic in question for any fellow readers.
What I do see is that the FET is operated fairly close to its nominal voltage rating, about 15 V d-s (it's a 20 V part). I would prefer to see a 30 V part here, and if in doubt I'd guess it's better to err on the side of slightly higher Idss. (Lowest distortion, of course, is generally obtained with a part having lower capacitance, run at higher Vgd, supporting about exactly the contrary.) In the end what we want is an operating point as close to the original as possible (Ids = 2.22 mA or even a bit higher at Vgs = -3.0 V, under no circumstances lower than about 1.7 mA at this point).
Attachments
My has the same problem, I just detected that was caused because of my treadmill. Recently I had the logic board changed. Just turned off the cable from the outlet and it disappeared. Vfet are very sensitive I think.
Thanks sgrossklass!
Seems the 2SK117GR really works out well. I ordered 10 of them. All where between 2,7mA and 4,2mA
I pulled the J-fet which I thought was the noisy one. It was Q301. After removing the Idss measures almost 17mA!
After that I replaced the other ones as well. (8 total including the Phono stage)
All the fets where way out of spec. With this high Idss on nearly all the fets I pulled I was surprised that there still was music to hear between the noise!
The smallest measured Idss from the old fets was 12mA where it should be 3,7mA. The only ones that where somewhat close to the original specifications where the ones from the phono stage. They where about 6mA. Still more than double what it should be.
Later on I will make some pictures of my finished and very quiet vintage amplifier!
Seems the 2SK117GR really works out well. I ordered 10 of them. All where between 2,7mA and 4,2mA
I pulled the J-fet which I thought was the noisy one. It was Q301. After removing the Idss measures almost 17mA!
After that I replaced the other ones as well. (8 total including the Phono stage)
All the fets where way out of spec. With this high Idss on nearly all the fets I pulled I was surprised that there still was music to hear between the noise!
The smallest measured Idss from the old fets was 12mA where it should be 3,7mA. The only ones that where somewhat close to the original specifications where the ones from the phono stage. They where about 6mA. Still more than double what it should be.
My advise is to order a bunch of 2SK117Y and 2sk117GR. In my case I ordered only the GR. I was lucky to receive a batch very close to what I needed. But if they where all 6.5mA what is possible with the GR rank, then it would be to high.
Later on I will make some pictures of my finished and very quiet vintage amplifier!
If your Idss measurements are anything to go by, my guess WRT part markings must have been at least partly wrong - while it does seem like higher number mean higher Idss, it is clear that a -37 could be 12 to 18 mA of Idss, and a -22 would still be over 6. (I don't think FET Idss magically goes up when they age.) Did you make a list of which kind of FET measured what?
From what I found 2SK23 could be anywhere between 2.7 mA and 12 mA of Idss, but with parts that old and information being hard to come by, who knows. The 2SK192A, for example, could be between 3 and 24 mA and is/was being sold in 3 Idss groups (Y= 3-7, GR = 6-14 and BL = 12-24 mA).
So I think you should put all the FETs back that weren't actually defective, taking care not to kill them through ESD. High gm, low capacitance FETs tend to be a bit hard to come by these days. NEC made some that essentially were the equivalent of two 2SK192s, with a highest Idss rank of 25-50 mA at Crss = 1.6 pF (Vbr,gds only 15 V though) - Sony used these in their portable shortwave receivers, probably about as common as the proverbial hen's teeth these days.
I'm not sure how you came up with 2SK117, these have pretty high Ciss/Crss - no surprise given the whopping 50 V breakdown voltage. For Q301 it seems you'd need at least a 2SK117BL, 2SK192BL would be better. What are voltages around it like right now?
From what I found 2SK23 could be anywhere between 2.7 mA and 12 mA of Idss, but with parts that old and information being hard to come by, who knows. The 2SK192A, for example, could be between 3 and 24 mA and is/was being sold in 3 Idss groups (Y= 3-7, GR = 6-14 and BL = 12-24 mA).
So I think you should put all the FETs back that weren't actually defective, taking care not to kill them through ESD. High gm, low capacitance FETs tend to be a bit hard to come by these days. NEC made some that essentially were the equivalent of two 2SK192s, with a highest Idss rank of 25-50 mA at Crss = 1.6 pF (Vbr,gds only 15 V though) - Sony used these in their portable shortwave receivers, probably about as common as the proverbial hen's teeth these days.
I'm not sure how you came up with 2SK117, these have pretty high Ciss/Crss - no surprise given the whopping 50 V breakdown voltage. For Q301 it seems you'd need at least a 2SK117BL, 2SK192BL would be better. What are voltages around it like right now?
Last edited:
Hello Again guys!
A few months later, picking up the project again. The Sony was stored for a while because of moving to another place.
At this moment the amp is running nicely. It sounds very good!
There is only one problem left:
When I power up the amplifier, there is a loud "pop" going through the speakers. A little is not so bad, but this one is..
I think the problem lies in the mutingboard.
There is a pcb in the amp special for the muting circuit which should connect the preamp to ground when powering up.
It seems that it acts to slow. If i measure between the red and black wire, or the white and black wire from the muting board, i see on the meter that after a second or so, the meter gives about 20 ohms. It is not near short circuit to ground, what i think should be. Then a few seconds later, it says open connection. This is how it should be.
So in the first second, the "pop" is already been there, before the preamp was muted. Or maybe my multimeter is just to slow.
Please look at the schematics below
The other possibility is that the problem lies in the replacement Jfets, 2sk117
As sgrossklass mentioned;
I came up with these because of what i thought was a close idss match, and another forum where they used it succesfully to repair an old Sony Pre-amp which also had the 2sk23 originally.
So maybe someone can help me locate this problem.
A few months later, picking up the project again. The Sony was stored for a while because of moving to another place.
At this moment the amp is running nicely. It sounds very good!
There is only one problem left:
When I power up the amplifier, there is a loud "pop" going through the speakers. A little is not so bad, but this one is..
I think the problem lies in the mutingboard.
There is a pcb in the amp special for the muting circuit which should connect the preamp to ground when powering up.
It seems that it acts to slow. If i measure between the red and black wire, or the white and black wire from the muting board, i see on the meter that after a second or so, the meter gives about 20 ohms. It is not near short circuit to ground, what i think should be. Then a few seconds later, it says open connection. This is how it should be.
So in the first second, the "pop" is already been there, before the preamp was muted. Or maybe my multimeter is just to slow.
Please look at the schematics below
The other possibility is that the problem lies in the replacement Jfets, 2sk117
As sgrossklass mentioned;
I came up with these because of what i thought was a close idss match, and another forum where they used it succesfully to repair an old Sony Pre-amp which also had the 2sk23 originally.
So maybe someone can help me locate this problem.
Attachments
Original fets are back in place, only the noisy one is now replaced bij 2sk192a. Problems solved..!
- Status
- Not open for further replies.