Nakamichi TA-2A - power supply board
- Thread starter northpaw
- Start date
Mannegizen
Active Member
Yeah, it makes sense that you can't drain reset down to 0 with the battery in.
I tried to drain on my unit too with the battery in, first I didn't get why I couldn't get it below 3V.
I removed the battery, drained pin 1 to below 1V and disconnected reset cable from CN-9.
I could do one startup to green with reset cable disconnected, second went directly to orange.
So I finally managed to simulate your units condition.
So that strengthens my belief that your reset circuit is bad. (I'm learning stuff too ;-) )
I tried to drain on my unit too with the battery in, first I didn't get why I couldn't get it below 3V.
I removed the battery, drained pin 1 to below 1V and disconnected reset cable from CN-9.
I could do one startup to green with reset cable disconnected, second went directly to orange.
So I finally managed to simulate your units condition.
So that strengthens my belief that your reset circuit is bad. (I'm learning stuff too ;-) )
Last edited:
northpaw
Super Member
Start up is functioning normally now. Looks like it was Q405 was the problem, the last thing I checked.
The measurements of other components seemed normal. I got 538, 535, 540 for D407-409. R410 and R411 measured dead on (in circuit, but with CN-9 detached; best I could do without unsoldering). R412 measures a bit low at 94.5K (spec is 100K), and while also in circuit, one end is ground. I don't know if this is worth replacing; possibly it is in a stage of deteriorating, being a bit more than 5% off spec. I'll have to check the tolerance band on that one.
Q405: I got continuity between a leg of C409 (connects to the collector of Q405) and pin 1 on CN-9 (however, this is misleading; see below). I could not get in there well enough to get leads on Q405 itself for measurement, so I unsoldered it. I used my desolder braid on it, but I have to pick at it a bit because there is a noticeable bend on two of the pins. As I pick at the middle one, I get dust, which is odd. I ease it out, but it only has two legs. Well, there is a third, the center, but it is just a rusted tiny nib. There is a trace of brown color around the middle hole on the board. I don't bother to try as measure Q405, just replaced it.
Check pin 1 before starting, and it is 0, as expected, given a new Q405. Starts fine, to green. A minute later shut it off. Wait 15 sec. Check the voltage on pin 1, and it is 3.5V, I leave it as is, and restart. An instant of yellow, then on to flashing green, and then green.
Mannegizen, you are a wizard!
The measurements of other components seemed normal. I got 538, 535, 540 for D407-409. R410 and R411 measured dead on (in circuit, but with CN-9 detached; best I could do without unsoldering). R412 measures a bit low at 94.5K (spec is 100K), and while also in circuit, one end is ground. I don't know if this is worth replacing; possibly it is in a stage of deteriorating, being a bit more than 5% off spec. I'll have to check the tolerance band on that one.
Q405: I got continuity between a leg of C409 (connects to the collector of Q405) and pin 1 on CN-9 (however, this is misleading; see below). I could not get in there well enough to get leads on Q405 itself for measurement, so I unsoldered it. I used my desolder braid on it, but I have to pick at it a bit because there is a noticeable bend on two of the pins. As I pick at the middle one, I get dust, which is odd. I ease it out, but it only has two legs. Well, there is a third, the center, but it is just a rusted tiny nib. There is a trace of brown color around the middle hole on the board. I don't bother to try as measure Q405, just replaced it.
Check pin 1 before starting, and it is 0, as expected, given a new Q405. Starts fine, to green. A minute later shut it off. Wait 15 sec. Check the voltage on pin 1, and it is 3.5V, I leave it as is, and restart. An instant of yellow, then on to flashing green, and then green.
Mannegizen, you are a wizard!
Last edited:
Mannegizen
Active Member
Yeah, that is the normal start up behavior.
Finally a break tru
, I celebrate that with a really thick imperial porter. 
Mannegizen
Active Member
Well, when I kill stuff I work on, I regard my self as a "Trial and terror troll", but practice makes perfect, cheers
northpaw
Super Member
I feel a desire to supply you with a few of those stark öls. I think there may be a way for me to do that, but I will discuss that with you later, privately.
The 30V/12V circuit still needs my attention (and your help), but I think I am also ready for a beer too. And it is sunny out, and the backyard beckons, so I will head there shortly.
I did check the diagrams for the AC feed to the 30V/12V circuit, and I measured 16.5VAC between the two "blu" secondaries coming off the big transformer. I wanted to know that in case it makes sense to feed it lower voltage when diagnosing that circuit. Since it seems like feeding the entire unit reduced voltage to lower the 16.5VAC secondaries might not be a good idea, and the fuse that is blowing is a 0.5A one, not the main fuse, I do have a Variac that I could use to separately feed the 30/12V circuit. I would need to cut into the "blu" leads, and retie them later.
But as I say, for another day.
The 30V/12V circuit still needs my attention (and your help), but I think I am also ready for a beer too. And it is sunny out, and the backyard beckons, so I will head there shortly.
I did check the diagrams for the AC feed to the 30V/12V circuit, and I measured 16.5VAC between the two "blu" secondaries coming off the big transformer. I wanted to know that in case it makes sense to feed it lower voltage when diagnosing that circuit. Since it seems like feeding the entire unit reduced voltage to lower the 16.5VAC secondaries might not be a good idea, and the fuse that is blowing is a 0.5A one, not the main fuse, I do have a Variac that I could use to separately feed the 30/12V circuit. I would need to cut into the "blu" leads, and retie them later.
But as I say, for another day.
northpaw
Super Member
I felt refreshed after that beer and dinner, and decided to give the 30V/12V circuit another try.
I replaced the components Mannegizen indicated should be replaced earlier today, but still had the trashed D401 rectifier in place. Since the circuit overdraws current (blows 0.5A F104) with a failed D401 in place, I had no choice but to install a new rectifier and try to test the circuit. To minimize the possibility of trashing D410 again, I did not replace F104 initially, but used my multimeter in ammeter mode as a "manual fuse". By placing one lead on either side of the fuse holder, I could monitor the current draw on the meter, and if I see it go more than momentarily above 0.5A, I simply pull the leads off to interrupt the circuit.
After an initial moment at about 0.20 to 0.25A, it settles down to a steady 0.17A. I kept holding the leads that way for almost 10 min, and it was very stable. I shut it off and installed the fuse, and restart. Voila, now I can see that the rest of the display has lit up (freq numbers were there previously, but not the other tuning functions). Tuning scanning works, and the signal strength meter and stereo indicator work. Oddly, previously the decimal point in the FM freq was missing, but now is present. And the volume control function on the remote now works. I have the unit hooked up to some OLAs and playing NPR on the radio now. All functions appear to be working. All of this came as a great surprise to my wife, but she is quite happy as well (the prospect that the dining room table will return to normal...).
So I declare this thing fixed. Many thanks to all who commented and helped me, with my deepest gratitude of course to Mannegizen for selflessly sharing his knowledge and diagnostic skills
, and for going far far beyond the call of duty in taking apart his own equipment to help figure out what was wrong with mine 
. I learned a tremendous amount from this experience.
I replaced the components Mannegizen indicated should be replaced earlier today, but still had the trashed D401 rectifier in place. Since the circuit overdraws current (blows 0.5A F104) with a failed D401 in place, I had no choice but to install a new rectifier and try to test the circuit. To minimize the possibility of trashing D410 again, I did not replace F104 initially, but used my multimeter in ammeter mode as a "manual fuse". By placing one lead on either side of the fuse holder, I could monitor the current draw on the meter, and if I see it go more than momentarily above 0.5A, I simply pull the leads off to interrupt the circuit.
After an initial moment at about 0.20 to 0.25A, it settles down to a steady 0.17A. I kept holding the leads that way for almost 10 min, and it was very stable. I shut it off and installed the fuse, and restart. Voila, now I can see that the rest of the display has lit up (freq numbers were there previously, but not the other tuning functions). Tuning scanning works, and the signal strength meter and stereo indicator work. Oddly, previously the decimal point in the FM freq was missing, but now is present. And the volume control function on the remote now works. I have the unit hooked up to some OLAs and playing NPR on the radio now. All functions appear to be working. All of this came as a great surprise to my wife, but she is quite happy as well (the prospect that the dining room table will return to normal...).
So I declare this thing fixed. Many thanks to all who commented and helped me, with my deepest gratitude of course to Mannegizen for selflessly sharing his knowledge and diagnostic skills
, and for going far far beyond the call of duty in taking apart his own equipment to help figure out what was wrong with mine . I learned a tremendous amount from this experience.Replying to a 4 year old thread can be fruitless, but I am hopeful that since one of my problems is similar to northpaw's, that this is as good a place as any to ask for help. I recently bought a broken TA-2A on CL. It was doing the typical flashing power light indicator and not able to move out of protection mode. A simple replacement of R309 22K resistor on the main board fixed that problem.
Now I am dealing with two more issues - one of which I will explain here. (The other issues is a very noisy FM when an antenna is attached which seems to be due to trouble on the main PC board next to the antenna jacks. I was able to pick up noise there with an antenna connected to another tuner.)
The power supply problem I am having is on the lower voltage side that feeds the logic board. The output of D401 is only 9.5 volts rather than 10.59 marked in the service manual. Pin 7 on CN 9 (which feeds the logic board) measures 7.8 volts rather than 10 with the power on. So the question I have is whether I have to replace more than D401. I have most of the capacitors that are on this board but the board is such a pain to work on that I am not interested in replacing them all. I am afraid things might be worse rather than better if I try to remanufacture the whole board while working on it partially flipped from the bottom of the case.
I found only one other value that was significantly lower than expected on the power supply board, That is the voltage between D414 and D415. It is supposed to be 39.6 volts. I measured 32.7 v. But the output to the main board on CN-21 is 28.9 v very close to the expected 30 volts. I don't think it is worth changing components to gain 1 volt when the amplifier seems to be working well.
Can someone please help with my low voltage to the logic board problem?
Now I am dealing with two more issues - one of which I will explain here. (The other issues is a very noisy FM when an antenna is attached which seems to be due to trouble on the main PC board next to the antenna jacks. I was able to pick up noise there with an antenna connected to another tuner.)
The power supply problem I am having is on the lower voltage side that feeds the logic board. The output of D401 is only 9.5 volts rather than 10.59 marked in the service manual. Pin 7 on CN 9 (which feeds the logic board) measures 7.8 volts rather than 10 with the power on. So the question I have is whether I have to replace more than D401. I have most of the capacitors that are on this board but the board is such a pain to work on that I am not interested in replacing them all. I am afraid things might be worse rather than better if I try to remanufacture the whole board while working on it partially flipped from the bottom of the case.
I found only one other value that was significantly lower than expected on the power supply board, That is the voltage between D414 and D415. It is supposed to be 39.6 volts. I measured 32.7 v. But the output to the main board on CN-21 is 28.9 v very close to the expected 30 volts. I don't think it is worth changing components to gain 1 volt when the amplifier seems to be working well.
Can someone please help with my low voltage to the logic board problem?
northpaw
Super Member
Yeah, it is a real PITA to work on that PS board.
I suppose D401 could be faulty, but before replacing it consider that it also seems possible that it is loaded down, reducing its voltage. This is just a guess, but it is based on the observation that there isn't much between the + pin of D401 and pin 7 of CN-9: only R401 (56Ω/1W). The other thing of note is that the output of D401 has C403 (1000µF/16V) across its + and - pins, and the health of C403 could affect the voltage across it.
So it seems there are two alternatives to D401 being bad: (1) something over on the logic board is drawing more current than it is supposed to, dropping more voltage than intended across R401 and also loading down D401, and (2) C403 is failing and causing a reduction of DC voltage across it. I think it is worth checking both of these out before replacing D401, assuming it is bad.
I think the way to check for case 1 is to disconnect CN-9, and see if the voltage rises to the correct range. As I recall, CN-9 is soldered on the PS PCB end, so just disconnect it on the Logic PCB end.
I think the way to check for case 2 is set your DMM to ACV mode and measure the ripple across C403. I am not sure what the max value should be, but I would start being suspicious if it was more than 10-20 mv. This thought is driven by my experience on other units, that as a PS smoothing cap fails, the DC voltage across its terminals will drop as the AC ripple across its terminals increases. For a smoothing cap on the main PS, this condition will be accompanied by hum so is easy to recognize, but here, that 10V DC is not driving speakers so you will not hear the result of its failing. Anyway, if there is any question about the cap, just replace it. Also I would use a cap rated at 25V (or more) instead of 16V. Also note that 16V and other similar low V caps tend to not have the best longevity, and these are over 30 years old.
I don't have any good ideas concerning the D414/D415 issue. The 30V branch of that circuit (actually, the schematic shows it as 29.55V, so the 30V is nominal and you are only 0.65V off) bypasses D410, and seems to only be half rectified by D413 and smoothed by C416 (470µF/25V) (this is my best guess). I can't quite decipher what is going on around Q408 (which is tied in part to the low pin of D410, as ground) except possibly some extra smoothing provided by C417 and C419, but you mentioned that you get good agreement with other marked voltages (such as 30.4V for the base of Q408?), so I would not assume things are amiss there. For lack of anything better to suggest, check and/or replace C416 would be my first step if I decided to try anything.
I suppose D401 could be faulty, but before replacing it consider that it also seems possible that it is loaded down, reducing its voltage. This is just a guess, but it is based on the observation that there isn't much between the + pin of D401 and pin 7 of CN-9: only R401 (56Ω/1W). The other thing of note is that the output of D401 has C403 (1000µF/16V) across its + and - pins, and the health of C403 could affect the voltage across it.
So it seems there are two alternatives to D401 being bad: (1) something over on the logic board is drawing more current than it is supposed to, dropping more voltage than intended across R401 and also loading down D401, and (2) C403 is failing and causing a reduction of DC voltage across it. I think it is worth checking both of these out before replacing D401, assuming it is bad.
I think the way to check for case 1 is to disconnect CN-9, and see if the voltage rises to the correct range. As I recall, CN-9 is soldered on the PS PCB end, so just disconnect it on the Logic PCB end.
I think the way to check for case 2 is set your DMM to ACV mode and measure the ripple across C403. I am not sure what the max value should be, but I would start being suspicious if it was more than 10-20 mv. This thought is driven by my experience on other units, that as a PS smoothing cap fails, the DC voltage across its terminals will drop as the AC ripple across its terminals increases. For a smoothing cap on the main PS, this condition will be accompanied by hum so is easy to recognize, but here, that 10V DC is not driving speakers so you will not hear the result of its failing. Anyway, if there is any question about the cap, just replace it. Also I would use a cap rated at 25V (or more) instead of 16V. Also note that 16V and other similar low V caps tend to not have the best longevity, and these are over 30 years old.
I don't have any good ideas concerning the D414/D415 issue. The 30V branch of that circuit (actually, the schematic shows it as 29.55V, so the 30V is nominal and you are only 0.65V off) bypasses D410, and seems to only be half rectified by D413 and smoothed by C416 (470µF/25V) (this is my best guess). I can't quite decipher what is going on around Q408 (which is tied in part to the low pin of D410, as ground) except possibly some extra smoothing provided by C417 and C419, but you mentioned that you get good agreement with other marked voltages (such as 30.4V for the base of Q408?), so I would not assume things are amiss there. For lack of anything better to suggest, check and/or replace C416 would be my first step if I decided to try anything.
northpaw,
Thank you for the two suggestions. I just tried to check the voltage with CN 9 disconnected and unfortunately the receiver won't power on when I pull the plug off the logic board.
As for measuring the ripple on C403, that is one of the capacitors glued to the board and short legs. I just checked this and the "glue" on the board must be leakage. Only two of the capacitors measured high and C403 is one of those. C416 I believe is the other. (I edited out my first erroneous response to your post.)
I do like the idea of changing out C403 as a first step. Once I turn the board over I can check to see if any of the solder looks bad, too.
Thank you for the two suggestions. I just tried to check the voltage with CN 9 disconnected and unfortunately the receiver won't power on when I pull the plug off the logic board.
As for measuring the ripple on C403, that is one of the capacitors glued to the board and short legs. I just checked this and the "glue" on the board must be leakage. Only two of the capacitors measured high and C403 is one of those. C416 I believe is the other. (I edited out my first erroneous response to your post.)
I do like the idea of changing out C403 as a first step. Once I turn the board over I can check to see if any of the solder looks bad, too.
Last edited:
northpaw
Super Member
Did you test for the voltage, or just noted the receiver didn't power up? As best as I can trace the circuit from the AC plug to power switch and relay RY401 on the PS PCB (it is a bit of a maze with crossing lines), the main transformer (PT402) will not power up if relay RY401 doesn't close, but there is no relay between the AC mains and the small transformer on the PS PCB (PT401), and that is what powers the 10V circuit. So I think the 10V circuit should be powered as long as the power switch is on. At least, that is what I think. That transformer has to be powered or the rest of the logic circuits wouldn't be powered, and they need to function in order to turn the unit on.
To measure ripple on C403, you can measure from equivalent points on the circuit. I am not looking at the actual board now, but from the schematic you should be able to measure across C403 by placing one probe on right lead (on the schematic) of R401, and the other probe on the - leg of D401 (or on pin 6 of CN-9). Those should be accessible while the board is seated in place.
To measure ripple on C403, you can measure from equivalent points on the circuit. I am not looking at the actual board now, but from the schematic you should be able to measure across C403 by placing one probe on right lead (on the schematic) of R401, and the other probe on the - leg of D401 (or on pin 6 of CN-9). Those should be accessible while the board is seated in place.
You are correct that the CN9 points are live even when the receiver does not power up. Some of the voltages are weird compared to the two sets of values that Mannegizen gives on the first pages of this message thread,
I have:
1 0.29v
2 5.5 v
3 -2.1v
4 0 v
5 6.0v
6 0.6 - 0.8 v
7 13.6 v
I also tested the voltages in the orange standby state by hitting the off position on the remote:
1 5.0v
2 5.3v
3 0.06v
4 0.2v
5 5.7v
6 0.0v
7 11.8v
So the common finding is that contact point 7 is low compared to Mannegizen's numbers.
Measuring the ripple from R401 to D401 I get 2.9 volts AC (to measure C403).
The measurement I incorrectly made for ripple current was from the metal top of the cannister of C403 to ground reading 1.38v and the top of C416 to ground at 8.9v AC. All the rest of the metal topped capacitors on the board and on the logic board read .6 v AC or less. Does it make sense to condemn them if the top of the metal area on the cap is showing AC vs ground?
Also I have the question again as to whether the dark gunk on the board near C403, C416, C405 and C412 is glue or leakage. None of these are bulging or look bad otherwise.
I have:
1 0.29v
2 5.5 v
3 -2.1v
4 0 v
5 6.0v
6 0.6 - 0.8 v
7 13.6 v
I also tested the voltages in the orange standby state by hitting the off position on the remote:
1 5.0v
2 5.3v
3 0.06v
4 0.2v
5 5.7v
6 0.0v
7 11.8v
So the common finding is that contact point 7 is low compared to Mannegizen's numbers.
Measuring the ripple from R401 to D401 I get 2.9 volts AC (to measure C403).
The measurement I incorrectly made for ripple current was from the metal top of the cannister of C403 to ground reading 1.38v and the top of C416 to ground at 8.9v AC. All the rest of the metal topped capacitors on the board and on the logic board read .6 v AC or less. Does it make sense to condemn them if the top of the metal area on the cap is showing AC vs ground?
Also I have the question again as to whether the dark gunk on the board near C403, C416, C405 and C412 is glue or leakage. None of these are bulging or look bad otherwise.
Last edited:
northpaw
Super Member
With CN-9 disconnected, some of those voltages would be different, because as I learned earlier in this thread, the PS board is not the only source of voltages that appear on CN-9. But not to worry, the primary issue is pin 7. You had 7.8V before. Now you have 13.6V with CN-9 disconnected, very close to Mannegizen's value. So it seems possible that there is some excess loading going on, and that might need to be chased, but first the cap situation should be sorted out (see further below).
For clarity, are the caps still in place; i.e., did you make those measurements you report for CN-9 above with the caps in place? And were the standby readings with CN-9 connected, or not?
On to the caps. The 2.9 volts of AC when measuring from R401 to D401 seems excessively high (measurements of the cap tops to ground aren't meaningful, as the outer casing of the caps are not [supposed to be] part of the circuit, although leaked electrolyte might change that; what you measured could possibly have a component of ambient AC fields in the air). That the ripple (in the circuit) seems too high is based on my experience with measurements I've made on much larger main PS caps, where the DC voltage is 45-60V and measured ripple is less than 100mv AC. The lower voltage and lower power situation here is different, but volts of AC still seems very high to me (I can't check what the ripple is on my units right now to see what they show, but I might be able to in a few days).
Leaked electrolyte, the rare times I have seen it, usually looks grimy and is typically thin, not like the thick blobs that glue looks like. You haven't posted a photo, so I can't offer much of an opinion on what yours is, but from looking at my saved photos, there is no glue on this board, at least on my units (edit: see next post). I think it is reasonable to conclude that those caps have leaked and they should be replaced.
So I think the next thing to do is remove the caps if they are still on the board, clean up the board well, install new caps, reassemble everything (including CN-9), and see what the voltages on CN-9 looks like. If things on CN-9 are still off, we can address them then.
For clarity, are the caps still in place; i.e., did you make those measurements you report for CN-9 above with the caps in place? And were the standby readings with CN-9 connected, or not?
On to the caps. The 2.9 volts of AC when measuring from R401 to D401 seems excessively high (measurements of the cap tops to ground aren't meaningful, as the outer casing of the caps are not [supposed to be] part of the circuit, although leaked electrolyte might change that; what you measured could possibly have a component of ambient AC fields in the air). That the ripple (in the circuit) seems too high is based on my experience with measurements I've made on much larger main PS caps, where the DC voltage is 45-60V and measured ripple is less than 100mv AC. The lower voltage and lower power situation here is different, but volts of AC still seems very high to me (I can't check what the ripple is on my units right now to see what they show, but I might be able to in a few days).
Leaked electrolyte, the rare times I have seen it, usually looks grimy and is typically thin, not like the thick blobs that glue looks like. You haven't posted a photo, so I can't offer much of an opinion on what yours is, but from looking at my saved photos, there is no glue on this board, at least on my units (edit: see next post). I think it is reasonable to conclude that those caps have leaked and they should be replaced.
So I think the next thing to do is remove the caps if they are still on the board, clean up the board well, install new caps, reassemble everything (including CN-9), and see what the voltages on CN-9 looks like. If things on CN-9 are still off, we can address them then.
Last edited:
northpaw
Super Member
I just looked over my notes, and it is possible there was some glue on the original board. The photos I was talking about were taken after I replaced the caps so no glue was apparent, and my notes reference some glue on the board (although I was quite green then, in 2017, and I might have misidentified it). I also have a note that C403 had 9V DC on its top metal piece; the other caps had none. In retrospect, my PS board was in very sorry shape, with quite a bit of rust about on exposed metal parts.
Here are my notes to myself on the caps I replaced, in case they are helpful, or perhaps they offer fodder for commiseration:
Desolder C403. Marked 1000µF/16V, but measures only 100µF, with Vloss of 38% and ESR of 150Ω. So it is toast, but except for glue on bottom, looks ok. Replace with 1000µF/50V.
Desolder C412, a 1000µF/25V cap. Cap shows bulges near base, and oxidation on top cap. Tested as damaged/unknown component. Replace with 1000µF/50V.
Desolder C405, a 1000µF/10V cap. Tested ok, Vloss slightly high. Replace with 1000µF/50V.
Desolder C416, a 470µF/50V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C417, a 470µF/25V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C415, a 220µF/16V cap. Tested ok, Vloss high (>1%) and ESR slightly high. Replace with 220µF/50V.
Desolder C418, a 10µF/35V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C419, a 100µF/35V cap. Tested ok, Vloss high (>1%) and ESR >1Ω. Replace with 100µF/50V.
Leave C407 and C408 (2.2µF/50V) alone as do not have good subs for them, but they look decent.
Here are my notes to myself on the caps I replaced, in case they are helpful, or perhaps they offer fodder for commiseration:
Desolder C403. Marked 1000µF/16V, but measures only 100µF, with Vloss of 38% and ESR of 150Ω. So it is toast, but except for glue on bottom, looks ok. Replace with 1000µF/50V.
Desolder C412, a 1000µF/25V cap. Cap shows bulges near base, and oxidation on top cap. Tested as damaged/unknown component. Replace with 1000µF/50V.
Desolder C405, a 1000µF/10V cap. Tested ok, Vloss slightly high. Replace with 1000µF/50V.
Desolder C416, a 470µF/50V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C417, a 470µF/25V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C415, a 220µF/16V cap. Tested ok, Vloss high (>1%) and ESR slightly high. Replace with 220µF/50V.
Desolder C418, a 10µF/35V cap. Tested ok, Vloss slightly high (>1%). Replace with 470µF/50V.
Desolder C419, a 100µF/35V cap. Tested ok, Vloss high (>1%) and ESR >1Ω. Replace with 100µF/50V.
Leave C407 and C408 (2.2µF/50V) alone as do not have good subs for them, but they look decent.
Thanks for all these notes. I think the black gunk on my board is leakage of the capacitors. I neglected to mention thatI have a very pristine TA-2A whyI I took apart tonight to compare the voltages. The small amount of glue on the capacitors was very different from the black tar on the circuit board. I plan to replace all the 4 or more caps that leaked before any other work on the board.
CN9 on the more pristine board reads very different from my sick TA-2A. These are readings with the socket unplugged.
1 1.06v
2 4.66v
3 -0.3 v
4 0.0
5 4.5v
6 0.76v
7 11.6v
I take this to mean the good power supply is better regulated as the 12v reading is present whether hooked to the logic board or not.
Thanks again for your work on my behalf.
CN9 on the more pristine board reads very different from my sick TA-2A. These are readings with the socket unplugged.
1 1.06v
2 4.66v
3 -0.3 v
4 0.0
5 4.5v
6 0.76v
7 11.6v
I take this to mean the good power supply is better regulated as the 12v reading is present whether hooked to the logic board or not.
Thanks again for your work on my behalf.
northpaw
Super Member
The black tar is probably glue contaminated with electrolyte; look for additional grime on the PCB surrounding that, and be sure to clean that up with alcohol or similar.
I think with the compromised caps, the PS isn't able to maintain voltages under the designed load (when CN-9 is connected). Those issues might disappear once you get new caps in there.
One thing to watch out for as you work on the PS PCB is the two voltage regulator ICs. They look like TO-220 transistors, and have tall aluminum heat spreaders bolted to them. They run hot, and sometimes their solders suffer. But a bigger problem for me was that I tended to get one of the cables caught on the tall heat spreaders as I jockeyed the PS PCB around to work on its underside. The tension I put on them ended up tearing a solder eyelet and part of a trace, which I had to fix with a jumper.
I think with the compromised caps, the PS isn't able to maintain voltages under the designed load (when CN-9 is connected). Those issues might disappear once you get new caps in there.
One thing to watch out for as you work on the PS PCB is the two voltage regulator ICs. They look like TO-220 transistors, and have tall aluminum heat spreaders bolted to them. They run hot, and sometimes their solders suffer. But a bigger problem for me was that I tended to get one of the cables caught on the tall heat spreaders as I jockeyed the PS PCB around to work on its underside. The tension I put on them ended up tearing a solder eyelet and part of a trace, which I had to fix with a jumper.
Thanks for the tips. When I have the time I'll work on this some time this week.
I attempted to repair the power supply board by changing the capacitors with only one incident. The diode closest to C412 broke when I was cleaning the goo off the leaky capacitors off the board. Once I fixed that the receiver worked as expected. I want to add the voltage changes here with the listed voltages from the manufacturer's schematics.
CN9 Numbers
Contact ......Before.....After........Spec from Schematic
1..................4.9...........4.9
2.................11.4.........11.4
3..................5.4...........5.6
4..................5.2...........5.5..........4.5
5..................5.4...........5.6..........5
6..................0..............0.............0
7..................7.8...........9.8........10
So my conclusion is that the repair was at least somewhat successful. Contact #7 on CN9 is improved and close to the spec on the schematic. The voltages to the main board were fine before and after the "repair". The issue that I still have is that the display segments on the tuner display are still dim.
northpaw:
Above you mentioned that something might be wrong on the logic board. But I am at a loss as to what to look for there. I have done just two things on the logic board. I replaced the 3 volt lithium battery and I replaced C503.
(As an aside, the gunk on the power supply board was definitely leakage from C403,C405, C412, C416 and C417. That means all of the larger capacitors leaked. Another aside is that I am posting pictures for two reasons. One is to show the black crud on the board. The other is to show that I had less trouble getting to the bottom of the board than I expected. In addition moving the large transformer I moved the power switch board out of the way.)
My next step might be to open up the working nearly pristine TA-2A and compare voltages on the logic board to the one with the sick display.
CN9 Numbers
Contact ......Before.....After........Spec from Schematic
1..................4.9...........4.9
2.................11.4.........11.4
3..................5.4...........5.6
4..................5.2...........5.5..........4.5
5..................5.4...........5.6..........5
6..................0..............0.............0
7..................7.8...........9.8........10
So my conclusion is that the repair was at least somewhat successful. Contact #7 on CN9 is improved and close to the spec on the schematic. The voltages to the main board were fine before and after the "repair". The issue that I still have is that the display segments on the tuner display are still dim.
northpaw:
Above you mentioned that something might be wrong on the logic board. But I am at a loss as to what to look for there. I have done just two things on the logic board. I replaced the 3 volt lithium battery and I replaced C503.
(As an aside, the gunk on the power supply board was definitely leakage from C403,C405, C412, C416 and C417. That means all of the larger capacitors leaked. Another aside is that I am posting pictures for two reasons. One is to show the black crud on the board. The other is to show that I had less trouble getting to the bottom of the board than I expected. In addition moving the large transformer I moved the power switch board out of the way.)
My next step might be to open up the working nearly pristine TA-2A and compare voltages on the logic board to the one with the sick display.
Last edited:
I attempted to repair the power supply board by changing the capacitors with only one incident. The diode closest to C412 broke when I was cleaning the goo off the leaky capacitors off the board. Once I fixed that the receiver worked as expected. I want to add the voltage changes here with the listed voltages from the manufacturer's schematics.
CN9 Numbers
Contact ......Before.....After........Spec from Schematic
1..................4.9...........4.9
2.................11.4.........11.4
3..................5.4...........5.6
4..................5.2...........5.5..........4.5
5..................5.4...........5.6..........5
6..................0..............0.............0
7..................7.8...........9.8........10
So my conclusion is that the repair was at least somewhat successful. Contact #7 on CN9 is improved and close to the spec on the schematic. The voltages to the main board were fine before and after the "repair". The issue that I still have is that the display segments on the tuner display are still dim.
northpaw:
Above you mentioned that something might be wrong on the logic board. But I am at a loss as to what to look for there. I have done just two things on the logic board. I replaced the 3 volt lithium battery and I replaced C503.
(As an aside, the gunk on the power supply board was definitely leakage from C403,C405, C412, C416 and C417. That means all of the larger capacitors leaked. Another aside is that I am posting pictures for two reasons. One is to show the black crud on the board. The other is to show that I had less trouble getting to the bottom of the board than I expected. In addition moving the large transformer I moved the power switch board out of the way.)
My next step might be to open up the working nearly pristine TA-2A and compare voltages on the logic board to the one with the sick display.
As an addendum the tuner display may not be working at full illumination but the noise in the FM tuner is NOW GONE!!! So the power supply issues were causing horrible noise in the tuner which seems to have vanished.
When comparing the two logic boards the only area in which I found an issue is at the jumpers near Q513, Q516 and Q517. J597 show 0 v in the "good TA-2A" with the bright display and .73v in the sick display unit. J598 has .73v in the good TA-2A and 0v in the sick one. Finally J599 has -0.1v in the good receiver and -0.23v in the sick one.
Last edited:
northpaw
Super Member
Congrats on getting it fixed. Your board looks very clean; too bad so many of the caps went.
That was in the context of trying to think why pin 7 was too low, one possibility being too great a drain on it (from the Logic board). The other possibility was a bad cap was making the voltage low (see post #69). Since that turned out to be the case, and the voltage is now in agreement with the specs, the idea of the drain on the Logic board isn't valid/needed.
On the faded display segments, I can only repeat what I have read, and that is that it is display aging issue without any fix short of transplanting a better display. I don't know if the low voltages that you measured are a symptom of dim segments on a display, or the cause of it.
That was in the context of trying to think why pin 7 was too low, one possibility being too great a drain on it (from the Logic board). The other possibility was a bad cap was making the voltage low (see post #69). Since that turned out to be the case, and the voltage is now in agreement with the specs, the idea of the drain on the Logic board isn't valid/needed.
On the faded display segments, I can only repeat what I have read, and that is that it is display aging issue without any fix short of transplanting a better display. I don't know if the low voltages that you measured are a symptom of dim segments on a display, or the cause of it.
