Nakamichi TA-2A - power supply board

Thanks for all this advice. I located C536 and it was bent over, but looks ok, leads are firm, and there isn't any corrosion. I will check the underside later today, though. I will desolder it and check its values; I think I have a comparable ceramic cap in my collection and can try a substitute.

On U503, that IC does have some mild corrosion and tarnishing on the top plate, which my other TA-2A does not have. Perhaps a close inspection of the pins is in order, although at first glance they seem fine, and the entire top board seems fine, in contrast to the power supply board. I will try the depowering/shorting procedure (seems to be my specialty today). And based on prior experience with a battery on a SR-2A, I will desolder from the board rather than trying to take the battery itself out.

Thanks again for your help and insights - tack så mycket. It is very kind of you take the time.
 
Well, Mannegizen, you are some sort of wizard. I'm not out of the woods yet, but your suggestions have given some very promising indications.

I followed your instructions for shorting out the leads on U503, hitting them a few times with foil from underneath (leads good and spiky there) and followed up on top with touching each pin with a needle probe connected to chassis ground. I left the removed battery off for the time being. I touched up the solders on a cap, but looking later it turned out to be C537, not C536 (hard to see/work on the underside). I also touched up some solders on the underside of the power supply board, particularly Q407.

When I plugged the unit back in, I got a green light immediately (no flashing phase), but I don't recall if there was a relay click or not, I was so surprised. The display was also lit, and I could switch between AM and FM, and scan the freq range, and change the scan mode. The bottom row of buttons (input selectors) were not lit, and I had no way to see if they actually functioning. I put my meter on some key spots on the power supply board to see what it was doing. On Q407, I got readings of 0.03/0.09/0.77V for E/C/B, which seems perfectly normal with the schematic. R404 had the same 0.77V on the end connecting with Q407's base, and 5.47V on the other. D405 had 0.09 and 11.0V, again consistent with the schematic. I made sure RY401 was activated (manually) and was able to measure DC offset on the speaker terminals; not sure I believe the readings at this point (<10mv on each side), but they both went to 0 when turned the speaker selector to off. I turned the unit off at this point.

I turned it back on 5 min later, remembering that I should measure those voltages on U503 too, but, alas, the steady yellow light returned. I looked at the voltages that I had just measured on the power supply board, and they too had also returned to the peculiar values I had earlier today.

What I take from this is that the unit is capable of functioning at some reasonable level, the display is not dead, U503 is not dead, Q407 is not dead, and the power supply (the parts of it not on D410) can also work properly. When I first got the green light, I did not know if it was the discharging procedure that led to it, or the solder work. But now I think it is pretty clear it was the discharging, as I touched nothing on the unit from the point of turning it off until turning it back on 5 min later.

So if that is the case, perhaps C536 is faulty? Is replacing it my next best move for tomorrow?

I had not removed C536 earlier, but It would be simple enough to do. I have some no-name, positively tiny 20nF ceramic caps that I could replace C536 (22nF) with, but I can't attest to their quality. I also have very good 10nF Murata high voltage ceramic caps in beautiful blue enamel and 84nF EPCOS/TDK high voltage film caps (bought both of them from Mouser for snubbers) that I could use if the capacitance value is not too critical in this application.

Thanks again for your insights.
 
That is good news, if you managed it to start it once it seems promising that it will be fixable.

The non blinking at the power on you had, is due to the missing +12V. I get the same behaviour if I disconnect CN-1.

I'll rub my grey matter some more when I get back from work, got some ideas, mostly around eliminating potential candidates of problem sources.

The tinfoil trick have I learned from "Mooly" over at diyaudio.
 
OK, thanks.

I discovered this morning that last night, I failed to reconnect CN-7 on the Logic board (it is a little 2-wire connector and was hidden from my view by the rest of the cable harness. The schematic indicates it goes to the Balance board, so probably inconsequential. I reconnected it.

This morning I repeated the foil treatment and restarted, and it is solid green again. I grabbed the TA-2A remote I have from my other unit, and the TA-2A does respond to several remote functions. The display responded to button presses by flashing a green LED, and it would switch AM-FM, scan up and down, mute (the little green LED next to the mute button would light) and unmute. No response to volume, but that is probably not powered right now.

I made some additional measurements in the green state, in case they are useful:

U503
3_____5.45
28____0.12
29____5.26

CN-9
1____4.91
2____5.22 ****(this is the only one that differs from yours: you had 10V here)
3____5.52
4____5.42
5____5.56
6____0
7____11.55

U401
11.56—0.59—5.55 ****(more normal than the 13.9V I measured yesterday while in the orange state)

D405
0.06—11.56

And there is 11.56V on the + leg of rectifier D401
 
CN-7 is for the volume motor, does not need to be connected for startup.
As you connected this one before last tinfoil reset, you should have been able to adjust volume with remote.
But I think we save that problem for later...

U503 measurements looks OK.

CN-9
2____5.22 ****(this is the only one that differs from yours: you had 10V here)

This is because you don't probably don't have the +12V, since D410 ( not D401 that I wrote first) gives you 11V (I have 20V there.)

I get 5V there if I disconnects CN-1 (which comes from what I think is a mute control circuit on the main board, Q121-123. The purpose for that circuit, (I think, pure speculation here), is to mute sound out put by setting voltage on pin19 on U204 to zero before turning of speaker relay to avoid sparks in the relay contacts.)
It is also needed to make the power led flash green on power on, how that works exactly I don't know since the flashing (and the solid green) is actually produced by pin 29 on U503, feeding base of Q504 on the control board.

But that circuit is not needed to turn unit on tho.

D401 has probably a diode that is open, but before replacing it I would replace all components in the 30V circuit, since my in my experience any problem with the 30V circuit will kill D401 eventually.

D413,414,415 C416,417,418,419 Q408 ZD416,417 and check that R416 is OK.

Actually you'll probably replace all old electrolyte caps in he power supply.

But before you do that, simply remove F401 to eliminate that a bad 12V is causing your power on problem. (It shouldn't but it's wort a test.)
 
Last edited:
So back to actually troubleshooting the start up problem.

The cause can be many things, I would start by eliminating as much as probably by disconnecting as many of the connectors on the control board that isn't needed for the unit to start up.

CN-1 (Power of muting circuit)
CN-2 (30V for tuner, 12V for some stuff on display, stereo indicator led )
CN-3 (Mostly tuner related)
CN-4 (Tuner voltage control)
CN-5 (External units remote control board, tape, CD)
CN-6 (External units remote control board, tape, CD)
CN-7 (Volume pot motor)
CN-8 (For input selectors, Phono, CD, ...)
CN-11 (Tone board mute control)
CN-14 (Tuner "Voltage control oscillator" stuff )

With CN-8 disconnected you will not be able to test if you get any sound from the unit since you can't choose any input.

So if the unit will turn on reliably with these connectors disconnected, you can reconnect the connectors one by one until the problem reoccurs.

If the power on problem persist with all these connector off, I would the check/replace C408. I think the purpose of this one is to create a short delay before reset voltage goes to 5V.
The purpose of the 5V reset voltage (pure speculation here tho) is to reset/start U503 after power on. This for the supply voltage to have time to stabilize for a few milliseconds before turning U503 on. If the reset comes to early, U503 might receive noise that can cause it to lock up. (I hope my explanation makes any sense...)

I've tested the function by disconnecting the reset cable on CN-9, the unit goes directly to orange and is turned on when I inserts the cable into CN-9.
But the reset voltage is present on your unit, but it might come to early.
 
Last edited:
On the continued troubleshooting based on the last post: I disconnected all those CN and tried a restart, still yellow. I pulled C408; it tested ok, but I replaced it with a no-name 2.2/50 I had on hand. Touched up some solders on the underside of the power supply board. Restart still gives yellow. Do a grounding of pins on U503, and restart gives green, but I get very peculiar voltages on some of the points on the power supply board that I measured yesterday (e.g., pins on Q407, and the + leg of D401). I switched the unit off, and restarted 15 sec later, and was yellow again. Checked underside of board again, confirming continuity between my solders and elsewhere on the traces that my solders were connected to, and all OK.

Reconnected all the unhooked CNs on the Logic board, drained all the pins on U503, and restarted, got solid green. Rechecked the voltages on the power supply board and all OK again (D401, Q407).

I will comment on the earlier post (#25) in a following post, as I have some confusion on a few points.
 
CN-9, pin 2, 5.22V: low value due to not having 20V on D401?
The + leg of D401 does give me 11.56V (when operating correctly), but my reading of the schematic indicates leads me to believe that is the right value (voltage label just to the right of U401 on the schematic for PS PCB). Could you have been thinking about the other rectifier on the PS board, D410, which is marked to give 20V? At this point, my D410 is dead (died in middle of post #1, and my F401 fuse went out there, so I do not have any of the voltages that are supplied downstream of D410 on CN-21 (30V and 12V). As to what is put out by D401, is it possible that yours gives 20V because you have a TA-2E, vs. my TA-2A?

Looking further at the PS schematic, I see that CN-9 pin 2 (labeled "Power Off") connects to the collector of Q403, and should have 11.65V on it, but I am getting only that problematic 5.22V. Probably a problem there (as this should be independent of the D410 circuit) and when I finish this post I will see if I can check the voltages on Q403 to see if I have a problem there.

D401 - possibly a diode open?
I am thinking you meant D410 here, but not sure. If D401 should have 20V, mine is faulty. Anyway, I did order several W02M yesterday from Digi-Key (they only had a few, and Mouser was out of them until at least next month).

The rest
My Digi-Key order includes the diodes and zeners to do the 30V circuit. It has already shipped, and I should have all the parts mid-week. I have already replaced all the E-caps on the PS board except C408 and C407 as I did not have quality caps on hand for those (but checked and replaced C408 this morning, as noted above). For the ceramic caps on the 30V circuit, I only have no-name ones on hand, but I will see how things are once I get the new D410 (W02M) installed and F401 replaced. I am mindful of the first things I discovered on the PS board was that C403 and C412 had failed before I got the unit. These two caps, if I understand the circuit, function as smoothing caps for the output of D401 and D410, respectively, and if so, the entire rest of the PS board received very ripply DC for however long it was switch on with those bad caps.

Lastly, I just want to express how thankful I am for the time and effort you have been putting in to help me. I greatly appreciate it. I am also amazed at how much you know about the digital part of the circuits, which are mostly quite murky to me. I thought those large ICs on the Logic board were mostly tending the LCD display, but I have now learned how much else they do. I had forgotten that by the late 1980s, the digital end had already penetrated even stereo gear (outside of the CD, of course).

BTW, I also ordered replacement for the SPST relay, RY401, because I am suspicious of it. Finding a replacement with the right electrical characteristics and pin geometry was tedious, but I think this will work (I hope so, because it is what I ordered). The original had a TV5 spec, but I was unable to find current one with that spec (perhaps the spec is obsolete?).

https://www.digikey.com/product-det...er-brumfield-relays/RT334012/PB964-ND/1095288

datasheet pdf is here:
http://www.te.com/commerce/Document...srchrtrv&DocNm=RT1&DocType=DS&DocLang=English
 
I have worked on 5 Nak SR and TA receivers and have found that the relays that malfunctioned were due to bad solder joints at the pin to mother board connection, and not the contacts or coils. I my have mentioned this before, but just a reminder. Since I reheated these joints, I have not had any more trouble. I repaired 3 of the 5 units this way. I know relays can go bad but it wasn't the case in my units.
 
OK, makes a lot more sense. Anything to do with D410 is not functional until I get those parts replace.

However, the 5.22V on pin 2 (CN-9) seems to be only dependent on the part of the PS board supplied by D401, and I should get 11.5V or so there, as indicated by the schematic. I can't quite get to the collector of Q403, but I can read D408, and on the side that connects to Q403 it is 5.20V (5.30V on the other side of that diode). The schematic says it should be 11.5V, so Q403 is bad, or something else is making it bad.

My Digi-Key order includes a bunch of 2SC945 transistors, so I can replace it if needed. I will be checking C407 when in there and replace it if needed, if it could be the cause of Q403 malfunctioning.
 
I have worked on 5 Nak SR and TA receivers and have found that the relays that malfunctioned were due to bad solder joints at the pin to mother board connection, and not the contacts or coils. I my have mentioned this before, but just a reminder. Since I reheated these joints, I have not had any more trouble. I repaired 3 of the 5 units this way. I know relays can go bad but it wasn't the case in my units.
Thanks for that, Freddy. I remember your posts on those solders and I will check them more carefully. I looked at them initially and they looked good, but have not yet reflowed them. I will do that before I consider pulling it.
 
Just been checking around the several Q40x on the PS board, and the only anomaly I found is the aforementioned 5.3V on the collector of Q403, which connects to pin 2 on CN-9 (Power Off). The voltage on pin 1 (Reset) is just a bit low (4.87V vs. 4.96V in the schematic; only of note because in my limited experience, voltages are usually a bit higher than marked).

At this point, i figure it is probably best to wait for the parts order to come in, and after replacing D410, Q403, and better C407 and C408, look around again at these voltages, esp. on CN-9 pin 2 and pin 1. Nonetheless, I am happy to chase any other suggestions if they come up before then.
 
As the unit starts up once after the thin foil trick, it makes me think something happens when you turn the unit off.
I wonder if a bad flyback diode, D405 at relay RY401, can cause a problem like this.
 
Last edited:
That's an intriguing couple of ideas. If something is going bad when the power is turned off, perhaps that 5.3V on the "PowerOff" pin (pin 2 on CN-9), instead of 11.6V, has something to do with it? And that D405 looked awful when I first saw it (it is in the photo in post #1). I thought it was unlikely to be working, but when I pulled it out and cleaned it up, it tested OK, and I reinstalled it as I have no spares. Voltages in circuit seemed OK then, but it could still behave wonky at switch off, I guess. My Digi-Key order has 100 1N4002 diodes in it (they are so cheap), and I will swap it out then.

But if there is one surprising thing I have learned under your tutelage, it is how interconnected/interdependent things are in the circuits in these units, including between the analog and digital sides. It well may be that fixing D410 and restoring +30V, +12V, etc., will clear up a lot of things. The interconnectedness/interdependentness reminds me of the days I spend chasing down a failout on a German mass spectrometer I was using. I never had any formal electronics training (I am an academic, and retired now), but I was being guided by an electronics technician who was oversubscribed and couldn't be there to baby sit the thing. There, there were several different consoles with all their interconnect cable bundles, plus computer control amongst them on an HP-IB bus, and two large binders with all the schematics and cabling diagrams, which is why it took days of effort to trace the problem down. I didn't think I'd see that kind of thing in an old receiver, but at least it is all on one chassis. But come to think about it, the vintage of that mass spectrometer and this TA-2A are within 1 year...
 
No response to volume, but that is probably not powered right now.

Checked the schematics for the remote volume control, it needs 5V, 10V and 12V.
And 12V is missing on your unit atm, this is also the cause for 5V at pin 2 on CN-9 instead of 10V.

Also found some interesting when reading the data sheet for uPD75104CW (U503).

"NOTES ON USING THE P00/INT4, AND RESET PINS
In addition to the functions described in Sections 3.1 and 3.2, an exclusive function for setting the test mode,
in which the internal fuctions of the μ PD75108 are tested (solely used for IC tests), is provided to the P00/INT4 (pin18)
and RESET pins.
If a voltage exceeding V DD is applied to either of these pins, the μ PD75108 is put into test mode. Therefore,
even when the μ PD75108 is in normal operation, if noise exceeding the V DD is input into any of these pins, the
μ PD75108 will enter the test mode, and this will cause problems for normal operation.
As an example, if the wiring to the P00/INT4 pin or the RESET pin is long, stray noise may be picked up
and the above montioned problem may occur.
Therefore, all wiring to these pins must be made short enough to not pick up stray noise. If noise cannot
be avoided, suppress the noise using a capacitor or diode as shown in the figure below"

I managed to get my unit to lock up a few times by inserting 10V on pin 18 ( Used power off voltage on jumper JS107 (next to R517). Turning my unit of and on cleared the "test mode" tho.

For what I can see, pin 18 on U503 don't have any capacitor connected to ground. Reset pin 45 has it.
Pin 18 is connected to 5V via R517.
So noise on the 5V during power off could set U503 into "test mode".
It should be cleared when the unit is powered off, but you might also have a grounding problem.
 
Good to know about the remote.

Thanks for looking further on the U503, and pin 18. I might have a ground issue, something to check. I saw that pin 18 also routes to the "Down", "Mute", "4-9" and "Video" buttons, but it is probably just powering those, and I would tend to agree with your thinking that it is something happening on power-off, not during normal "on" operation. I guess my general thought about this is I wonder if, with the present lack of +30V and +12V, that something is happening that is not supposed to, be that noise on pin 18 or just something wrong elsewhere.

On that last note, what about pin 2 on CN-9? First, this is connected to the "power off" pin on U503. Sort of sounds like it could be a red flag. Tracing from pin 3 on U503, it is touched by R5114 (connected to +5V on the other side), passes through D512 and then touched by the downstream side D513 before it gets to pin 2 of CN-9; not sure what all that does, but the schematic does indicate that 5.22V is the normal state for pin 3 of U503 while the unit is on. Is it that during power off, that pin needs to go high, to 10V, to effect a proper power down cycle, and because that is not happening, it will not restart properly?

Second, and of course related, you mention above that me getting only 5.22V on pin 2 of CN-9 (earlier you mentioned you have 10V there), and that the reason for my low value is related to the loss of 12V (presumably from the D410-fed circuit). OK, if that is what it is, great, because when I fix the D410 circuit, that will be restored. But I just reviewed again the PS schematic and as best as I can tell, pin 2 on CN-9 is fed solely by Q403 and that is solely downstream of D401. I think my Q403 is faulty as I measure only 5.2V on its collector, instead of the 11.65V indicated on the schematic. Once I replace Q403, hopefully it will be putting out the proper 11.65V, and I will have the right voltage on pin 2.

Either way, the pin 2 issue should be fixed once I have my parts. I guess it boils down to the question: is the present inability to supply 10-11V to pin 2 of CN-9 the root cause of my startup problem? Is that 10-11V is what needs to be there to send a proper "power off" signal to U503, so that it resets properly?
 
Follow up: I have this other working TA-2A, and I should use it more, now that I am starting to understand more. I checked its Q403 has the proper 11.6V on its collector, and delivers that to pin 2 on CN-9. While that 11.6V sits on pin 2 of CN-9 on the Logic board, pin 3 of U503 has the right 5.3V on it. So I am hoping that is what is needed.

When checking that voltage on pin 3, I slipped the probe accidentally onto pin 2 of U503 (which is ground), and the unit went right into yellow on the power switch. Turning off, and then on again, restored the green, after the suitable flashing for a few seconds. It would be great to put a o-scope on pin 3 of U503 and try to catch a voltage change during power off, but I don't have one.
 
I'm pretty sure that the 5V on pin 2 on CN-9 isn't a problem, I get 5V also if I disconnects CN-1 and my unit works fine then.

The voltage on pin 2 at CN-9 has two sources, 5V from pin 3 on U503 via D512.
10V from the circuit made up by Q121,122,123 on the main board which goes via CN-1 to D513 and from there to pin2 on CN-9.

I'm not clear on what all the purposes of the "power off" is.

It mutes sound when turning the unit off. Tested this by disconnecting the cable on pin 2 at CN-9.
The result was that I got some noise from the speakers as I turned the unit off.

What Q403 and Q404 does in the power supply is unknown to me, perhaps they are a protection circuit that are supposed to turn of the unit if U401 fails and lets 10V trough or if the voltage from D402/403 gets too high.
Or if they just have a regulation function too keep the "power off" voltage stable.

But if Q403 or Q404 is faulty they might be cause of immediate turn off of the unit or produce noise on the 5V rail. (well they don't seem create any problem if you tin foil U503 prior to turn on...)

A oscilloscope might detect noise on the 5V rail?
 
Last edited:
Follow up: I have this other working TA-2A, and I should use it more, now that I am starting to understand more. I checked its Q403 has the proper 11.6V on its collector, and delivers that to pin 2 on CN-9.

It is the other way around, Collector on Q403 is feed from the 10V from CN-1 via pin 2 on CN-9
If I disconnects the cable on pin 2 at CN-9, the voltage on collector drops too 5V as it's only feed from D408.
 
Back
Top Bottom