Power supply board damage - reversed polarity on two CONs

[vladc77]

Hello everyone,

I accidentally incorrectly connected two connectors from the power supply board to the logic board of the Nakamichi CR-7A vintage deck. They both look the same. These are the CN4 to the CN5 cable connectors on schema. It resulted in some damage to the power supply board that I cannot identify. After connecting them properly, the deck stopped playing, rewind, or fast forward. I found out that may be reverse polarity is happening on the two connectors on the power supply board. These are CN40 and CN5. I need help to identify the reason for the voltage swap from +12V to -12V on pins 1 and 2 of the CN40 power supply board connector and the opposite on Pins 5 and 6 from -12V to +12V. In addition, the same voltage values are in the CN5 power supply board while they supposed to be different. It may be that the same failure affects both. I marked these connectors in red on the schema. I checked the capacitors and most resistors on the circuit board. they all seem to be OK. All fuses are good. I hope anyone more experienced can help to suggest what can cause this reverse polarity. Thank you in advance!

Below are CN40 and CN5 voltage measurements on the power supply board.
-------------------------------------------
CN40
-------------------------------------------
Pin 1: -12.2v (should be +12v)
Pin 2: -12.2v (should be +12v)
Pin 3:  0v      (should be GND)
Pin 4:  0v      (should be GND)
Pin 5: +12.3v (should be -12v)
Pin 6: +12.3v (should be -12v)
-------------------------------------------
-------------------------------------------
CN5
-------------------------------------------
Pin 1: -12.2v (should be +12v)
Pin 2: -12.2v (should be +5.6v)
Pin 3:  0v      (should be -12v)
Pin 4:  0v      (should be GND)
Pin 5: +12.3v (should be +15v)
Pin 6: +12.3v (should be AC)
-------------------------------------------

[MathWizard]

What voltages to you get after the 3 rectifiers, D401, 404, 407 ? So that would be from GND to the top of the caps C401, 410, 413.

[vladc77]

Thanks MathWizard!

Here are my voltage testing results. I also tested the voltage on rectifiers and IC401 since it is related to CN5. IC401 voltage is correct for all pins.

-------------------------------------------
+ sign is for positive pins and - for negative.
-------------------------------------------
Capacitors
-------------------------------------------
C401: + (+20.5V) | - (0V)     
C402: + (+20.5V) | - (0V)
C406: + (0V)        | - (-21V)
---
C410: + (+9.3V)   | - (0V)
C413: + (+25.4V) | - (0V)
C414: + (0V)        | - (-24.5V)
---
-------------------------------------------
Rectifiers voltage
-------------------------------------------
D401: + (+20.5V) | - (-21V)    | voltage with leads connected to + and - = +41.5v
---
D404: + (+9.4V)   | - (0)         | voltage with leads connected to + and - = +9.4v
---
D407: + (+25.4V) | - (-25.5V)  | voltage with leads connected to + and - = +50.4v
-------------------------------------------

-------------------------------------------
IC 401
-------------------------------------------
Pin 1 has to be -2v         I have -2.1v
Pin 2 has to be 0v          I have  0v
Pin 3 has to be 0v          I have  0v
Pin 4 has to be -12v       I have -12v
Pin 5 has to be +5v        I have +5v   
Pin 6 has to be +5v        I have +5v
Pin 7 has to be +2          I have +2v
Pin 8 has to be +12v      I have +12v
-------------------------------------------

[MathWizard]

Those sound ok, the op-amp IC401 has it's Vcc =+12V on p8 and should be directly connected to CN40-p1,2 and CN5-p1. IC401's Vee=-12V is p4, and that should go right to CN4-p5,6 and CN5-p3

[vladc77]

Do all voltage values sound ok, including caps and rectifiers? Please clarify. What then can cause getting incorrect voltage values for CN40 and CN5? Those are not correct for all pins. Thanks.

[MathWizard]

Yeah the cap and diode voltages all sound fine. Sounds more like CN40 was measured backwards, either with the DMM probes with the polarity backwards, or the labelling on the PCB, or on the cable itself ?

Unhook CN40 and 5, and with it unpowered for awhile, you should measure well under 1 ohm, from IC401-p8, to CN40-p1,2 and CN5-p1 on the PCB.

For the 15V rail, that's coming from ZD403, C412 should be just over 15V, and 15V goes right to CN5p5 and CN6-p2. Likewise for the 5.6V rail, it should be on C411, and go right to those CN connectors. Zener regulator circuits aren't super accurate, so they could be up or down by 0.5V, and vary even more with a lot of output current.

[vladc77]

MathWizard, thank you for checking this for me! It is very appreciated!

Regarding CN40 being measured backward. I checked it multiple times. I connected the black probe to the GND pins. I also connected the black probe to the metal frame of the deck. The results are the same and voltage values are in reverse.

For resistance, I  measured resistance and got different and unstable values.

The values for IC401-p8 - to CN40-p1,2 start from 225 kOhm and go slowly down. At other times, it starts from lower values and goes up, like 4.8 kOhm to 80 kOhm
The values for IC401-p8 - to CN5-p1 behave similarly. It starts from 60kOhm and goes slowly up to 210 kOmh. At other times, it starts from higher values and goes down.
Can be this due to a capacitor in this line?

Regarding 15V rail, ZD403 has 16.8v.

CN5p5 gets +12.3v - supposed to have +15v
CN5p3 gets 0v        - supposed to have -12v
CN6p2 gets +12.3v - supposed to have +15v

C412 on + has +19.5v
C411 on + has +6.3v

Do all these measurements help to understand what is wrong? Any ideas?

[MathWizard]

Ok I never tried the circuit in a sim, but those PSU voltages sound close enough. I'd expect maybe 1-2kOhm on my DMM from GND to 12V rail, or -12V. But according to the schem. there should be just a trace /wire from those pins to IC401-p8. So what does IC401-p4 to CN40-p1,2 measure.  Besides some PCB trace or solder joints breaking around those connectors, IDK what's wrong.

What's on CN-7 ?

[vladc77]

Other tests indicate a good connection. I assume it is OK to test this way. Here are the results:

CN40 measurements:
IC401-p4 -> CN40-p1 = 0.18ohm
IC401-p4 -> CN40-p2 = 0.14ohm

CN7 gets all the correct voltage
CN7-p1 = -15.5v
CN7-p2 =  GND (0)
CN7-p3 = +19v

I did a few other measurements:
IC401-p4 -> CN5-p1 = 0.15ohm
IC401-p4 -> CN5-p2 = 0.17ohm

There is a good connection between IC401-p8 and CN40p5,6 and CN5p5,6.
IC401-p8 -> CN40-p5 = 0.17ohm
IC401-p8 -> CN40-p6 = 0.28ohm

IC401-p8 -> CN5-p5 = 0.15ohm
IC401-p8 -> CN5-p6 = 0.15ohm

Do these values make sense?

[vladc77]

I think the Nakamichi documentation may have errors. According to the attached screenshot (PowerSupplyBoard_03.png) of the power supply board trace and locations view, IC401p8 (+12v) goes to CN40p5,6. It seems it also goes to CN5p5,6. The same goes for IC401p4 (-12v) to CN40p1,2 and CN5p1,2. If that is true, it looks like CN40 is within specs. This view explains why CN40,p5,6 get +12v.

If assume the documentation error exists, then the problem is only with CN5.
1. CN5p2 gets +12v but should have +5.6v
2. CN5p3 gets 0v but should have -12v
3. CN5p5 gets -12v from IC401p4 but should +15v.
4. CN5p6 gets -12v but in schema, it is marked as AC power.

I hope this explains the confusion about incorrect voltages for CN40. With that being noticed, what would be the cause of CN5p2, p3, and possibly p5 and p6 being outside the correct voltage?

[vladc77]

According to the attached "PCB-Trace-View.png" diagram CN-5pin1 (+12v) and CN-5pin2 (+5.6v) are directly connected to one trace line that is from IC401p8 (+12v). There is no trace line from C411 to CN-5pin2 to deliver 5.6v as it can be seen in the electrical schema. CN-5pin1 and CN-5pin2 are directly connected. How CN-5pin2 can have (+5.6v) but CN-5pin1 (+12v)? I missing something here.

The same for CN5pin3. How CN5pin3 can get -12v? CN5pin3 is not connected to IC401pin4 but to the ground accoring to "PCB-Trace-View.png". Where -12v are coming from to CN5pin3?

Finally, CN5pin5 (+15v). Looking at the PCB Trace view, CN5pin5 (+15v) is connected to IC401pin4 (-12v) and has no connection to ZD403 (+15v). How CN5pin5 can get +15v? Currently, it receives -12v which is not as specified in the electrical schema. It should be receiving +15v from ZD403.

I hope you can educate me on this. It may help to trace the current issue. I already removed all transistors for testing, most caps, and some diodes and they all seem to be OK. The caps are not 100% perfect but were working fine before. So, I do not know what else to check. Thank you again!

[vladc77]

I still cannot resolve the issue with connector CN-5 pins 2 and 3 not getting the correct voltage. It seems like all components in the chain are good while testing.
---------
CN-5
---------
1. CN5p2 gets +12v but should have +5.6v
2. CN5p3 gets 0v but should have -12v
3. CN5p5 gets -12v from IC401p4 but should +15v
4. CN5p6 gets -12v but in the schema, it is marked as AC Power

The voltage values are the same as for CN40 pins. Looking at PCB trace lines, CN40 pins are connected to the same trace lines as CN-5 pins. The most difficult is to understand how these values can be different. The way I test voltage is simply by connecting a black probe to the ground and a red probe to the pins.

According to the electric schema diagram, CN-6 connector pins 2 and 3 are getting voltage from the same sources as CN-5 connector pins 2 and 3 and even directly connected. I cannot find trace lines on the PCB that connect CN-5 p2 and  CN-5 p3 to the sources as CN-6 pins. CN-6 pins get voltages very close to the specified in the schema while CN-5 pins are not.
---------
CN-6
---------
CN6p2 - (+15v) from ZD403
CN6p3 - (+5.6v) from C411

I did not find damaged components. I tested them on the PCB circuit board and removed the caps, diodes, and transistors. I am stuck here. Can anyone help to understand what is wrong?

[MathWizard]

It's a lot to follow along with, but really I'd say forget the schematics, and make your own. I find it fun especially on big old stuff that's easy to trace out. I mapped out most all of a 1980's stereo this winter or spring, as part of a repair/modification. I also removed all the electrolytic caps, to test them, but also just to make it way easier to map and check the rest of the circuit unpowered with a DMM. At some point I started powering different sections of my bench PSU, making sure it all seemed good.

But I keep forgetting to ask or say that I'd fully disconnect the PSU from the rest of the stereo. And then map it all out and make sure those voltages make sense. And if you have any power resistors, or dummy loads, measure the DC and AC ripple.

And for the rest of the stereo, if you like soldering, and writing down which cap goes back where, I'd remove the electro's, and then fly through rest of the circuits with a DMM. Either way, I'd separate each board, and just go through unpowered with a DMM.

Some people would just apply power, and just go check all over the places that way. I'd want a lot of wires soldered on for that, but I'd want to know where I was putting them. But sometimes too, with a current limited voltage source, it's ok to slowly up the voltage to a board, and see if it starts using way too much current. Depending on your external PSU, that can be pretty safe and quick way to narrow down a problem. Or inject voltages into just 1 section of a circuit. Sometimes I'll cut traces, and remove a few parts, just to power up 1 section to see what it does.

And I'd make the circuit or parts of it in LTSpice, anything I can't calculate or don't know about, LTSpice usually does.

[vladc77]

Thank you, MathWizard!

To clarify, are you suggesting to try connecting the Power Supply Board to the rest of the boards and checking the voltage with loads along with other deeper investigation methods? I am afraid to damage anything else.

When I installed connectors incorrectly, a 12V load could have been coming from the logic board to the CN-5 connector on the PSU which itself was supplying voltage. it is unlikely it damaged much. I think that because all fusible resistors are good, all fuses are also not damaged and I did not find any damaged parts yet. I removed almost all parts of the chain. All other connectors on the PSU have the correct voltage even IC401.

The most confusing is understanding how CN5p2 and CN5p3 get their voltage. Looking online, I've read that is possible that pins connected to the same trace may have different voltages similarly as specified in the schema. I cannot find a good article explaining this. Regardless, these pins do not have the voltage as specified in the schema. Maybe, I need to measure the voltage somehow differently.

[MathWizard]

No I'm saying to disconnect the PSU as much as possible, from everything else, and check all the PSU voltages. Yeah I'm forgetting this is an old tape deck, I'd say it could be a nightmare to disassemble, and another to get back together correctly. But that might be the only way to narrow it down, if it's that bad.

But yeah unplug as much as you can, like the motors, see what they read with a DMM, maybe try powering them off some 5-12V supply, but yeah I'd put some resistor in series if I was just going apply power to those motors. If you hooked up some connector backwards at the start, then it would be pretty easy to pop polarized caps, like maybe around a motor, or base-emitter junctions, IC's.

[vladc77]

Yes, all my previous testing was done with all connectors unplugged from other boards. I tested the reel motor with a 9v supply. It works well. I localized my investigation fully to the power board supply.

12v could be loaded into the power board supply from the logic board. I assume pop polarized caps should be only on the PSU board. Correct? Is that possible to correct the polarization of the caps?

At the same time, there are not many caps there. I believe I tested most of them. All connectors except CN-5 pins get proper voltage. I hope I am not missing anything here.

With all that said, I am wondering if there are any other ideas on why CN-5p2 and CN-5p3 may not be getting the voltage as needed while this voltage is available for CN-6 pins. Thank you again!

[vladc77]

MathWizard, I took your advice to forget the schematics and make my own. Thank you!

I decided to trace the source which is receiving +5.6v from the CN-5p2 in the logic board of this Nakamichi. I was surprised to learn that this area is working on +12v voltage. The values indicated in the schematic are just not correct. I am puzzled how this could happen, especially from this top-end manufacturer as Nakamichi. Based on the schematic, I was tracing something that did not exist. The power supply was working correctly all the time. It was a bit of bad luck I connected two connectors incorrectly with one of them had incorrect values listed in the schematic. I was 100% sure something was damaged.

As soon as I realized that, I got a few ideas about what could cause the deck not doing FF, Rewind, and Play. The first idea was a low-hanging fruit option to replace burned light bulbs in the transport that is supposed to provide light for the inside cassette compartment. It looks more like a decorative thing but it is not as I learned now. There are two photosensors there. I thought this light was the only source of stable light for them to operate. As as soon as I replaced the bulb, everything started to work. In the end, it was a relatively easy fix. However, I think the design using the photosensors with that bulb is not the brightest idea. That bulb is not reliable. As soon as it is gone, the deck is gone too.

The only remaining issue is auto-calibration which does not pass on the tapes I have. The rest seems to be working well.