I've come across a non-functional Nakamichi PA-7II amp. It will not power on. I found that R504, the 20W resistor on the limiter PCB is blown. So I replaced it, and it immediately blew again. The thermal fuse in series with it is intact. Does anyone have any ideas as to what might be causing the resistor to blow?
I'm returning to this project after years of letting it sit. I replaced the blown R504 with a light bulb, as suggested. When powered on, the voltage back from the secondary seems to vary from 0 to 1.3 VAC. If I disconnect the secondary and measure it, I get 1.7ohms. Is this telling me the transformer is toast?
Does the light bulb you installed in place of the current limiting resistor stay brightly lit or dim?
1.7 ohms across one of the secondaries sounds about right.
According to the schematic the transformer primary has an internal thermal fuse inline with it. If you're only seeing 1.3VAC across one of the secondaries under power then you may want to ensure that the xfrmr primaries are not open due to this fuse.
1.7 ohms across one of the secondaries sounds about right.
According to the schematic the transformer primary has an internal thermal fuse inline with it. If you're only seeing 1.3VAC across one of the secondaries under power then you may want to ensure that the xfrmr primaries are not open due to this fuse.
The bulb didn't seem to light at all, though it definitely allowed the 120 VAC to reach the other side of the resistor pads (I measured it), so it effectively took the place of the resistor. It was a 75W incandescent, for what it's worth.
Do I understand the purpose of the bulb correctly, in that if it didn't light up, there is no short in the rest of the connected circuit?
Chamberman, thanks for the suggestion. The next step for me is to desolder the 2 wires to the primary from the limiter PCB and measure those. If they are open, that implies the thermal fuse is blown? If the fuse is ok, should I get something in the low single-digit ohms like on the secondary I measured? Are there any options for a replacement transformer if the thermal fuse is blown?
Do I understand the purpose of the bulb correctly, in that if it didn't light up, there is no short in the rest of the connected circuit?
Chamberman, thanks for the suggestion. The next step for me is to desolder the 2 wires to the primary from the limiter PCB and measure those. If they are open, that implies the thermal fuse is blown? If the fuse is ok, should I get something in the low single-digit ohms like on the secondary I measured? Are there any options for a replacement transformer if the thermal fuse is blown?
You are correct in that if the lamp doesn't stay lit then there is no load on it. The more brightly it is lit the higher the load. Typically a class ab amp with low idle bias and a decent amount of filtering capacitance will brightly illuminate the lamp for a few secs then it will dim way down to a low level. You have all of the electronics after the transformer disconnected so you should not see any illumination unless there's a problem.
Also, yes if the primaries measure open then the thermal fuse opened.
Also, yes if the primaries measure open then the thermal fuse opened.
As Wayne had said back in 2015; leave bulb in circuit, disconnect secondary(ies) from rectifier(s) and power up. If bulb glows and goes out (or VERY dim) then the xformer is ok and the problem is in the rectifier(s) or beyond.
The bulb is there to limit the current the circuit can draw and prevent blowing up components. The brighter the bulb gets the more current is flowing. With an open secondary the xformer should draw a bunch on initial turn on and drop to a low level (bulb out or very dim) very quickly (< a second).
TJ
The bulb is there to limit the current the circuit can draw and prevent blowing up components. The brighter the bulb gets the more current is flowing. With an open secondary the xformer should draw a bunch on initial turn on and drop to a low level (bulb out or very dim) very quickly (< a second).
TJ
I soldered the 2nd thermal fuse (on the PCB) back in place after removing the ballast resistor, however I was working perhaps too quickly and it seems that my solder joints were not so good. So with the limiter thermal fuse properly soldered, I see a solid 120vac on the primary.
Powered off, I measure 0.5 ohms on the primary in-circuit (primary still soldered to limiter). So it seems the transformer thermal fuse is not blown!
Now, when I power back on and measure across the two pin connector from the secondary, I get 27.64vac. I connected the secondary back to the limiter PCB, turned power on, and very quickly heard the relay click on! That's great! I also verified that, with the front panel connected to the limiter, the LED goes green.
The question remains why the ballast resistor blew in the first place. I wonder if I had a short to chassis somewhere when I tested that? I have a spare. Should I put it back in place of the bulb and see what happens?
What should I check next? With the PCB below the caps connected to the transformer, but with the amplifier boards disconnected, should I see any voltage on the caps at this point?
Powered off, I measure 0.5 ohms on the primary in-circuit (primary still soldered to limiter). So it seems the transformer thermal fuse is not blown!
Now, when I power back on and measure across the two pin connector from the secondary, I get 27.64vac. I connected the secondary back to the limiter PCB, turned power on, and very quickly heard the relay click on! That's great! I also verified that, with the front panel connected to the limiter, the LED goes green.
The question remains why the ballast resistor blew in the first place. I wonder if I had a short to chassis somewhere when I tested that? I have a spare. Should I put it back in place of the bulb and see what happens?
What should I check next? With the PCB below the caps connected to the transformer, but with the amplifier boards disconnected, should I see any voltage on the caps at this point?
If the slow start relay fails to engage and short out the resistor then the resistor will overheat. This is the reason for the thermal fuse on the limiter board but if the thermal fuse isn't touching the resistor body then it'll not open as it should and will allow the resistor to overheat. Maybe something is not right with the relay circuit and its not engaging all the time or the relay contacts are in bad shape causing a high resistance there.
The lamp you have in place of the resistor will not save you if the limiter board relay engages and bypasses it so I would remove the lamp from the circuit and build the lamp into a dim bulb tester at this point. Then use the DBT from this point forward and have the amplifier connected to the wall through the DBT for all of your troubleshooting. It may save you from doing some really bad damage. Looking at the schematic this amp doesn't use fuses in the DC rails so it can do a lot of damage in the time it takes to blow the AC mains fuse and discharge the primary caps. You may also want to install a lower value AC mains fuse while you're testing the unit.
You should disconnect the channel boards before connecting the secondaries to the rectifiers and testing the power supply. In answer to your question, If the main caps are connected to the secondaries you should see DC rail voltages across them even with the channel boards disconnected.
Remember the DBT lamp should glow brightly for 3 - 4 secs while charging the PS caps and then go dim, if it doesn't dim then a short is present in the PS. If the power supply comes up okay with the channels disconnected and the DBT inline then discharge the supply caps and connect one channel at a time. If the DBT lamp glows brightly with one channel or the other connected then try turning the bias potentiometer to its minimum position for that channel.
If at any point the DBT does not go dim then you've found your problem area to dig deeper into.
The lamp you have in place of the resistor will not save you if the limiter board relay engages and bypasses it so I would remove the lamp from the circuit and build the lamp into a dim bulb tester at this point. Then use the DBT from this point forward and have the amplifier connected to the wall through the DBT for all of your troubleshooting. It may save you from doing some really bad damage. Looking at the schematic this amp doesn't use fuses in the DC rails so it can do a lot of damage in the time it takes to blow the AC mains fuse and discharge the primary caps. You may also want to install a lower value AC mains fuse while you're testing the unit.
You should disconnect the channel boards before connecting the secondaries to the rectifiers and testing the power supply. In answer to your question, If the main caps are connected to the secondaries you should see DC rail voltages across them even with the channel boards disconnected.
Remember the DBT lamp should glow brightly for 3 - 4 secs while charging the PS caps and then go dim, if it doesn't dim then a short is present in the PS. If the power supply comes up okay with the channels disconnected and the DBT inline then discharge the supply caps and connect one channel at a time. If the DBT lamp glows brightly with one channel or the other connected then try turning the bias potentiometer to its minimum position for that channel.
If at any point the DBT does not go dim then you've found your problem area to dig deeper into.
Last edited:
I tried a bunch of power cycles with the bulb still in place of the 20W resistor, and the relay always seems to engage in about a quarter of a second. This timing might be different if I swap the bulb for the resistor, but the relay seems to be working at a basic level. Also, when I removed the resistor, the thermal fuse was touching the 20W resistor body, and was held in place with a sort of strap formed by what looks like heat shrink material. It doesn't seem like a very good thermal connection to me, but it seemed to be original.
Also, the schematic says the 20W resistor is a 2.2ohm. The original part I pulled off is a Noble RGU20L, and it says 2.2ohmK. The K is throwing me, and seems to imply a 2.2k ohm resistor. But I'm pretty sure that can't be right because the current would be fairly low through that. And it's blown so I can't measure it.
This weekend when I have time I will head to the hardware store to buy parts and then build a DBT. Thanks for all the help so far!
Also, the schematic says the 20W resistor is a 2.2ohm. The original part I pulled off is a Noble RGU20L, and it says 2.2ohmK. The K is throwing me, and seems to imply a 2.2k ohm resistor. But I'm pretty sure that can't be right because the current would be fairly low through that. And it's blown so I can't measure it.
This weekend when I have time I will head to the hardware store to buy parts and then build a DBT. Thanks for all the help so far!
I built a DBT with some instructions I found online. This will be very handy in future projects!
Placing the DBT inline with the amp, and with the amplifier boards still disconnected, I powered up the amp. The bulb was lit briefly, and then the relay clicked on. I measured the voltage at the power supply caps and got roughly 80v on each, which seems about right.
After that I connected both amplifier boards. The DBT bulb did not dim so I turned it off. With the left channel disconnected, the bulb still did not dim. With the right channel disconnected and the left channel reconnected, the bulb dimmed.
I'm going to focus on the right channel for the moment and disconnect the left. First I will try turning down the bias pot to it's minimum as Chamberman suggests, but I am out of time for the day. If that doesn't fix it, I must have a short somewhere on the board.
Placing the DBT inline with the amp, and with the amplifier boards still disconnected, I powered up the amp. The bulb was lit briefly, and then the relay clicked on. I measured the voltage at the power supply caps and got roughly 80v on each, which seems about right.
After that I connected both amplifier boards. The DBT bulb did not dim so I turned it off. With the left channel disconnected, the bulb still did not dim. With the right channel disconnected and the left channel reconnected, the bulb dimmed.
I'm going to focus on the right channel for the moment and disconnect the left. First I will try turning down the bias pot to it's minimum as Chamberman suggests, but I am out of time for the day. If that doesn't fix it, I must have a short somewhere on the board.
It sounds like you've got the problem isolated. If it was previously blowing the AC mains fuse then I'd expect a catastrophic failure of that right channel with most likely one or more outputs shorted. Luckily the outputs and drivers are the same in this version and they're readily available through digikey. Be sure you replace both the drivers and outputs if an output is shorted.
Wow, I just measured the transistors mounted to the heat sink and compared right versus left. Pretty much all of them are different, but I have some shorted pins on the left channel on transistors that are not shorted on the right, on 7 of the C3856 and 2 of the A1492. I think I'll start by replacing those, then see if some of the differences in the megaohm regime on the other transistors go away.
Just replace the whole left channel outputs on both the 1492 & 3856's. Also the 4 drivers (same devices).
You should also check Q114/Q115 and make sure they didn't fail. The A3298/A1306 devices are long out of prod but the 2SC4793/A1837 took the place of those older devices in a lot of amps and are produced by UTC now.
When you put it back together I'd just put one driver in and one P & N output transistor. Make sure one of the devices you install is Q126 as according to the schematic it's the one that bias is measured across. Then power up with the DBT. This will limit the damage if there's another issue. If DC offset is good with one device in and it seems like bias will adjust then populate the full channel.
I'd also take the opportunity while it's apart to replace the bias and offset pots with sealed multi turns.
You should also check Q114/Q115 and make sure they didn't fail. The A3298/A1306 devices are long out of prod but the 2SC4793/A1837 took the place of those older devices in a lot of amps and are produced by UTC now.
When you put it back together I'd just put one driver in and one P & N output transistor. Make sure one of the devices you install is Q126 as according to the schematic it's the one that bias is measured across. Then power up with the DBT. This will limit the damage if there's another issue. If DC offset is good with one device in and it seems like bias will adjust then populate the full channel.
I'd also take the opportunity while it's apart to replace the bias and offset pots with sealed multi turns.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.