NAP-140 Clone Amp Kit on eBay

There is no pinout information printed on board. Do you know the pin info? 20180219_220227.jpg
 
Member
Joined 2010
Paid Member
I've got zerozone nap140 clone boards. I tried everyting but cannot succeed about bias....I checked every component.Couldnt find anything yet. Transistor next to bias pot is bc550- i couldnt find ztx384..

What am I missing. Any idea?....
If you substitute parts, you need to be sure of pinouts and these are not always as original supplier datasheets show, sometimes varying between manufacturers. When the substitutes are fakes and copies too, you could have a big problem unless you learn to test for pinout (lead identity) yourself. You don't say anything about the type of current source or VI limiter transistors you are using, so I presume these are copies of original MPSA types or more of the supplier's substitutes.

Get yourself any cheap transistor tester toy from Ebay - that will identify pinouts of separate transistors correctly if you can't work them out from multimeter readings and methods shown on many beginner forums, blogs and websites. It will also give you an indication of Hfe, the DC gain of the transistors, which has to be un-matched a little between the input transistor pair in order to get the output offset voltage down to 15mV or less.

Soldering can look good and still not be making contact due to technique, contamination, stress on connectors etc. Check your work and reflow joints. When powered, measure the Base to Emitter voltage of all transistors - you should be getting around 0.65V across all small transistors, a little less for the power types.

Another surprise for Naim cloners who don't read enough before buying:
The amplifier needs to be enclosed in a suitably small aluminium case for the bias control to work. The Vbe mutiplier transistor is a long way from the output stage and takes 20 mins or more to stabilise after power-up and that applies because the case is the heatsink with a heat spreader bar distributing heat more evenly to the 3.2 mm case, like the original. Even when you have a response to adjustment, bias setting is pointless until the amplifier has warmed up with the cover in pace.
 
Ian, thanks for your reply..

I'm using mpsa transistors for current source&limiter positions. But I don't know if these devices are genuine or fake, also i don't have any idea about their pinouts. I know how to check&identify pinouts of bjts, I checked some of them and use them as is, but the thing is board does not have any info about pinouts. I didn't bought as a kit, only boards and used my parts stash for assembly.

Soldering is ok, I checked every solder joint and after soldering I always check continuity and resistance check between spots.

My suspicion is placement of bjts on board and pinouts is different than mine. And I cannot be %100 sure if its ok or not. Seller does not provide any info about it..

I noted your precious advices.. Thanks a lot.
 
I reversed bc550. Now across the pot reading around 1v and its changing with turning pot. still no voltage across base to base of the drivers... Pot is 2k multi turn..

Here's my thinking....

Let's look at the amplfier circuit in the attached schematic, although itmay not be an exact match to your board but it must be close.

To turn on the output transistors (TR9 & 10) we need around 0.65V across their base-emitter junctions - it's the price of entry. The upper device has a driver (TR7) in the traditional Darlington arrangement where one transistors output (emitter) drives the base input of the other - and both transistors must be on. The two devices need a combined voltage of at least 2 x 0.65V = 1.3V from the base of the driver to the emitter of the output device. The lower device (TR8) is in a Sziklai arrangement instead but all the same the driver transistor output (collector) drives the base of the other. In this case all we need to ensure is that the driver device is turned on and it's power device partner will follow in it's footsteps. There's a diode in series with the emitter of the lower driver, both it and the transistor must be turned on. A diode needs the same 0.65V to turn on and so, again, we'll need around 2 x 0.65V = 1.3V. So between the two bases of the driver transistors (TR7 & 8) we want around 2.3V of bias voltage in total.


The bias voltage is generated by flowing a current through the bias generator (TR5), which produces a voltage drop. This voltage drop is the bias. The bias generator is placed between the collector of the Vas transistor (TR4) which is source of current, and the collector of the lower transistor, TR6 which acts as a current sink. The bias potentiometer controls how much voltage drop the bias transistor will generate. The voltage drop across the bias pot is not the bias voltage by itself. You can measure the bias voltage across the bias generator by measuring the voltage between the collectors of TR4 & TR6. This bias voltage should appear at the bases of the driver transistors, connected through base- resistor networks (the diagram I attached shows a network of resistors and capacitor, most amplifiers use only a single resistor). If these resistors are open circuit then the bias voltage will not appear at the bases of the driver transistors. You can check for this fault easily, there should be no significant voltage between the collector of TR4 & base of TR7, and again no significant voltage between collector of TR6 and TR8. There could be other faults. It would be helpful if you could post voltages at as many places around the circuit as possible - mark up a schematic with the measured voltages and post it.

Once you have adequate bias voltage you can check to see if there is current flowing through the output devices (no load, no speaker connected at the output) by measuring the voltage drop across the two 0R22 power resistors on the power devices. Measure the voltage between the emitter of TR7 and collector of TR8, you want around 50mV to 80mV (others can advise on 'optimal' bias later).
 

Attachments

  • NCC200Notes.jpg
    NCC200Notes.jpg
    363 KB · Views: 476
Last edited:
There is no pinout information printed on board. Do you know the pin info? View attachment 663571

I guess all is good if it works first time. Otherwise, you have to put in the spade work. You should be able to trace out the circuit by comparing a printed circuit schematic, like the one I posted, with your pcb. The pin-out on the pcb will become clear if you do that. Unless you are lucky and find a bad solder joint or something you will need to spend some time on it. What you learn will be very useful for future repairs and future projects.
 
Low bias

Finally I have 2 boards up and running.

I had to give up on one board as it blew a TR4 (ZTX753) and the driver TR7. Even after replacements, and before powering up, I was getting diode readings on the reverse bias of the output diode that was significantly different to the other boards and despite pulling and checking individual components, there was just something not right with that PCB.

I also had an issue with a high DC offset on one of the boards which turned out to be a cooked cap across TR5. I fell in to the 'NAP140 Mystery' trap as described on PFM - I had disconnected the positive rail from the ammeter on that board before the caps had bled down fully.... Lesson learned.

Which brings me to the question - I can only get a maximum of around 20mA current flowing at idle. The bias trimpot does alter the current from around 12mA to around 20mA. This is with the amp closed up and running for quite some time.

Could this be due to probably highest quality chinesium non-genuine output transistors? (2SC2922) or perhaps is the case too good a heatsink and not letting the transistors warm up enough to stabilise (there are many comments on the thread regarding bias being carried out while enclosed).

Is it worth forking out for genuine (again, possibly not) transistors or is there a better choice I can drop in that is more likely to be genuine?
 
Other than the 2922, are the other transistors supplied with the board too? If they were they're probably fakes as well.

I bought a NAP200 board that came as populated and tested. Due to the fake devices they underbiased it to around 7mV instead of the recommended 13mV (30mA). I did not attempt to bias it to the recommended value. Even then the sound however sounded okay.

I've since replaced all sands and capacitors.
 
Other than the 2922, are the other transistors supplied with the board too? If they were they're probably fakes as well.

I bought a NAP200 board that came as populated and tested. Due to the fake devices they underbiased it to around 7mV instead of the recommended 13mV (30mA). I did not attempt to bias it to the recommended value. Even then the sound however sounded okay.

I've since replaced all sands and capacitors.

I bought all new genuine transistors that matched the schematic (ztx, mje and BC). I replaced the ztx384 with BC550C and definitely got the pinout correct. I put all new tants as well.

I didn't replace the 2sc2922 due to not knowing where to source a genuine article.

Also the mpsa06 I couldn't get hold of in the short term, but I replaced the mpsa56.

Even on my 25w test speakers it sounds pretty good so far, but just wonder how I can get that bias up further.
 
I would power it up a section at a time starting with the low current stuff.
Eg: remove drivers and output transistors. Temporarily connect the output feedback point to the collector of the ZTX753. This circuit should draw less than 10mA from the supplies so you can stick a series resistor in the negative supply rail as an emergency stop, like 100 ohms, in case transistors are shorting. You can measure the voltage across this resistor to confirm correct total current. Then power this up and check the voltage at the ZTX753 collector is pretty much zero (the feedback point); by just about zero I mean within +/-50mV, and +/-10mV if its working very well. Check the other voltages in the circuit look right (see Bigun's post). Check that you can vary the voltage across the biasing transistor and the range looks right. Set the bias to its lowest before the next step.

If something is wrong, and you have checked your resistor values, it is probably a transistor that is in the wrong way or is duff. Make sure electrolytic caps are polarized the right way. Don't use tanatlum capacitors yet because they can go open or short circuit if something is wrong.

Once that looks good, disconnect the feedback from the ZTX753 collector and reconnect to the output proper. Solder the drivers and output transistors in place. Replace the 100 ohm resistor in the negative supply (if you used one) with a 10 ohm power resistor (unless you have a current-limited bench supply). This will help prevent melt-down of the output transistors if there is a fault and gives you a convenient way to measure total current. Once powered-up, set the bias current to 10mA or so and measure all the voltages around the output stage.

If something is wrong it is most likely a duff transistor or wrongly oriented one or a diode is the wrong way around. The voltage across the PNP driver emitter diode should be measured. Make sure the thing isn't oscillating - use a scope if you have one or a multimeter set on AC. You should end up with a dc offset like before.

AND before you do any of this, make sure your grounding arrangements are correct. This is really important to avoid instability. Do it the way Naim does it; I'm sure there are photos somewhere in this thread.

:)
 
Last edited:
I would power it up a section at a time starting with the low current stuff.
Eg: remove drivers and output transistors. Temporarily connect the output feedback point to the collector of the ZTX753. This circuit should draw less than 10mA from the supplies so you can stick a series resistor in the negative supply rail as an emergency stop, like 100 ohms, in case transistors are shorting. You can measure the voltage across this resistor to confirm correct total current. Then power this up and check the voltage at the ZTX753 collector is pretty much zero (the feedback point); by just about zero I mean within +/-50mV, and +/-10mV if its working very well. Check the other voltages in the circuit look right (see Bigun's post). Check that you can vary the voltage across the biasing transistor and the range looks right. Set the bias to its lowest before the next step.

If something is wrong, and you have checked your resistor values, it is probably a transistor that is in the wrong way or is duff. Make sure electrolytic caps are polarized the right way. Don't use tanatlum capacitors yet because they can go open or short circuit if something is wrong.

Once that looks good, disconnect the feedback from the ZTX753 collector and reconnect to the output proper. Solder the drivers and output transistors in place. Replace the 100 ohm resistor in the negative supply (if you used one) with a 10 ohm power resistor (unless you have a current-limited bench supply). This will help prevent melt-down of the output transistors if there is a fault and gives you a convenient way to measure total current. Once powered-up, set the bias current to 10mA or so and measure all the voltages around the output stage.

If something is wrong it is most likely a duff transistor or wrongly oriented one or a diode is the wrong way around. The voltage across the PNP driver emitter diode should be measured. Make sure the thing isn't oscillating - use a scope if you have one or a multimeter set on AC. You should end up with a dc offset like before.

AND before you do any of this, make sure your grounding arrangements are correct. This is really important to avoid instability. Do it the way Naim does it; I'm sure there are photos somewhere in this thread.

:)

Thanks for all the tips, I'll try to work my way through the circuit again.

I find it strange that both boards are biasing the same with roughly the same amount of resistance on the trimpot.. but just lower than expected. That's what makes me suspect the output transistors.

I'll see if I can examine the voltage range of the biasing transistor.

The amp is working and sounds great.. I just feel it is being throttled if I am only getting 20mA max.

The grounding is good and as per the Naim arrangement as illustrated here a few times.

I do have an oscilloscope but I managed to let the smoke escape from a 100ohm resistor and the top driver transistor (silly mistake, managed to show the 40v a path to ground via the mains earth of the scope). I think I'll stick to the DMM!
 
The amp is working and sounds great.. I just feel it is being throttled if I am only getting 20mA max.
I haven't been keeping up with your posts but I thought I'd just throw in a general debug process to help you or anyone else reading this stuff.

20mA is plenty through the output transistors. It seems odd that that is the max you can set, though, and is suspicious for that reason. However, I haven't checked you schematic to see what the range ought to be. If it is 20mA total for the amp then that is not enough.
 
Last edited:
I had the same problem too with 2922 and I got them from Digikey. For Malaysia (my country) they have free shipping for orders over $50. Try checking there and you can get other stuff along to reach $50 if they got that for your country.

$10 per transistor, free shipping $60 AUD. I'll have to consider whether to roll the dice on more output transistors or enjoy what the (possibly) fakes are producing.
 
I haven't been keeping up with your posts but I thought I'd just throw in a general debug process to help you are anyone else reading this stuff.

20mA is plenty through the output transistors. It seems odd that that is the max you can set, though, and is suspicious for that reason. However, I haven't checked you schematic to see what the range ought to be. If it is 20mA total for the amp then that is not enough.

I got some 140 clone boards from eBay that seem to follow the schematics posted here.

The schematics say I should be getting around 30 max 34mA current draw from the positive rail but I get around 20mA max. Both channel boards are similar in draw.

I was having trouble getting a mV read across the 2 0.22 ohm output resistors so I switched to reading the current draw directly instead. Makes me wonder now if the difficulty getting a mV read was due to oscillation. I was getting about 1.1-2mV showing as the bias range (should be about 10-14mV across both) however when I had the scope on it, I was getting about 7mV max and the signal was pretty dirty.

I'll check for oscillation at the pnp diode next I think..
 
The input and VAS stages draw about 10mA. At idle, the drivers should draw about 7mA. So if the supplies draw 20mA total then you would expect the output transistors are drawing only 3mA or so.

The voltage between the driver bases should be in the region of 2.8V. If it is higher than this there is a fault in the output stage. Is the diode open-circuit?
 
Last edited:
Member
Joined 2010
Paid Member
...... it sounds pretty good so far, but just wonder how I can get that bias up further.
Don't try to increase bias on a NAP or Nait design amplifier because they are a quasi complementary design which not only has lower THD at a very low bias current but will be more thermally stable there. Raise that current to where you might bias a typical complementary EF design (like every other design you see) and you'll likely soon see the temperature keep rising by itself in thermal runaway followed by smoke .

For best results and stability, the complete amplifier should be run in a compact, closed aluminium case which is also the heatsink, like the original product. They run fairly cool in an average room and are quite reliable if you get the right size case and materials.

The link below is to an interesting experiment on bias level in a Nait2 model - same circuit, lower power. It makes it pretty clear why you don't need to raise bias current, even though can, provided there is more efficient heatsinking located beneath the PCB, to support the higher dissipation whilst maintaining thermal "feedback"to the Vbe multiplier transistor. That's critically important if you raise bias level significantly.

Just one thing about music - when it hits you feel no pain
Scroll down the subject index to the "More Naim stuff" and click on the item "Bias settings in Naim amplifiers". This will open a new page and the article. There's lots of info. on Naim and other UK design products on that archive and also the parent Pink Fish Media forum, where you find commercial UK products and DIY projects discussed together, which seems to suit the active community they've built up over many years. Give at least the d.i.y. forum section an occasional read. d.i.y. | pink fish media