Hi again .
I inspected every cap inside NAD 306 amp and also the service manual .
In the main amp section all electrolytics are polar ones .
I was expecting to find some non-polar electrolytics to replace those with film caps but I can't find any of those .
I added a photo with the scheme .
Maybe this amp have some huge compromises ?.
And maybe the signal is passing thru some polar electrolytics and those can be changed with non-polar ones ?
What needs to be replaced ?
How can I improve this amp ?
Regards
I inspected every cap inside NAD 306 amp and also the service manual .
In the main amp section all electrolytics are polar ones .
I was expecting to find some non-polar electrolytics to replace those with film caps but I can't find any of those .
I added a photo with the scheme .
Maybe this amp have some huge compromises ?.
And maybe the signal is passing thru some polar electrolytics and those can be changed with non-polar ones ?
What needs to be replaced ?
How can I improve this amp ?
Regards
Attachments
Its a very unusual front end but as to improvements, well I would say to leave well alone. Certainly just swapping parts randomly will not achieve anything.
That said, there is one thing that could be improved (as a generalisation) and that would be to replace the speaker relays with solid state versions (which would have to be diy'ed, they are not available as an off the shelf part). Problems due to contacts tarnishing and giving rise to various forms of non linear distortion seem to be getting more common.
That said, there is one thing that could be improved (as a generalisation) and that would be to replace the speaker relays with solid state versions (which would have to be diy'ed, they are not available as an off the shelf part). Problems due to contacts tarnishing and giving rise to various forms of non linear distortion seem to be getting more common.
Quite agree,. If you are an audio design engineer, (engineer means qualified and trained to at least HNC), you could find improvements to be made, as I have done in the past but they are pretty much a good design. RS Components sell some wonderful contact cleaning strips. That will solve the relay issues.
Made by Electrolube.
Made by Electrolube.
It could make a difference yes. Distortion from relay contacts is an insidious problem that has become surprisingly common across a range of equipment. Paradoxically it is worst at low levels and can be temporarily improved by playing the amp loud as this allows the higher current to 'burn' through the oxidation.
It is very real though as witnessed by the many threads that appear and where that is the cause. What is even worse is that this has happened to me.
Its not a design that you will improve just by swapping parts imo.
Bipolar caps are only used where there is a need for a large value cap that can see voltages of either polarity. To use them in other areas would simply be bad engineering. A correctly specified polar cap used correctly is essentially transparent to audio.
If I remember correctly there are 2 or 3 bipolars on tone control pcb. Main pcb originally has tiny tiny 1uF Elna's, yellow/gold I believe, if still on your pcb all will be defective. Many other Elna's can be bad, but the little tiny ones will be completely gone, I haven't found one measuring like a capacitor.
The smaller ones are measuring between 8-12 % Vloss .
First I already changed every cap with just Nichicon KZ all over the amp .
The sound was so transparent and neutral that was boring .
Second step was to take some KZs out and replace those with Elna Cerafines in order for the sound to become more sharp and agressive and it worked .
In my experience, NAD amps react very kindly to mods, as long as you know what you're doing.
A word of warning: do anything at your own risk, make sure you followup every little change with extensive testing, using good measurement equipment. If you don't have any such equipment, I would highly recommend that you REFRAIN from touching this delicate equipment, and skip the text below ENTIRELY.
So, if you have the right equipment and like to play dangerously, I would focus on the following:
1. Main caps:
NAD usually go for dirt-cheap caps, which show their age very ungracefully. A good replacement would be the excellent Mundorf MCAP, ideally 15,000uF/63v.
2. Reducing noise floor level.
This part is a little more complicated and rather tricky: you need to locate the amp's soft spots, and provide them with the best voltage regulators you can afford.
For example, the preamp section is supplied by a dual-tracking regulator called M5230L made ages ago by Mitsubishi Electric. Its spec sheet suggests ripple rejection of 68dB, and output noise of 12uVRMS. These are VERY impressive numbers, which I would treat with a grain of salt. I tend to trust modern regulators much better, such as Linear Tech's LT3015 (Neg) and LT1963 (Pos), or TI's amazing little TPS7A47 (Pos) and TPS7A33 (Neg) - DIYA's commercial section has someone selling PCBs with those little wonders. I would take out the original reg and insert a small pcb with one of the above regs, using as short as possible wires to connect them into the board.
3. Redundant circuitry:
This amp has two functions which are totally irrelevant in everyday use, and add much complexity to the signal path (and personally, I hat complexity). These are the Soft-Clipping and Bridging. Therefore, I would follow the signal path, starting at the relevant switches (BTL = bridging, Soft Clipping), and remove components that are directly connected and are situated at the signal path. This would require a long and exhaustive research at the (complicated) schematic, and will increase the risk of damaging the amplifier if you do it wrong, but it gets you a simpler, and usually faster amplifier.
4. Tonal balance: this is rather easy, and is a matter of personal taste. NAD usually introduce massive low-pass filtering across the signal path, so as to create the household "warm sound". It's usually overdone and results in losing the top-end of the spectrum, so no details, no micro-dynamics, nothing. I suggest fiddling around with the values of the input caps and serial resistors. A right-on value will rip so many benefits and will get you a warm BUT detailed sound.
I'm sure there is much more that can be done. Take away the over engineering and you get yourself a sweet sounding amplifier with amazing dynamics.
Goodluck!
Yair
A word of warning: do anything at your own risk, make sure you followup every little change with extensive testing, using good measurement equipment. If you don't have any such equipment, I would highly recommend that you REFRAIN from touching this delicate equipment, and skip the text below ENTIRELY.
So, if you have the right equipment and like to play dangerously, I would focus on the following:
1. Main caps:
NAD usually go for dirt-cheap caps, which show their age very ungracefully. A good replacement would be the excellent Mundorf MCAP, ideally 15,000uF/63v.
2. Reducing noise floor level.
This part is a little more complicated and rather tricky: you need to locate the amp's soft spots, and provide them with the best voltage regulators you can afford.
For example, the preamp section is supplied by a dual-tracking regulator called M5230L made ages ago by Mitsubishi Electric. Its spec sheet suggests ripple rejection of 68dB, and output noise of 12uVRMS. These are VERY impressive numbers, which I would treat with a grain of salt. I tend to trust modern regulators much better, such as Linear Tech's LT3015 (Neg) and LT1963 (Pos), or TI's amazing little TPS7A47 (Pos) and TPS7A33 (Neg) - DIYA's commercial section has someone selling PCBs with those little wonders. I would take out the original reg and insert a small pcb with one of the above regs, using as short as possible wires to connect them into the board.
3. Redundant circuitry:
This amp has two functions which are totally irrelevant in everyday use, and add much complexity to the signal path (and personally, I hat complexity). These are the Soft-Clipping and Bridging. Therefore, I would follow the signal path, starting at the relevant switches (BTL = bridging, Soft Clipping), and remove components that are directly connected and are situated at the signal path. This would require a long and exhaustive research at the (complicated) schematic, and will increase the risk of damaging the amplifier if you do it wrong, but it gets you a simpler, and usually faster amplifier.
4. Tonal balance: this is rather easy, and is a matter of personal taste. NAD usually introduce massive low-pass filtering across the signal path, so as to create the household "warm sound". It's usually overdone and results in losing the top-end of the spectrum, so no details, no micro-dynamics, nothing. I suggest fiddling around with the values of the input caps and serial resistors. A right-on value will rip so many benefits and will get you a warm BUT detailed sound.
I'm sure there is much more that can be done. Take away the over engineering and you get yourself a sweet sounding amplifier with amazing dynamics.
Goodluck!
Yair
Yes I think this is the main problem . Not enough microdinamics .
Are you speaking about the Line In components that are tacking the signal from the aux source ?
I should add lower cap values to diminish the bass or something ?
Sounds a little complicated .
Sound more simple to play with the bass and highs knobs ?.
I mean by reducing the bass and increase the highs and turn off the soft clipp is not the same thing ?
By reducing the bass and increase the highs the low-pass filter can be change right ?
I was just thinking to replace some parts with the same values like original .
To change the values sound very risky
Last edited:
Ok but the amp uses only polar electrolytics .
To replace all those polar electros bellow 22uf with non polar ones ?
I don;t want to kill the amp
Thks
Hi, I just got one of these amps from a customer who had bought a new NAD as replacement. I have just had a look at it, and done a quick test to see if it was working properly.
It is quite a complex design for a cheap(ish) small amplifier: All the outputs are IRFP240, and at first glance there appears to be four devices per channel (!), but then I saw it has dual rail voltages (41 and 72V), and it looks like two of the IRFP240 are switching the rail voltages to the other two devices (Class G/H stuff). So in effect only two outputs per chn. And of course it has the typical NAD soft clipping and protection circuits, as well as both MM and MC phono stages, so lots and lots of components in it.
After cleaning the volume pot it works perfectly, and has a very clean output, with THD better than -90dB. Harmonics are mixed, but with low-order becoming clear with increasing load. Frequency response is about -1dB at both 20hz and 20Khz, so "nice" symmetry there, but could perhaps benefit from a larger input cap.
So objectively a (surprisingly) good little amplifier, although I don't really like the topology - but that is perhaps just because I am more familiar with the common BJT push-pull EF layout.
If only NAD had spent just $2 more on the plastics for the front panel and knobs, instead of using the lowest quality stuff they could find...
Have a nice weekend
It is quite a complex design for a cheap(ish) small amplifier: All the outputs are IRFP240, and at first glance there appears to be four devices per channel (!), but then I saw it has dual rail voltages (41 and 72V), and it looks like two of the IRFP240 are switching the rail voltages to the other two devices (Class G/H stuff). So in effect only two outputs per chn. And of course it has the typical NAD soft clipping and protection circuits, as well as both MM and MC phono stages, so lots and lots of components in it.
After cleaning the volume pot it works perfectly, and has a very clean output, with THD better than -90dB. Harmonics are mixed, but with low-order becoming clear with increasing load. Frequency response is about -1dB at both 20hz and 20Khz, so "nice" symmetry there, but could perhaps benefit from a larger input cap.
So objectively a (surprisingly) good little amplifier, although I don't really like the topology - but that is perhaps just because I am more familiar with the common BJT push-pull EF layout.
If only NAD had spent just $2 more on the plastics for the front panel and knobs, instead of using the lowest quality stuff they could find...
Have a nice weekend
Dry & baked solder joints..
Well I went back to do a few more tests, set the Bias and align the Soft clipping circuit, and then I got some "funny" measurements - suddenly not working so perfectly after all.
I then had a look on the bottom side of the PCB, and found a lot of dried out solder joints: The worst areas were under the two clusters of power resistors (around R425 and R426 - see the two pictures), but there were many joints all around the board with a ring just starting to show. Almost an hour of reflowing/soldering.
So if anyone has one of these, with unstable sound or some intermittent problems, I would advise to check the condition of all the solder joints first.
If you are aiming to keep it for years, the caps and small transistors around the hottest resistors should probably be replaced, as they will be pretty well baked by now. R425 and R426 (22K) should also have more distance to the PCB, for better airflow around them.
Well I went back to do a few more tests, set the Bias and align the Soft clipping circuit, and then I got some "funny" measurements - suddenly not working so perfectly after all.
I then had a look on the bottom side of the PCB, and found a lot of dried out solder joints: The worst areas were under the two clusters of power resistors (around R425 and R426 - see the two pictures), but there were many joints all around the board with a ring just starting to show. Almost an hour of reflowing/soldering.
So if anyone has one of these, with unstable sound or some intermittent problems, I would advise to check the condition of all the solder joints first.
If you are aiming to keep it for years, the caps and small transistors around the hottest resistors should probably be replaced, as they will be pretty well baked by now. R425 and R426 (22K) should also have more distance to the PCB, for better airflow around them.
Attachments
Have the same amp. Redoing everything. Every single cap either shorted or way out of spec. Lot of work... maybe Nads worst design ever. Let see how it sounds after im done.
All 1uf/100v replaced with Wima Mks or other polycaps. All elecs Nichicon HE or similar. All powerres upgraded to atleast 5w because this is a very hot amp. My mainboard is black almost eveywhere.
All 1uf/100v replaced with Wima Mks or other polycaps. All elecs Nichicon HE or similar. All powerres upgraded to atleast 5w because this is a very hot amp. My mainboard is black almost eveywhere.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.