Hello,
I read the forum very often but I've never had the impulse to post a message ... anyway ... I think it's time to contribute a little bit to the DIY community.
I have built and perfectly working Naim NAP 140 based amp (as sold on eBay).
Having a Naim CD5i (the cheapest one
I often have the impulse to go and buy a Naim amp and to finish my quests, but right now I'm ready to put my clone next to any NAP 150/180, and I'm sure I won't be ashamed.
This is the best amp that I've had in my system and I am in this hobby for more than 15 years. I started with amps created by a friend of mine, electronics engineer – some solid state class A and tube preamps. Later, when Internet has grown and was populated with useful and useless stuff, I started building and upgrading existing schematics ... all that most of you already know: Quad 405 (heavy modified many times, thanks to the abundant info available), few class A/B designs, GC, Leach amp, McIntosh C22 preamp, active crossovers, CD players mods, lots of speakers etc. After all, I was in pursuit of the music – I consider myself music lover with electronics background and not a typical audiophile. My current system is 2-channel: Naim CD5i, Naim clone (via B1 Nelson Pass buffer, many thanks Mr. Pass) and Triangle Altea TW speakers.
Having said all that, I think people who don't know (or not interested in) electronics pay huge money for audio and are often mislead by the marketing tricks of some companies. I don't put Naim in the list because their gear, being of course quite expensive, is at least well engineered and built to last and work at its best...
To keep the post shorter I go directly to the point:
Reading the forum I see a lot of people somehow struggling with this amp, which is not that complex one but has few important steps to follow:
1.Verify that the kit you bought has the correct paths on the board. On my board I had few transistors with wrong (mirrored) prints, which can cause their wrong orientation. On the schematics going with the kit, you'll find a dot that connects one of the protection circuit bias resistors with the base of the top driver transistor and shouldn't be there. However, on the board the wiring is correct.
2.Find all info in the web about this amp. Basically, all schematics are same – some of them have protection and some of them don't. Some use the old Zetex and Naim “proprietary” transistors and some don't. I, personally, like Acoustica.org site. The mods on Naim Nait are suitable for this amp too. I've made some of them and the sound, if not better, is not worst. There is also a lot of information about best parts, substitutions etc. Some of them worth it, some of them don't. Decide for yourself, knowing your budget.
3.Parts: in my kit I replaced all small transistors with 2SC2240, being low noise and audio grade (just to be on the safe side). I matched their hfe and also replaced the drivers with faster – MJE 243 and 253. As far as I understand this schematics, the speed is not that much of factor here, but I still don't like the cheap and dirty TIP 41, 42
I replaced most of the capacitors – the input bypass one with a non-electrolytic (who knows how your DC component will be polarized), the input 330pF (or 220pF) with styroflex and the two in the drivers' base with styroflex type too, because I think the two drivers' base precision and total similarity is critical for the sound of this amp.
4.Everything else is from the kit – I have some SILMICs and some decent resistors, power transistors are fine, fake or not as some people claim, they are working fine in my design.
5.Then solder everything, BUT the power transistors.
6.Build your power supply. ±40Vdc is best. Mine is ±31Vdc from 225VA toroid and is working. About the “original” power supply with double rectifiers, one on each secondary, find an article in TNT. Google something like “TNT power supply” and you'll be right there. There are few pages of very interesting information. I can't add anything more. I have built a power supply like this, it works perfectly.
7.The grounding, the Naim magic!
Read the user manuals available on their site. They mention that the Naim equipment is grounded only in one point, usually the source, hence the CDP. Mine has huge thick AC cord. All other AC plugs of your equipment contain only safety grounds (only the enclosure chassis is grounded). The boards themselves are using “floating” grounding, which means they are at 0V potential, but not connected to ground until you put your whole system together and then everything is grounded via your source. To make this work use only isolated from the chassis RCA chinches (with plastic washers). You can keep everything “neutral” in your amp in one point – the reservoir capacitors common point.
Have in mind that in this kit the input “neutral” is not connected to the output (speaker “neutral”). You have to connect them or better, to pull wires from each of them to the capacitor's common. If not, you should expect some big DC voltage on your speaker output. I realized this from experience Fortunately, I have the habit to always measure the speakers' DC component before connecting the speakers.
My ground configuration is slightly different from the original Naim idea. I've put a 15 ohms resistors between the amp's input and output “neutrals”, and I grounded in the AC plug the capacitors' common. This way I have de facto properly grounded amp, but the inputs are not grounded there, because the 15 ohms resistors should force the inputs to be grounded through the source (my CD5i) less resistive path to ground, travelling into the interconnect cables shielding etc. etc.
I don't know which way is best, but I don't have any hum, and in the speakers I find only very quiet and uniform white noise.
The important thing is to avoid ground loops. For more info about ground loops I found very extensive paper on Hammond transformers site and, of course, on the FirstWatt site. Thank you, Mr. Pass!
8.Connect your power to the boards. Do it correctly!!!
9.Measure the voltage across the 100-ohm resistor between the base and emitter of the top output transistor (Remember, you still don't have the output transistors on the boards). Turn the trimmer and set this voltage to be small: 0.25V-0.30V, or less if possible, is fine. Now you can safely connect the power transistors and they won't have dangerous currents through them. They shouldn't have current at all in this situation.
10.Turn slowly the trimmer with the power transistors in place until you see about 5.5-7 mV across the output resistors (in the output transistors' emitters)
11.Leave the amp for some time and measure again. Mine doesn't drift and doesn't get hot. One more indication for people without precise measuring equipment that there is not ugly high-frequency oscillation.
12.I found by experimenting that the sound is by far better if you use 12-15 turns 0.75-1.0 mm output coil+10 ohms/3W resistor instead of the 0.22-ohm resistors. Some of the kit versions are with coils, some other are “original” with resistor. I thought the “original” should be better, but it's not. Try for yourself if you wish. For me the difference is night and day (or music - no music)! I wound the coil on a pen and didn't put the resistor inside, but kept them separated. I think we don't need to affect the resistor with some minor magnetic fields. It's so easy to avoid it, so, let's avoid it... put one on top of the other and not into.
Important: The amp is mildly oscillating with my DIY speaker cables. With QED Silver Anniversary it is perfect. So, it is true! The speaker cables must be high-ish inductance, low-ish capacitance, which is translated in relatively high characteristic impedance (75-100 ohms) Zchar. = SQRT (Lcable/Ccable)
This is it! Sorry for the huge post! I hope it could be useful for you, guys!
Finally, how does it sound? It sounds like music. Don't make me write cliché Its very expressive, open, deep, refined, tight, never dull or harsh sounding amp. This is not just good attempt, but sounds like real hi-end that I've listened to at some audio expos and in friend's systems.
How does it compares to other amps, including JLH? - I don't know. It's better than everything I've had (see the list on top). I have a friend of mine who is building right now the Naim clone and F4, simultaneously. He also have JLH and GC. Eventually, he'd give some opinion. I believe my Naim clone is at least same like JLH (which is to me a reference point) but with more power and no heat I'm mostly listening to jazz and some acoustic stuff. My friend says, he can never use JLH as a party amp. I don't know about the Naim, I don't listen to high sound levels, but the clarity at lower levels is very impressive.
If some questions arise, I'll try to answer, but I don't have a lot of free time. Please, don't ask me about theory of operation or basic electronic stuff. Buy some books instead! I can answer only practical questions related to my experience in audio.
Regards
I read the forum very often but I've never had the impulse to post a message ... anyway ... I think it's time to contribute a little bit to the DIY community.
I have built and perfectly working Naim NAP 140 based amp (as sold on eBay).
Having a Naim CD5i (the cheapest one
I often have the impulse to go and buy a Naim amp and to finish my quests, but right now I'm ready to put my clone next to any NAP 150/180, and I'm sure I won't be ashamed.This is the best amp that I've had in my system and I am in this hobby for more than 15 years. I started with amps created by a friend of mine, electronics engineer – some solid state class A and tube preamps. Later, when Internet has grown and was populated with useful and useless stuff, I started building and upgrading existing schematics ... all that most of you already know: Quad 405 (heavy modified many times, thanks to the abundant info available), few class A/B designs, GC, Leach amp, McIntosh C22 preamp, active crossovers, CD players mods, lots of speakers etc. After all, I was in pursuit of the music – I consider myself music lover with electronics background and not a typical audiophile. My current system is 2-channel: Naim CD5i, Naim clone (via B1 Nelson Pass buffer, many thanks Mr. Pass) and Triangle Altea TW speakers.
Having said all that, I think people who don't know (or not interested in) electronics pay huge money for audio and are often mislead by the marketing tricks of some companies. I don't put Naim in the list because their gear, being of course quite expensive, is at least well engineered and built to last and work at its best...
To keep the post shorter I go directly to the point:
Reading the forum I see a lot of people somehow struggling with this amp, which is not that complex one but has few important steps to follow:
1.Verify that the kit you bought has the correct paths on the board. On my board I had few transistors with wrong (mirrored) prints, which can cause their wrong orientation. On the schematics going with the kit, you'll find a dot that connects one of the protection circuit bias resistors with the base of the top driver transistor and shouldn't be there. However, on the board the wiring is correct.
2.Find all info in the web about this amp. Basically, all schematics are same – some of them have protection and some of them don't. Some use the old Zetex and Naim “proprietary” transistors and some don't. I, personally, like Acoustica.org site. The mods on Naim Nait are suitable for this amp too. I've made some of them and the sound, if not better, is not worst. There is also a lot of information about best parts, substitutions etc. Some of them worth it, some of them don't. Decide for yourself, knowing your budget.
3.Parts: in my kit I replaced all small transistors with 2SC2240, being low noise and audio grade (just to be on the safe side). I matched their hfe and also replaced the drivers with faster – MJE 243 and 253. As far as I understand this schematics, the speed is not that much of factor here, but I still don't like the cheap and dirty TIP 41, 42
I replaced most of the capacitors – the input bypass one with a non-electrolytic (who knows how your DC component will be polarized), the input 330pF (or 220pF) with styroflex and the two in the drivers' base with styroflex type too, because I think the two drivers' base precision and total similarity is critical for the sound of this amp.
4.Everything else is from the kit – I have some SILMICs and some decent resistors, power transistors are fine, fake or not as some people claim, they are working fine in my design.
5.Then solder everything, BUT the power transistors.
6.Build your power supply. ±40Vdc is best. Mine is ±31Vdc from 225VA toroid and is working. About the “original” power supply with double rectifiers, one on each secondary, find an article in TNT. Google something like “TNT power supply” and you'll be right there. There are few pages of very interesting information. I can't add anything more. I have built a power supply like this, it works perfectly.
7.The grounding, the Naim magic!
Read the user manuals available on their site. They mention that the Naim equipment is grounded only in one point, usually the source, hence the CDP. Mine has huge thick AC cord. All other AC plugs of your equipment contain only safety grounds (only the enclosure chassis is grounded). The boards themselves are using “floating” grounding, which means they are at 0V potential, but not connected to ground until you put your whole system together and then everything is grounded via your source. To make this work use only isolated from the chassis RCA chinches (with plastic washers). You can keep everything “neutral” in your amp in one point – the reservoir capacitors common point.
Have in mind that in this kit the input “neutral” is not connected to the output (speaker “neutral”). You have to connect them or better, to pull wires from each of them to the capacitor's common. If not, you should expect some big DC voltage on your speaker output. I realized this from experience
Fortunately, I have the habit to always measure the speakers' DC component before connecting the speakers.My ground configuration is slightly different from the original Naim idea. I've put a 15 ohms resistors between the amp's input and output “neutrals”, and I grounded in the AC plug the capacitors' common. This way I have de facto properly grounded amp, but the inputs are not grounded there, because the 15 ohms resistors should force the inputs to be grounded through the source (my CD5i) less resistive path to ground, travelling into the interconnect cables shielding etc. etc.
I don't know which way is best, but I don't have any hum, and in the speakers I find only very quiet and uniform white noise.
The important thing is to avoid ground loops. For more info about ground loops I found very extensive paper on Hammond transformers site and, of course, on the FirstWatt site. Thank you, Mr. Pass!
8.Connect your power to the boards. Do it correctly!!!
9.Measure the voltage across the 100-ohm resistor between the base and emitter of the top output transistor (Remember, you still don't have the output transistors on the boards). Turn the trimmer and set this voltage to be small: 0.25V-0.30V, or less if possible, is fine. Now you can safely connect the power transistors and they won't have dangerous currents through them. They shouldn't have current at all in this situation.
10.Turn slowly the trimmer with the power transistors in place until you see about 5.5-7 mV across the output resistors (in the output transistors' emitters)
11.Leave the amp for some time and measure again. Mine doesn't drift and doesn't get hot. One more indication for people without precise measuring equipment that there is not ugly high-frequency oscillation.
12.I found by experimenting that the sound is by far better if you use 12-15 turns 0.75-1.0 mm output coil+10 ohms/3W resistor instead of the 0.22-ohm resistors. Some of the kit versions are with coils, some other are “original” with resistor. I thought the “original” should be better, but it's not. Try for yourself if you wish. For me the difference is night and day (or music - no music)! I wound the coil on a pen and didn't put the resistor inside, but kept them separated. I think we don't need to affect the resistor with some minor magnetic fields. It's so easy to avoid it, so, let's avoid it... put one on top of the other and not into.
Important: The amp is mildly oscillating with my DIY speaker cables. With QED Silver Anniversary it is perfect. So, it is true! The speaker cables must be high-ish inductance, low-ish capacitance, which is translated in relatively high characteristic impedance (75-100 ohms) Zchar. = SQRT (Lcable/Ccable)
This is it! Sorry for the huge post! I hope it could be useful for you, guys!
Finally, how does it sound? It sounds like music. Don't make me write cliché
Its very expressive, open, deep, refined, tight, never dull or harsh sounding amp. This is not just good attempt, but sounds like real hi-end that I've listened to at some audio expos and in friend's systems. How does it compares to other amps, including JLH? - I don't know. It's better than everything I've had (see the list on top). I have a friend of mine who is building right now the Naim clone and F4, simultaneously. He also have JLH and GC. Eventually, he'd give some opinion. I believe my Naim clone is at least same like JLH (which is to me a reference point) but with more power and no heat
I'm mostly listening to jazz and some acoustic stuff. My friend says, he can never use JLH as a party amp. I don't know about the Naim, I don't listen to high sound levels, but the clarity at lower levels is very impressive.If some questions arise, I'll try to answer, but I don't have a lot of free time. Please, don't ask me about theory of operation or basic electronic stuff. Buy some books instead!
I can answer only practical questions related to my experience in audio. Regards
Thanks.
First can i thank all and one member who helped me get my amps to work.
I asked a member if he would look at my kits i had built and he said he would,he did check the kits and i had fitted a couple of resistors the wrong way,he changed them for me and sent the kits back to me.
I checked the kits on bench/computer desk wires all over the place switched on and music from one amp,so the other amp was checked and that to worked, i was well pleased that they worked.
Now they need a case look around the room and i find a couple of amps i built from a maplin kit it they had transformers and switch fitted,so of i went removing parts and fitting the new naim amp boards wanting to make them look good i took my time.
I had to make a newish power supply as there was no bridge rectifier fitted, as i had some new ones they were fitted along with 6x 4700uf caps,i sorted the wiring so it looked good must have spent a good few hours sorting it out.
It looked good so now the big switch on ,knowing that they did work i plugged the amp in and switched on,i got a bang flash and smoke before i could un plug it,and then the wife walks in asking what the smell was and were is this smoke from!!!!!!!!!!
After about 7 hours work and just seen it go up in smoke i was well pixxxd off.
So to day i checked the other amp with a supply that i have on the bench and that worked ok, i check the up in smoke amp and some of the tracking has gone, i still need to find out why it went up in smoke.
I check the transformer in the case that works so i fit the bridge rectifier, now the trans starts to click/hum so off comes the rectifier,now i did buy 4 off ebay a few months back and i have checked all of them and i get the noise if i fit them,i also tried a old one i had and that worked ok,no noise or hum and 34v.
Anyone had any problems with bridge rectifiers?
While i was on the net i also priced some parts to fix the dead amp,its cheaper to buy a new kit off xbay.
So i have bought some new amps, and i will get them to work it has become a challenge now to get two working amps.
And when the amp went bang it took a cheap speaker with it,so i now need a cheap pair of speakers to do my tests,before they near my main system.
So a big thanks to all who help me with this saga.
Stay tuned for the next big switch on,should be good for a laugh.
First can i thank all and one member who helped me get my amps to work.
I asked a member if he would look at my kits i had built and he said he would,he did check the kits and i had fitted a couple of resistors the wrong way,he changed them for me and sent the kits back to me.
I checked the kits on bench/computer desk wires all over the place switched on and music from one amp,so the other amp was checked and that to worked, i was well pleased that they worked.
Now they need a case look around the room and i find a couple of amps i built from a maplin kit it they had transformers and switch fitted,so of i went removing parts and fitting the new naim amp boards wanting to make them look good i took my time.
I had to make a newish power supply as there was no bridge rectifier fitted, as i had some new ones they were fitted along with 6x 4700uf caps,i sorted the wiring so it looked good must have spent a good few hours sorting it out.
It looked good so now the big switch on ,knowing that they did work i plugged the amp in and switched on,i got a bang flash and smoke before i could un plug it,and then the wife walks in asking what the smell was and were is this smoke from!!!!!!!!!!
After about 7 hours work and just seen it go up in smoke i was well pixxxd off.
So to day i checked the other amp with a supply that i have on the bench and that worked ok, i check the up in smoke amp and some of the tracking has gone, i still need to find out why it went up in smoke.
I check the transformer in the case that works so i fit the bridge rectifier, now the trans starts to click/hum so off comes the rectifier,now i did buy 4 off ebay a few months back and i have checked all of them and i get the noise if i fit them,i also tried a old one i had and that worked ok,no noise or hum and 34v.
Anyone had any problems with bridge rectifiers?
While i was on the net i also priced some parts to fix the dead amp,its cheaper to buy a new kit off xbay.
So i have bought some new amps, and i will get them to work it has become a challenge now to get two working amps.
And when the amp went bang it took a cheap speaker with it,so i now need a cheap pair of speakers to do my tests,before they near my main system.
So a big thanks to all who help me with this saga.
Stay tuned for the next big switch on,should be good for a laugh.
hi geo
looking at your pix of the boards i notice you have 10uf 50v input cap in the wrong place, you have it across the c2240 which should have the 47uf 50v which you have currently on the input.
duncan
looking at your pix of the boards i notice you have 10uf 50v input cap in the wrong place, you have it across the c2240 which should have the 47uf 50v which you have currently on the input.
duncan
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Thanks to everybody for an interesting thread, with many good tips, and I used the Ruwe recommendation to check some of the voltages before adding the output transistors.
I am only an amateur, so please do not take everything I say below too seriously, but here are some thoughts anyway:
I have tried out the "new and improved version" for about a week, and there was no problem building it and make it work convincingly. It is silent, does not get hot, sounds perfect with mono, and also measures OK with 15mV offset on both speaker terminals (30mV with open input). Also both amps showed minus 75mV on the capacitor side of R8 in the feedback loop, more about this later.
Sound (very subjective):
I was impressed with the powerful and tight bass right from the start, but also found that it could sound a tad lean and laconic higher up through the tonal register. When listening to symphonic music, the woodwinds, in particular "the oboes", were pulled forward somewhat (idiotic, since it is where the ear is most sensitive anyway), and all the violinists seemed to have changed to steel-core strings. Not very pleasant on some classical recordings when the recordings already were unbalanced with too prominent treble and upper mids, but otherwise the amp has been sort of entertaining too listen to, with the right musical input. It is a kind of straight forward, unsentimental and masculine sound, perhaps one could say, if ultimately somewhat unsmiling, and lacking in "magic" in comparison with, say, my old Electrocompaniet, which also has better overall dynamic capabilities, I must say. The NAP hits you like a boxing glove, while the old 25watt Electro makes the whole world explode in front of you (in an enjoyable way). A Linn Klout I accidentally have standing about is as powerful and controlled in the bass as the NAP, but still manages to sound more "tuneful" in the low-ish department).
In any case, an interesting amplifier, highly recommended just because of that, and for the great value of the kit.
(Personally I suspect that one comes rather close to the sound of the original NAP with this. After all one must remember that the company did not always use stick to using the same components over the years. The components delivered with the kit seems to be of high quality, and the print outlay very close to that of the classical old NAPS).
Some things I wanted to try out, perhaps to make the treble more palatable:
First of all I had no need for the input capacitor, since my pre-amp has some big blocking polyprops in the output. (a finished version of the DIY NAP ought to have two inputs IMO, one of them bypassing this cap).
Then I found out I did not need the protection circuit either, and removed all the belonging components. I also soldered the loudspeaker leads to the board instead of some dubious connectors I had used at first.
Already it sounded cleaner and better! Most of the violinists had at this point changed back to nylon core!
The two old Nad3020i amps I have been using as temporary power sources may seem not too impressive, but at least they have been beefed up with 3 x the original capacitor battery, which actually means 6 x in comparison to the Nad used as a single stereo amp. It should be enough to get a fair impression of the kit amp's basic abilities, I think.
After a while I started to wonder about the capacitor C5 in the feedback loop. There is some confusion about the polarity here, when one looks at the different schematics and prints published on the net.. Sometimes minus is shown on the GND side of the cap. But the measurements (see above) and some photos of the original, for example this:
http://www.roadum.org.uk/MISC Pics/feedback cap nap160.jpg
and also the schematics on Mr.Neil McBride's pages, suggest that plus goes to GND. The supplier of the kit is very considerate and supplies bipolar electrolytics for this use, and for C1 also, but perhaps there could be made arguments for using tantalum caps instead, like Naim does. Maybe the added un-linearity of these particular chemical caps will be sonically benevolent in this circuit, who knows. After all, famous people in the audio business are adding stuff like tube buffers and transformers to their constructions for approximately the same reason. Naim's specialty was to haul the sonic optimum from these old-fashioned circuits, and they earned their money with this approach for many years, and probably had real reason for using tantalum.
Also there may possibly be arguments against using bipolar electrolytic capacitors. The location in the feedback loop is sensitive, and as far as I know this kind of capacitors consists of two capacitors in series (squeezed together in one can), and who would use two electrolytic capacitors in series? I mean, with added inductance and possibly other nasty things? 68uF film types could be natural thing to try here, if one believes in reducing un-linearities at any cost.
BTW, Mr.Neil McBride, who seems to be very knowing and competent in the area, is bypassing his polarized C5 cap with a small film-based type, and that is what I am trying out right now (with plus to GND). I had 100uF 25V at hand, plus 0,1uF polyprop. Initial impression: Probably better!
Cheers, and pleas forgive if my English is not correct in places (or everywhere).
I am only an amateur, so please do not take everything I say below too seriously, but here are some thoughts anyway:
I have tried out the "new and improved version" for about a week, and there was no problem building it and make it work convincingly. It is silent, does not get hot, sounds perfect with mono, and also measures OK with 15mV offset on both speaker terminals (30mV with open input). Also both amps showed minus 75mV on the capacitor side of R8 in the feedback loop, more about this later.
Sound (very subjective):
I was impressed with the powerful and tight bass right from the start, but also found that it could sound a tad lean and laconic higher up through the tonal register. When listening to symphonic music, the woodwinds, in particular "the oboes", were pulled forward somewhat (idiotic, since it is where the ear is most sensitive anyway), and all the violinists seemed to have changed to steel-core strings. Not very pleasant on some classical recordings when the recordings already were unbalanced with too prominent treble and upper mids, but otherwise the amp has been sort of entertaining too listen to, with the right musical input. It is a kind of straight forward, unsentimental and masculine sound, perhaps one could say, if ultimately somewhat unsmiling, and lacking in "magic" in comparison with, say, my old Electrocompaniet, which also has better overall dynamic capabilities, I must say. The NAP hits you like a boxing glove, while the old 25watt Electro makes the whole world explode in front of you (in an enjoyable way). A Linn Klout I accidentally have standing about is as powerful and controlled in the bass as the NAP, but still manages to sound more "tuneful" in the low-ish department).
In any case, an interesting amplifier, highly recommended just because of that, and for the great value of the kit.
(Personally I suspect that one comes rather close to the sound of the original NAP with this. After all one must remember that the company did not always use stick to using the same components over the years. The components delivered with the kit seems to be of high quality, and the print outlay very close to that of the classical old NAPS).
Some things I wanted to try out, perhaps to make the treble more palatable:
First of all I had no need for the input capacitor, since my pre-amp has some big blocking polyprops in the output. (a finished version of the DIY NAP ought to have two inputs IMO, one of them bypassing this cap).
Then I found out I did not need the protection circuit either, and removed all the belonging components. I also soldered the loudspeaker leads to the board instead of some dubious connectors I had used at first.
Already it sounded cleaner and better! Most of the violinists had at this point changed back to nylon core!
The two old Nad3020i amps I have been using as temporary power sources may seem not too impressive, but at least they have been beefed up with 3 x the original capacitor battery, which actually means 6 x in comparison to the Nad used as a single stereo amp. It should be enough to get a fair impression of the kit amp's basic abilities, I think.
After a while I started to wonder about the capacitor C5 in the feedback loop. There is some confusion about the polarity here, when one looks at the different schematics and prints published on the net.. Sometimes minus is shown on the GND side of the cap. But the measurements (see above) and some photos of the original, for example this:
http://www.roadum.org.uk/MISC Pics/feedback cap nap160.jpg
and also the schematics on Mr.Neil McBride's pages, suggest that plus goes to GND. The supplier of the kit is very considerate and supplies bipolar electrolytics for this use, and for C1 also, but perhaps there could be made arguments for using tantalum caps instead, like Naim does. Maybe the added un-linearity of these particular chemical caps will be sonically benevolent in this circuit, who knows. After all, famous people in the audio business are adding stuff like tube buffers and transformers to their constructions for approximately the same reason. Naim's specialty was to haul the sonic optimum from these old-fashioned circuits, and they earned their money with this approach for many years, and probably had real reason for using tantalum.
Also there may possibly be arguments against using bipolar electrolytic capacitors. The location in the feedback loop is sensitive, and as far as I know this kind of capacitors consists of two capacitors in series (squeezed together in one can), and who would use two electrolytic capacitors in series? I mean, with added inductance and possibly other nasty things? 68uF film types could be natural thing to try here, if one believes in reducing un-linearities at any cost.
BTW, Mr.Neil McBride, who seems to be very knowing and competent in the area, is bypassing his polarized C5 cap with a small film-based type, and that is what I am trying out right now (with plus to GND). I had 100uF 25V at hand, plus 0,1uF polyprop. Initial impression: Probably better!
Cheers, and pleas forgive if my English is not correct in places (or everywhere).
An externally hosted image should be here but it was not working when we last tested it.
Hi Sletol,
Very interesting observations! Thanks for spending your time to post them. They confirmed some of my thoughts and measurements.
I also have ~75 mV (negative) across the feedback capacitor. In my kit I use 22uF+22uF+1uF tantalum at that place and I returned the whole package in the opposite direction. I suppose it's better to have the capacitors correctly DC polarized, especially the tantalums, which are much more sensitive to reverse voltages than electrolytics. Probably, the best is to put there some small voltage ceramic capacitors. I found really small 10uF that can fit in a stack of 5, for example. I'm sure they are not the best quality out there, but for sure are better than any elecrolytic.
My output DC component is ~50mV without load and ~30mV with speakers attached.
You know, about the protection circuit, my idea is the same - to remove it one day, but first I have to make a real protection circuit with a relay and very high input-impedance OP Amp watching for anomalies at the output. Without any protection, if for some reason or accidentally, the output of the amp shorts, it will take the output transistors (or one of them), which will kill immediately the speaker/s. Mine are not extremely expensive, but I like them very much
Recently, I did some changes in the protection with hope that I would remove its influence over the audio. IMO, the protection circuit as is, doesn't look totally harmless for the sound of this amp. For some reason Naim have biased the base-emitter voltage of the protection transitors to 0.25-0.30V, which is enough to put them in very small conduction (according to the curves that I've seen in the books). On the other side of the resistors R37, R38 is the DC voltage created by the output transistors current – or about 6 mV. This means, that current will flow all the time from the protection transistors base bias circuit through the output. I am not sure that it would affect the sound but I don't like it!
So, I changed the resistors R32 and R33 to 10k and now the protection transistors have base voltage of about 0.1V, which reduces twice the current that they add at the output and keeps them deeper in non-conductive state. Probably, now the protection circuit will react few msec slower to abuse, because the capacitors C3,C4 won't be pre-charged that much, but I am ready to compromise here.
Your initial observations of the sound of this amp are exactly the same as the observations of my friend who already built it. He says the same – dynamic, deep, powerful, very impressive on some records, but also aggressive and harsh. He says that the amp cause a listener fatigue after a while. I think, you both have some sort of oscillations or the last kits doesn't supply good quality output transistors (mine are 2SC2837). My amp doesn't sound harsh and aggressive at all, it sounds right with my speakers, which are considered bright, according to the most reviews. I have Triangle Altea SW, and I don't agree that they are bright even with this amp, or may be I am already deaf! Anyway, if your amp is still aggressive, turns a string orchestra into rock band, or causes headache, you should try to look for oscillations and distortions on oscilloscope. BTW, you have one very serious amp, it's tough to beat an Electrocompaniet.
Regards and thanks again for the input.
Very interesting observations! Thanks for spending your time to post them. They confirmed some of my thoughts and measurements.
I also have ~75 mV (negative) across the feedback capacitor. In my kit I use 22uF+22uF+1uF tantalum at that place and I returned the whole package in the opposite direction. I suppose it's better to have the capacitors correctly DC polarized, especially the tantalums, which are much more sensitive to reverse voltages than electrolytics. Probably, the best is to put there some small voltage ceramic capacitors. I found really small 10uF that can fit in a stack of 5, for example. I'm sure they are not the best quality out there, but for sure are better than any elecrolytic.
My output DC component is ~50mV without load and ~30mV with speakers attached.
You know, about the protection circuit, my idea is the same - to remove it one day, but first I have to make a real protection circuit with a relay and very high input-impedance OP Amp watching for anomalies at the output. Without any protection, if for some reason or accidentally, the output of the amp shorts, it will take the output transistors (or one of them), which will kill immediately the speaker/s. Mine are not extremely expensive, but I like them very much
Recently, I did some changes in the protection with hope that I would remove its influence over the audio. IMO, the protection circuit as is, doesn't look totally harmless for the sound of this amp. For some reason Naim have biased the base-emitter voltage of the protection transitors to 0.25-0.30V, which is enough to put them in very small conduction (according to the curves that I've seen in the books). On the other side of the resistors R37, R38 is the DC voltage created by the output transistors current – or about 6 mV. This means, that current will flow all the time from the protection transistors base bias circuit through the output. I am not sure that it would affect the sound but I don't like it!
So, I changed the resistors R32 and R33 to 10k and now the protection transistors have base voltage of about 0.1V, which reduces twice the current that they add at the output and keeps them deeper in non-conductive state. Probably, now the protection circuit will react few msec slower to abuse, because the capacitors C3,C4 won't be pre-charged that much, but I am ready to compromise here.
Your initial observations of the sound of this amp are exactly the same as the observations of my friend who already built it. He says the same – dynamic, deep, powerful, very impressive on some records, but also aggressive and harsh. He says that the amp cause a listener fatigue after a while. I think, you both have some sort of oscillations or the last kits doesn't supply good quality output transistors (mine are 2SC2837). My amp doesn't sound harsh and aggressive at all, it sounds right with my speakers, which are considered bright, according to the most reviews. I have Triangle Altea SW, and I don't agree that they are bright even with this amp, or may be I am already deaf! Anyway, if your amp is still aggressive, turns a string orchestra into rock band
, or causes headache, you should try to look for oscillations and distortions on oscilloscope. BTW, you have one very serious amp, it's tough to beat an Electrocompaniet.Regards and thanks again for the input.
HI
I have the same kit but I modified it from the beginning .
I do not use the originally supplied transistors because many fake Toshiba transistors .Probably those are to since Toshiba stop-ed manufacturing them many years back .
Also I changed the driver transistors MJE1XXX.
You can read about more at ESP site .
For the input cap I use Black Gate 10uF bipolar , believed or not one of the best capacitor you can buy .
But I did more modification ,I use Wima capacitors etc.
Now I'm planing to build another amp with out the protection , I want to try different power transistors like Sanken 2SC3858 , from ON semi some another type .
Better (Dale) resistors , the best pars I can buy .
For the power supply I use 600VA 2x35VAC transformer , 4pc Mepco 50 000uF capacitors .
Pure silver wire .
I can say even with the cheap metal film resistors and with the protection is in I'm still more than happy with the amp .
It sound incredible , never heard deep bass, nice mids and also nice highs .
These is the best amplifier for that money you can get !! And you can do a lot of improvement .
I believe these amp can sound if carefully ans ambled and used the best available parts you can com pare the sound of the amp with 00000 value amp.
I'm sure about that .That is why I will build the next amp with the totally upgraded components ..
Of course using the Mc B layout , no protection etc .
Greetings
I have the same kit but I modified it from the beginning .
I do not use the originally supplied transistors because many fake Toshiba transistors .Probably those are to since Toshiba stop-ed manufacturing them many years back .
Also I changed the driver transistors MJE1XXX.
You can read about more at ESP site .
For the input cap I use Black Gate 10uF bipolar , believed or not one of the best capacitor you can buy .
But I did more modification ,I use Wima capacitors etc.
Now I'm planing to build another amp with out the protection , I want to try different power transistors like Sanken 2SC3858 , from ON semi some another type .
Better (Dale) resistors , the best pars I can buy .
For the power supply I use 600VA 2x35VAC transformer , 4pc Mepco 50 000uF capacitors .
Pure silver wire .
I can say even with the cheap metal film resistors and with the protection is in I'm still more than happy with the amp .
It sound incredible , never heard deep bass, nice mids and also nice highs .
These is the best amplifier for that money you can get !! And you can do a lot of improvement .
I believe these amp can sound if carefully ans ambled and used the best available parts you can com pare the sound of the amp with 00000 value amp.
I'm sure about that .That is why I will build the next amp with the totally upgraded components ..
Of course using the Mc B layout , no protection etc .
Greetings
HI
I sold the Krell KSA-50 stuffed boards .
I did not finished totally because I read about some n........... comments on it .
Some one who built the Krell clone also built the Symasym amplifier .
I also had the Symasym amp and I was not satisfied with the sound gave me .
The guy who built both amp stated that the Symasym beat the Krell Clone .
So I decided the Krell must go after the Symasym amp .
Now I have the large heat sink and I'm on my way to build a Aleph2 amplifier instead the Krell Clone .
I only can say the Naim is much better sounding amplifier than the Symasym .
In every aspect is better !!!
I think for me it will be better choice the Aleph2 than the Krell Clone .
I only can write about what I heard .
Greetings
I sold the Krell KSA-50 stuffed boards .
I did not finished totally because I read about some n........... comments on it .
Some one who built the Krell clone also built the Symasym amplifier .
I also had the Symasym amp and I was not satisfied with the sound gave me .
The guy who built both amp stated that the Symasym beat the Krell Clone .
So I decided the Krell must go after the Symasym amp .
Now I have the large heat sink and I'm on my way to build a Aleph2 amplifier instead the Krell Clone .
I only can say the Naim is much better sounding amplifier than the Symasym .
In every aspect is better !!!
I think for me it will be better choice the Aleph2 than the Krell Clone .
I only can write about what I heard .
Greetings
Very good.
I also have the Aleph 60 board stuffed already, have the chassis, tranny, caps, but the project on hold until I have a bit more time...
My speakers are a very tough load - Magnepan MGIIIa, so the Krell works fine for me.
I always wanted to build a symasym..and an AKSA.
One day I will have all of them and be able to compare all at once.
I think you may have liked the Krell had you finished and listened to it...
I definitely like the Class A a bit more than anything I have listed to yet...but speakers and surroundings make a big difference...
I also have the Aleph 60 board stuffed already, have the chassis, tranny, caps, but the project on hold until I have a bit more time...
My speakers are a very tough load - Magnepan MGIIIa, so the Krell works fine for me.
I always wanted to build a symasym..and an AKSA.
One day I will have all of them and be able to compare all at once.
I think you may have liked the Krell had you finished and listened to it...
I definitely like the Class A a bit more than anything I have listed to yet...but speakers and surroundings make a big difference...
Hi everyone, just some comments on this naim.
I have also built this amp soe years ago, not the kit but just a prototype. I think a lot of builders are having trouble with oscilations or the amps are very unstable. As one can see from schematics this amp does not have zobel or coil. I had problems with this too. When naim was selling these amps they recommended using speaker cables of their own design and make, these had some impedance and when this amp was used with their cables they tended not to have oscillating problems. Now most of the builders here are using their own cables which is gonna create problems.
I recommend that the builders here build these into their amps to minimize stability problems. When done properly it shouldnt have any impact on the sound if your amp was already working without oscillation problems, for the others a good improvement.
I have also built this amp soe years ago, not the kit but just a prototype. I think a lot of builders are having trouble with oscilations or the amps are very unstable. As one can see from schematics this amp does not have zobel or coil. I had problems with this too. When naim was selling these amps they recommended using speaker cables of their own design and make, these had some impedance and when this amp was used with their cables they tended not to have oscillating problems. Now most of the builders here are using their own cables which is gonna create problems.
I recommend that the builders here build these into their amps to minimize stability problems. When done properly it shouldnt have any impact on the sound if your amp was already working without oscillation problems, for the others a good improvement.
I build and use CAT 5 cable for speaker cables. I have a very expensive set of Martin Logan cables I don't use often since so cumbersome...
Since I have an LC meter, what capacitance/inductance/resistance should I strive for on this Naim clone?
Don't want an output zobel if I can design it out with cables...
Since I have an LC meter, what capacitance/inductance/resistance should I strive for on this Naim clone?
Don't want an output zobel if I can design it out with cables...
john65b said:
Good question!
I would not be surprised if you could reach the recommended values just by tearing apart the plus and minus leads of an ordinary, typical speaker cable...
Conserning the possible fake Toshiba transistors (pretty good sounding fakes I would say!), it would be interesting to see if this could be established as a fact. I mean, either are they fakes, or they are not, there ought to be few possibilities in between...
Could any of you who have some devices leftover after changing to other types possibly try to open them (squeeze in a vise or something) and post a picture of what is inside.
There are some examples in other threads, for example this:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=28868&perpage=25&pagenumber=1
HI
I did not have any experience the amp would oscillate . It cant be because I use Kimber 8TC speaker cable or the PC boards are well design .
About the fake transistor all I can say if I can use original ON semi or original large Sanken like 2SC3858 i do not want to have the possibility using fake transistors .Read the comments of Rode at the ESP site !
If it blows up will take the speaker with itself with out any protection .
The protection the kit designer idea . Originally it does not have the protection .
You can go to the Mc B site and read his advise about the power transistors or the driver transistors .
He did tested a few of them .
I will use the Mc Bridge layout for my next higher power vision Naim Clone .
Of course using the best pars available ! Now I'm waiting to get one more extra PC board from my fried so I can do all kind of tests on a spare board .After it settle done build the original .
Even I want to do comparison between the Ebay Naim Clone and the Mc B clone .
I have extra PC boards from the Ebay in case if that would sound better with out the protection of course I will populate that boards .
Right now I use the same on-semi like you can find at the Mc B site .
Greetings
I did not have any experience the amp would oscillate . It cant be because I use Kimber 8TC speaker cable or the PC boards are well design .
About the fake transistor all I can say if I can use original ON semi or original large Sanken like 2SC3858 i do not want to have the possibility using fake transistors .Read the comments of Rode at the ESP site !
If it blows up will take the speaker with itself with out any protection .
The protection the kit designer idea . Originally it does not have the protection .
You can go to the Mc B site and read his advise about the power transistors or the driver transistors .
He did tested a few of them .
I will use the Mc Bridge layout for my next higher power vision Naim Clone .
Of course using the best pars available ! Now I'm waiting to get one more extra PC board from my fried so I can do all kind of tests on a spare board .After it settle done build the original .
Even I want to do comparison between the Ebay Naim Clone and the Mc B clone .
I have extra PC boards from the Ebay in case if that would sound better with out the protection of course I will populate that boards .
Right now I use the same on-semi like you can find at the Mc B site .
Greetings
Hi,
About the cables.
I have large file with old Julian Vereker posts. This is what I found:
FOR INFO THE RELEVANT SPECS ARE:
1.3 - 1.5microHenries per metre (Loop)
MAX 20picoFarads per metre
About 25milliOhms per meter (loop)
MIN Length 3 metres
To me, it means that any cable with high characteristic impedance will work Zchar.=SQRT(L/C)>150-180 ohms
About the cables.
I have large file with old Julian Vereker posts. This is what I found:
FOR INFO THE RELEVANT SPECS ARE:
1.3 - 1.5microHenries per metre (Loop)
MAX 20picoFarads per metre
About 25milliOhms per meter (loop)
MIN Length 3 metres
To me, it means that any cable with high characteristic impedance will work Zchar.=SQRT(L/C)>150-180 ohms
Thank you for the information Ruwe! I must say that the narrow limits that he suggests are surprising (the narrowness makes
it sound more like a typical receipt how to achieve oscillation than how to avoid it). Or can it be that he was describing some spesific cable here, more than suggesting real, practical limits? (I would rather have expected him to use words as "highish" and "lowish" about these values, like you do yourself in your long message on the previous page).
In any case I will go for your solution sooner or later, using conventiohnal zobel. But did you say that you got a tendency
to oscillation, despite this (with your DIY cable)? Strange somewhat!
Not thinking about this peculiar cable aspect, I let the newly finished amp work very hard in a dummy load the whole night
through, to see if everything was OK. It was, despite the (possible) fake transistors, and despite using extremely short
pieces of cable, much less than one meter. No extreme heat or signs of trouble at all...
it sound more like a typical receipt how to achieve oscillation than how to avoid it). Or can it be that he was describing some spesific cable here, more than suggesting real, practical limits? (I would rather have expected him to use words as "highish" and "lowish" about these values, like you do yourself in your long message on the previous page).
In any case I will go for your solution sooner or later, using conventiohnal zobel. But did you say that you got a tendency
to oscillation, despite this (with your DIY cable)? Strange somewhat!
Not thinking about this peculiar cable aspect, I let the newly finished amp work very hard in a dummy load the whole night
through, to see if everything was OK. It was, despite the (possible) fake transistors, and despite using extremely short
pieces of cable, much less than one meter. No extreme heat or signs of trouble at all...