TDA7297
Maybe, I do not know, the Folsom's amp with the TDA7297. Audiocircle thread:
Folsom's great little 7297 Chip Amp
Here, in Diyaudio:
Folsom DIY7297 Amp & Antipole PSU
[PDF] https://www.st.com/resource/en/datasheet/tda7297.pdf
Maybe, I do not know, the Folsom's amp with the TDA7297. Audiocircle thread:
Folsom's great little 7297 Chip Amp
Here, in Diyaudio:
Folsom DIY7297 Amp & Antipole PSU
[PDF] https://www.st.com/resource/en/datasheet/tda7297.pdf
Now there are LJM L12-2 v3 and v4 (new logo is panda bear).
preamps and poweramps at along1986090 | eBay
LJM L12-2 at along1986090 | eBay
kit, ensambled with heatsink and full amp.
ljm l12-2 heatsink ljm l12-2 heatsink | eBay
The more important questions: which speakers, room, distance and type of music.
The others cheap chinese amplifiers are a lottery I think.
preamps and poweramps at along1986090 | eBay
LJM L12-2 at along1986090 | eBay
kit, ensambled with heatsink and full amp.
ljm l12-2 heatsink ljm l12-2 heatsink | eBay
The more important questions: which speakers, room, distance and type of music.
The others cheap chinese amplifiers are a lottery I think.
Last edited:
20 Vrms / 50 watts
Measurements by vzaichenko were at 20 Vrms / 50 watts.
L12-2 CFP Output amp 120W*2 8R
by vzaichenko
Measurements by vzaichenko were at 20 Vrms / 50 watts.
L12-2 CFP Output amp 120W*2 8R
by vzaichenko
Just got a pair of the boards from ebay - measured the key parameters yesterday.
Power rails were +/-36V DC.
All the spectrums are measured at 20VRMS, 8 ohm resistive load - equal to 50W output. I can't go higher due to my load resistor power, rather hot already
Square waves - 30V amplitude (p-p).
Very good performance, I must say
Last edited:
I guess the amplifier part is same on both, and same to datasheet application.
The black one has tons more bells and whistles, which may be very useful, but not straight changing audio quality.
As in: you buy 2 chipamps, one in a "white box" shop, where you order, pay, salesman gives you a little white cardboard box with it, and tells you to move along, there´s a long qyueue behind you.
Second shop charges 10 X as much, but is lavishly decorated, a HOT girl smiles at you, you wait sitting in a very comfortable sofa while being offered Abyssinian coffee, a little Cognac and a cigar while you wait for a white apron Engineer to deliver a detailed Nobel prize level speech on amp qualities, then a uniformed chauffeur drives you and the gorgeously packaged amplifier to your home, in a luxurious Limo .
Which one sounds best?
Back to the amps, I´m not so sure main rails are regulated, presumed regulators and heat sinks look too small for that.
Amp certainly looks designed to impress possible customers, that´s for sure.
ICs might do some Audio processing but I rather suspect they are Servos or part of some elaborate speaker/DC protection, not necessarily in the Audio signal path.
Of course, this whole thread is just an excercise on Imagination, since we lack any solid data
The black one has tons more bells and whistles, which may be very useful, but not straight changing audio quality.
As in: you buy 2 chipamps, one in a "white box" shop, where you order, pay, salesman gives you a little white cardboard box with it, and tells you to move along, there´s a long qyueue behind you.
Second shop charges 10 X as much, but is lavishly decorated, a HOT girl smiles at you, you wait sitting in a very comfortable sofa while being offered Abyssinian coffee, a little Cognac and a cigar while you wait for a white apron Engineer to deliver a detailed Nobel prize level speech on amp qualities, then a uniformed chauffeur drives you and the gorgeously packaged amplifier to your home, in a luxurious Limo .
Which one sounds best?
Back to the amps, I´m not so sure main rails are regulated, presumed regulators and heat sinks look too small for that.
Amp certainly looks designed to impress possible customers, that´s for sure.
ICs might do some Audio processing but I rather suspect they are Servos or part of some elaborate speaker/DC protection, not necessarily in the Audio signal path.
Of course, this whole thread is just an excercise on Imagination, since we lack any solid data
Right!
Many components, in a variety of colors..
When it comes to electronic modules, that come as a PCB with soldered components on it and no enclosure,
the looks of the whole thing does a part of the sales..
So it's very clear that they wanted to make it condensed and colorful.
And with big capacitors to impress.
That's why I asked about it
The other (green) board puts no emphasis on impressing.
I always hear "Servos" regarding amplifiers, but the Servos that I know are a kind of motor that can rotate slowly..
Obviously it's not the meaning here, so what does a Servo mean?
Well, these also "rotate slowly"
Point is, amplifier output does not exactly follow input signal, often by a large margin.
Negative feedback corrects a lot of that, by comparing output voltage to input voltage.
In a classic SS amplifier, Lin invented architecture is the most popular, you have basically 3 blocks: a large gain stage in the middle, a high current stage at the output to drive speakers, and an input stage which does many useful things, from providing an acceptable input impedance to accepting the negative feedback signal to keeping output DC offset near ground.
Sadly this cool little circuit has some offset by itself, for various reasons, typically a few mV.
IF amplifier has down to DC response (a few do), this small offset gets amplified to annoying levels, sometimes requiring an extra trimmer to zero it, or a carefully symmetrical design, with carefully matched components to minimize it.
IF amplifier does NOT have down to DC response, which is the most common option, by using a series capacitor in the NFB net connected to ground, (even LM3886 has it, just look at the datasheet schematic), then DC gain becomes 1X or 0dB ... offset is not amplified but it still appears at the output.
That´s why typical amplifiers have low mV offset, think 20 to 40 mV or so.
We can live with that, but if you want ZERO or, say, not more than 1 or 2 mV, you have 2 options:
a) add a DC offset adjust trimmer even to the "non DC" amplifier and adjust it as needed.
b) enter the mighty *Servo* ... which is nothing more or less than an Op Amp, which compares output DC to ground , generates an error signal, and injects it into the input stage to correct those 20 to 40mV into zero (or almost).
Call it an "amplified NFB" if you wish ... but now you are introducing a problem: ... "how much" is fed back for correction is critical, inevitable phase shifts along the path mean that if you do not do it carefully amp may become a self destroying power oscillator, or at least sound bad or do nasty things.
So servos are heavily limited in frequency response, they typically work only below 1 or 2 Hz ... which includes DC of course, what we need, but do NOT affect Audio frequencies at all.
This Boutique LM3886 kit must include them, I see no other reason, certainly chipamp by itself sounds very good.
There are a couple designs around in this very Forum including an Op Amp front end added to a "normal" amplifier for vanishing small distortion figures, not impossible but not likely they did that here, which takes **careful** compensation tweaking.
Point is, amplifier output does not exactly follow input signal, often by a large margin.
Negative feedback corrects a lot of that, by comparing output voltage to input voltage.
In a classic SS amplifier, Lin invented architecture is the most popular, you have basically 3 blocks: a large gain stage in the middle, a high current stage at the output to drive speakers, and an input stage which does many useful things, from providing an acceptable input impedance to accepting the negative feedback signal to keeping output DC offset near ground.
Sadly this cool little circuit has some offset by itself, for various reasons, typically a few mV.
IF amplifier has down to DC response (a few do), this small offset gets amplified to annoying levels, sometimes requiring an extra trimmer to zero it, or a carefully symmetrical design, with carefully matched components to minimize it.
IF amplifier does NOT have down to DC response, which is the most common option, by using a series capacitor in the NFB net connected to ground, (even LM3886 has it, just look at the datasheet schematic), then DC gain becomes 1X or 0dB ... offset is not amplified but it still appears at the output.
That´s why typical amplifiers have low mV offset, think 20 to 40 mV or so.
We can live with that, but if you want ZERO or, say, not more than 1 or 2 mV, you have 2 options:
a) add a DC offset adjust trimmer even to the "non DC" amplifier and adjust it as needed.
b) enter the mighty *Servo*
... which is nothing more or less than an Op Amp, which compares output DC to ground , generates an error signal, and injects it into the input stage to correct those 20 to 40mV into zero (or almost).Call it an "amplified NFB" if you wish ... but now you are introducing a problem:
... "how much" is fed back for correction is critical, inevitable phase shifts along the path mean that if you do not do it carefully amp may become a self destroying power oscillator, or at least sound bad or do nasty things.So servos are heavily limited in frequency response, they typically work only below 1 or 2 Hz ... which includes DC of course, what we need, but do NOT affect Audio frequencies at all.
This Boutique LM3886 kit must include them, I see no other reason, certainly chipamp by itself sounds very good.
There are a couple designs around in this very Forum including an Op Amp front end added to a "normal" amplifier for vanishing small distortion figures, not impossible but not likely they did that here, which takes **careful** compensation tweaking.
Thank you for a great explanation..
To which one are you referring?
When you say "chipamp by itself" you mean without adding 1 of the 2 options that you wrote for fixing the low mV offset?
The ACA kit that Bare mentioned a few posts back is definitely a wonderful sounding, and very well documented / supported design, and easy enough build - but certainly cut from entirely different cloth than a 3886 amp. I’ve built both the original ACA1.1 monoblocks as well as a pair of the recent stereo version - currently used in mono bridged mode - and have had a good exposure to 3 different Neurochrome designs. They do not sound the same, and I happen to prefer the Modulus variants; for those who can relate, I also happen to prefer the Alpair metal cones to the papers - specifically the Alpair10.3, A11MS or Pluvia11s vis a vis the A10P.
I'm not certain this was directed at me, but I'll try and answer.
- I believe that Tom's engineering expertise has probably resulted in one of the best designed "datasheet style" boards (the LM3886DR). Have a look at his CV here for some background. He has also made available a lot of the technical information and design criteria used to design a board that is as good as it can be for a datasheet design (see here).
- Tom's boards use good design to mean that there is little to be further gained by using boutique components. He does point out where you might make marginal gains though, if you fancy it.
- When you purchase from Tom, you get excellent documentation that includes everything you would need to get the boards constructed into a working amplifier. As well as the board building guide, this include a lot of background around the power supply, heatsink and how to correctly calculate power requirements based upon intended use and speaker impedance.
- The documentation from Tom also includes a link to a Mouser project which makes ordering all of the components a breeze. Although Mouser is based in the USA, they have offices worldwide and my order was shipped and arrived with me in the UK in around 4 days.
- Tom's products are backed by his support. On the few occasions I have had to contact Tom, he has responded quickly and has answered my questions perfectly, often giving me some extra background information as well.
I don't have experience with other LM3886 amplifiers, or even any other DIY amplifiers. I was - and still am - something of a novice at this. However, given my experience with the LM3886DR, I have ordered a set of the Modulus-86 boards for another project.
While I certainly accept that the Neurochrome boards appear quite expensive, I wasn't trying to build an amplifier on a shoestring and for me the extras I have detailed above make the Neurochrome offerings good value for money. And as I have said before, once you cost in the power supply components, chassis, other ancillary parts and perhaps even your time, the cost of the boards is far from the most expensive part. For a comparison, take a look at the Amp Camp kit here for a very rough expectation of what a whole amp might cost.
Do I think that everybody looking to build an LM3886 based amplifier should go the Neurochrome route? No of course not. Everybody has different requirements, available funds and levels of expertise. Ultimately, you've got to make your own mind up about the direction that is right for you.
Neil
Neil, thank you for your elaborated answer, but to be honest the request was for Chrisb.
I am happy that you are happy with the lm3886DR, which btw is a "vanilla" lm3886 amp designed by Tom. You should try other amps based on lm3886 to make sure you got the best for you .
From your comment I can see that you are a little bit too impressed by his background, the documentation is very good, I know that, the emails were very prompt, and his product is very good too. Tom is doing a great job!
Have fun building the mod86 and please let us know which one you like more.
I am happy that you are happy with the lm3886DR, which btw is a "vanilla" lm3886 amp designed by Tom. You should try other amps based on lm3886 to make sure you got the best for you .
From your comment I can see that you are a little bit too impressed by his background, the documentation is very good, I know that, the emails were very prompt, and his product is very good too. Tom is doing a great job!
Have fun building the mod86 and please let us know which one you like more.
asus - I’m guessing Tom wouldn’t object to “vanilla” as a description of his basic DR circuit - which I’ve heard as well. As a bit of background, “hi, my name is Chris, and I’ve been an audio addict since 1963”. I worked in the retail trade off and in for a good part of the 70s, during which time I made more than a few revolutions on the merry-go-round chasing after that brasss ring, and owned / heard lotsa interesting stuff, but didn’t dip my toes into the DIY kiddy pool until the late ‘90s.
Cutting to the chase as is relevant here, I’d owned/built at least half a dozen tube amps between approx ‘99 to as recently as about 5yrs ago - ranging from minimalist SE clones of Decware Zen, Bottlehead Paramour 2A3, DIY variant of Eli Duttman’s El Cheapo, stock/tweaked Jolida 302, custom 6CA7, plus probably a couple I’ve forgotten for the moment. There’s certainly a seductive allure to the combination of higher sensitivity “fullrange” speakers - such as Fostex, Alpairs etc, in back-loaded horns - and low power SE amps, whether tube or SS such as the ACA.
That said, my immediate impression on listening to the Modulus86 about 2 yrs ago- and more recently the 686 - was their transparency, “grip”, and particularly in the case of the 686, seemingly bottomless well of dynamics. Long an agnostic/ skeptic as to such exactly where the threshold for specmanship for its own sake might lie, the recent opportunity to directly compare several of Tom’s amps to the ACA, Hypex UCD180, and my two current mainstream products (Marantz CR611 and Onkyo TX NR818) on a range of speakers, some of which I thought I was intimately familiar, was revelatory. The speakers were Alpair10.3, A7.3, A10P/A7.3 MTM, the new A11MS, and Jordan Eikona2.
Cutting to the chase as is relevant here, I’d owned/built at least half a dozen tube amps between approx ‘99 to as recently as about 5yrs ago - ranging from minimalist SE clones of Decware Zen, Bottlehead Paramour 2A3, DIY variant of Eli Duttman’s El Cheapo, stock/tweaked Jolida 302, custom 6CA7, plus probably a couple I’ve forgotten for the moment. There’s certainly a seductive allure to the combination of higher sensitivity “fullrange” speakers - such as Fostex, Alpairs etc, in back-loaded horns - and low power SE amps, whether tube or SS such as the ACA.
That said, my immediate impression on listening to the Modulus86 about 2 yrs ago- and more recently the 686 - was their transparency, “grip”, and particularly in the case of the 686, seemingly bottomless well of dynamics. Long an agnostic/ skeptic as to such exactly where the threshold for specmanship for its own sake might lie, the recent opportunity to directly compare several of Tom’s amps to the ACA, Hypex UCD180, and my two current mainstream products (Marantz CR611 and Onkyo TX NR818) on a range of speakers, some of which I thought I was intimately familiar, was revelatory. The speakers were Alpair10.3, A7.3, A10P/A7.3 MTM, the new A11MS, and Jordan Eikona2.
Chris, I appreciate that you took your time and answer.
I built parallel86, modulus86, My_Ref Fremen Edition , my ref revC, lm4780 pa03, akitika gt101 and some other ones not worth mentioning here, all of these chip amps based on lm3886/4780. I cant say I liked one more than other one, for me all sounded kind of similar.
Maybe my ears are not so good as I was thinking they are, or maybe my audio sources are not that good.
I did not try the mod686, which I think is a different beast.
I built parallel86, modulus86, My_Ref Fremen Edition , my ref revC, lm4780 pa03, akitika gt101 and some other ones not worth mentioning here, all of these chip amps based on lm3886/4780. I cant say I liked one more than other one, for me all sounded kind of similar.
Maybe my ears are not so good as I was thinking they are, or maybe my audio sources are not that good.
I did not try the mod686, which I think is a different beast.
Try to do this easy test:
Can you tell original file from three times DA-AD converted?
I suspect that many people have not optimized the OS and soft players to play multimedia, I think that is the main cause and not the ears (I can not believe that there is so much deafness). Very few people optimize I think.
Other possibility is that you have a very bad electrical grid like me -> the BIG bottleneck. After the use of my gadgets to alleviate it I appreciate much more the differences in the recordings, much more. Well, I have tweaked all the componentes of my system to minimize any source of noise and distortion.
You eliminate a bottleneck and another one appears that was hidden and so on until there is none left. A lot of time and effort required!
You eliminate a bottleneck and another one appears that was hidden and so on until there is none left. A lot of time and effort required!
Yeah. The improvements in the front end of the Modulus-686 (and also -186 and -286) and the extra power reserve of the -286 and -686 make a big difference. I designed the Modulus-686 from the ground up. It's definitely a different animal, though its output stage shares some circuitry with the other Modulus amps. The Modulus-186 and -286 are improved derivatives of the Modulus-86.
Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.