Did you use the stock schematic or did you add anything?
I just got a set of pcbs for the MyRef version of LM3886 and I can see that a fully populated kit (ps included, but truly no transformer) runs for 200 euro with the premium BOM with all components/pcb. I dare to say that it's as much as yours only transformers missing. But that's a more powerful setup to be honest. Also you get two monoblocks for that price. But for that price you get nice parts and extra development for the circuit so it makes the most out of that chip (well, as feedback goes anyway).
I'm curious about the schematic if you could post it and explain the improvements. I'm intrigued to be honest.
I just got a set of pcbs for the MyRef version of LM3886 and I can see that a fully populated kit (ps included, but truly no transformer) runs for 200 euro with the premium BOM with all components/pcb. I dare to say that it's as much as yours only transformers missing. But that's a more powerful setup to be honest. Also you get two monoblocks for that price. But for that price you get nice parts and extra development for the circuit so it makes the most out of that chip (well, as feedback goes anyway).
I'm curious about the schematic if you could post it and explain the improvements. I'm intrigued to be honest.
Mine is 200 USA, that's nearly 400 euro. Mine includes shipping for a transformer, all the parts, and the CMC. For some perspective you'd be spending $9 for the CMC, $8 for the parts, $10 (maybe more) for the transformer, $4-8 for wire, and $8 for PCB's if it all came separate. That's at least $39 for shipping! My $200 includes priority USPS to the USA! (It's $12.95) That means you get it fast and in better shape. The only reason the price is even brought up is because the original was so cheap. You can't even save money by buying these parts individually once shipping is involved - yes that's how close I've been able to bring this deal to everyone if I can sell enough to buy the parts for a 100.
The basic schematic has things you can't not do, like the mute/stdby functions need power or you won't get sound. I've been through a lot of interesting discoveries like the ground for DC power in and the source signal have to be connected or the amp won't turn on under any circumstance! No one even knows about that... oh man I went nuts and soldered in a new chip to a PCB board prototype trying to figure that out!
The improvements are very much so involved in the design of the PCB with its star grounding, copper planes (pours) for very low inductance, and capacitor orientation. There are things going on I haven't seen with other kits. The Kmultiplier that provides the voltage regulation, improves PSRR to the DC capacitors, and lowers the AC crossing them, is only just starting to be appreciated! It does wonder for a multitude of things and offers even better filtration for CD players etc. Even more interesting it provides very low impedance, lower than most capacitor you can put on a chipamp besides the OSCON SEPF Panasonics I'm using. Implementing it meant greatly improved imaging, among other things.
The transformer really is a good one; there's a reason Peter Daniel's quotes the engineer for Mark Levinsons saying they only use dual bobbins for smaller devices. The PSU is unique, too. How many AC filtered post transformer PSU's have you seen? Dampened? Copper pours? Lay down rectifiers onto ground? Again, with special orientation for the capacitors. Less not forget the optional output taps for small devices! ChrisMMM used a version of this PSU on a TPA3116 and really enjoyed it. Others have too, but no one but him has reported it that I recall. (Mouser sold out of the recommend caps)
The basic schematic has things you can't not do, like the mute/stdby functions need power or you won't get sound. I've been through a lot of interesting discoveries like the ground for DC power in and the source signal have to be connected or the amp won't turn on under any circumstance! No one even knows about that... oh man I went nuts and soldered in a new chip to a PCB board prototype trying to figure that out!
The improvements are very much so involved in the design of the PCB with its star grounding, copper planes (pours) for very low inductance, and capacitor orientation. There are things going on I haven't seen with other kits. The Kmultiplier that provides the voltage regulation, improves PSRR to the DC capacitors, and lowers the AC crossing them, is only just starting to be appreciated! It does wonder for a multitude of things and offers even better filtration for CD players etc. Even more interesting it provides very low impedance, lower than most capacitor you can put on a chipamp besides the OSCON SEPF Panasonics I'm using. Implementing it meant greatly improved imaging, among other things.
The transformer really is a good one; there's a reason Peter Daniel's quotes the engineer for Mark Levinsons saying they only use dual bobbins for smaller devices. The PSU is unique, too. How many AC filtered post transformer PSU's have you seen? Dampened? Copper pours? Lay down rectifiers onto ground? Again, with special orientation for the capacitors. Less not forget the optional output taps for small devices! ChrisMMM used a version of this PSU on a TPA3116 and really enjoyed it. Others have too, but no one but him has reported it that I recall. (Mouser sold out of the recommend caps)
Last edited:
I was not talking about the removal of the components in the stock schematic, but to add some to make it better (if possible).
Here's some literature on what has been done to the LM3886 by Mauro:
http://www.diyaudio.com/forums/atta...73673-students-guide-myref-project_my_ref.pdf
I thought you improved the schematic design, but you say you only concentrated on the pcb layout instead.
And yes, 1€ = 1,11$. After looking again I noticed that the LM3886 populated pcbs will cost me 175€ (~195$).
Hope you don't mind the questions but I'm giving you a chance to explain and present your product
as well as the improvements you brought to it.
Did you make many iterations of the pcb until you arrived to the current one? How many versions have you gone through?
Also do you have a working final product that you could present some pictures of? And maybe a recording with a nice mic?
Here's some literature on what has been done to the LM3886 by Mauro:
http://www.diyaudio.com/forums/atta...73673-students-guide-myref-project_my_ref.pdf
I thought you improved the schematic design, but you say you only concentrated on the pcb layout instead.
And yes, 1€ = 1,11$. After looking again I noticed that the LM3886 populated pcbs will cost me 175€ (~195$).
Hope you don't mind the questions but I'm giving you a chance to explain and present your product
as well as the improvements you brought to it.Did you make many iterations of the pcb until you arrived to the current one? How many versions have you gone through?
Also do you have a working final product that you could present some pictures of? And maybe a recording with a nice mic?
Here's another thread on a nice layout based on the smaller LM1875/TDA2050 variants, made by Linuxguru:
http://www.diyaudio.com/forums/chip-amps/184165-miniref-schematic-pcb-layout.html
I attached his board prototype. The BOM for the board should be around 50$ (speaker protection relay included )
http://www.diyaudio.com/forums/chip-amps/184165-miniref-schematic-pcb-layout.html
I attached his board prototype. The BOM for the board should be around 50$ (speaker protection relay included
)Attachments
20x lunches! I'm actually surprised how well the price came out for the premium parts, and after reduces it two two PCB's. I've easily spent that with shipping for other chipamp kits and transformers that were not as impressive, no where nearly interesting enough to make me want to develop them in particular.
My friend who owns a Bryston 3BSST(2) and Piega 5.2 Premiums told me, and I quote, "if I didn't already have the Bryston, I would be begging you to build me one." But then he made a charitable contribution to get the demo going (expensive when order only a few PCB's, chassis, etc). I'll go back to his place and he'll want the final version of mine more than his Bryston
I can see how the difference in price is big, but if you look at the performance, shipped, I'd say it's a steal.
Why include a power supply?
My lunch money amp with it's "reject chip" never gets switched off and runs dead quiet on a 14v 4amp Muse switching ps.
Trileru, I have to point out that the basic schematic only has the most basic of parts on it. Essentially every amplifier ever made includes them. But I did remove things like the protection diode (that causes noise). What you're asking would be a great question if feedback or other built-into-the-chip were external.
The product will be the 4th generation PCB, 4th or more of the Kmultiplier version, at least 5 in parts, 3rd in PSU's, and 3rd in transformers. Surely I lost a few in there I didn't count
I had no idea the euro dropped like that!
The PSU and transformer greatly simplify the end users needed capability, multiple people asked for it, it allows me to show off the PSU so people can use it for other projects, and it ensures you get very good performance. Different voltages change the voltage regulation factor, and the saturation factors of the Kmultiplier.
You may not have audibly apparent noise Radiosmuck, that doesn't change the fact that noise is changing the sound of your music! This is one of the hardest things to explain to someone looking for or wanting to understand power conditioning. You hear what the noise does, every moment you listen, but you don't "hear" the noises typically on properly designed equipment. There isn't really a level where noise reduction isn't a benefit. But hearing is believing. I assure you I'm excited for the demo reviews to come in! The IndieGoGo has over a month, so you don't have to decide before anyone hears it!
The product will be the 4th generation PCB, 4th or more of the Kmultiplier version, at least 5 in parts, 3rd in PSU's, and 3rd in transformers. Surely I lost a few in there I didn't count
I had no idea the euro dropped like that!
The PSU and transformer greatly simplify the end users needed capability, multiple people asked for it, it allows me to show off the PSU so people can use it for other projects, and it ensures you get very good performance. Different voltages change the voltage regulation factor, and the saturation factors of the Kmultiplier.
You may not have audibly apparent noise Radiosmuck, that doesn't change the fact that noise is changing the sound of your music! This is one of the hardest things to explain to someone looking for or wanting to understand power conditioning. You hear what the noise does, every moment you listen, but you don't "hear" the noises typically on properly designed equipment. There isn't really a level where noise reduction isn't a benefit. But hearing is believing. I assure you I'm excited for the demo reviews to come in! The IndieGoGo has over a month, so you don't have to decide before anyone hears it!
I read you tested 23V PSU and in the end decided on 18V. You do not mention speakers, other than 91dB high efficiency, but looking at data they must be 8 ohm?
Did you try the 18v powered board with a 4ohm dummy load or real 4 ohm speaker and could you tell me what happens past the point when tda7297 reaches its 2A limit at 8v output into 4 ohm load ?
Did you try the 18v powered board with a 4ohm dummy load or real 4 ohm speaker and could you tell me what happens past the point when tda7297 reaches its 2A limit at 8v output into 4 ohm load ?
Actually the Piegas it was tested on at 91/db performed great, which are 4ohm. And that was 2v lower than 18v.
18v is the spec sheet limit. It's intended to live it's life at that, but tolerate 20v (hidden in data sheet, not continuous). 23v sounds fantastic but does it last 3 weeks or three decades at that? I can't answer that.
When the amp clips it sounds a little busier than normal, a little less focused, but won't stop you from listening. The best thing for best performance is voltage. It handles itself surprisingly well, perhaps due mto its balanced output.
18v is the spec sheet limit. It's intended to live it's life at that, but tolerate 20v (hidden in data sheet, not continuous). 23v sounds fantastic but does it last 3 weeks or three decades at that? I can't answer that.
When the amp clips it sounds a little busier than normal, a little less focused, but won't stop you from listening. The best thing for best performance is voltage. It handles itself surprisingly well, perhaps due mto its balanced output.
It clips. I didn't put 50v into the input or anything like that, but it softly clips when it exceeds it s power ability. It's actually a rather beneficial attribute. Amplifiers clip all the time under duress but some sound nice (7297) on the ears, others hurt.
No matter what the impedance curve looks looks like on the Premium 5.2's, it did awesome at higher tolerable levels. It had bass, force, imaging, all of it.
I purchased an LM317 voltage regulator and connected it to a spare transformer and my lunch money amp drove my very large efficient speakers from a supply range of 4.5v to 18vdc.
I noticed no audio difference when using the switching ps or the LM317.
If I don't hear it it's not there.
Anyway good luck with your project.
You're observation is not related to my project in any way what so ever, but thank you I'm excited.
PCB's will be printed at the same time, but honestly I don't expect any leap of faith! I have the demo that's about to be circulating.
That particular sample doesn't need higher voltage--it was suffering either overcompensation or overdampening. Higher voltage could offset those, but isn't an appropriate fix. Internally, the chip is not overcompensated at all (actually they're slightly sketchy, definitely not overcomp). Therefore, I propose that the sample you're speaking of, is suffering from overdampening.
Solving the overdampening mistake is within the areas of layout and power circuit. That will need a review and revisement to solve the fault.
Basically, you have detected a missing series element; but, then there is the puzzle of how much, what type, and the most effective locale for it. Good luck!
EDIT: I have but one clue for you, and I hope that it helps: When you put the right amount of loss in a just right locale, the headroom will INCREASE. That's what happens when you repair and/or cancel an overdamp fault. So, when on a search for the applicable locale, observe the practical headroom, which will let you know if the repair has been appropriate.
Last edited:
TDA7297 high noise, no detail, bass
Hello, I would like to ask for some help.
I bought a TDA7297 amplifier (this) and I tried 3 different power source (DC) and I got no bass.
I tried with
5V 1A (USB), no noise, but no bass
12V 12A (XBOX360 PSU), big noise, no bass
8.5V 1.5A (Camera charger), big noise, no bass.
Could you help me Guys why I have a crap sound quality? Everywhere I read this amplifier have good sound quality , but not for me.
Thanks!
Hello, I would like to ask for some help.
I bought a TDA7297 amplifier (this) and I tried 3 different power source (DC) and I got no bass.
I tried with
5V 1A (USB), no noise, but no bass
12V 12A (XBOX360 PSU), big noise, no bass
8.5V 1.5A (Camera charger), big noise, no bass.
Could you help me Guys why I have a crap sound quality? Everywhere I read this amplifier have good sound quality , but not for me.
Thanks!