My "audiophile" LM3886 approach

Results for My ref C.

Have to say i am very impressed with the amp so far. Very detailed without being harsh and it has a great control. It also also very quiet.

In a way it sounds like a really nice hybrid of the Tripath and LM3886 chip amp sound, the detail of the tripath but without out the slighly fatiguing sound and and the body of the LM based amps but with more detail and better bass control.

I really like the sound though i might trying fiddling with a more exotic input caps since that really seem to have a very positive effect on my Amp3.

I am currently using it with Rudi's dc nulling version of the AD815 preamp. What are people opition of using it with active vs passive preamp. I found the preamp made a very postive effect to the straight LM3886 subber setup though it had a less obvious effect on the Amp3.


Phil
 
I found I prefer the my_ref passive.There is plenty of gain and dynamics.Mauro suggests a low imp pot around 20k.
Now I have tried two pretty good preamps -one is based on ad 8610 plus EL buffer in a Walt Jung type nested feedback circuit and the other is the Broskie Aikido which is better.But both slightly colour the sound and there is a loss of transparency even though some may like that they bring some parts of the sound a little more forward.But for me they subtract too much info.
I am seriously thinking of investing in a good TVC passive pre such as the Django.Too bad that it costs three -four times more than a my_ref.
 
protos said:
I am seriously thinking of investing in a good TVC passive pre such as the Django.Too bad that it costs three -four times more than a my_ref.

A single ended Joshua Tree may be just what you are looking for. It works very well directly into the REVC as a passive attenuator, and the cost is quite a lot less then similar devices. The output impedance is a constant 750R and the input impedance in the worst case is about 2.2K, but for most of the range it is around 5K.

Cheers!
Russ
 
It would be interesting to compare a good step attenuator with a TVC.
At the moment I have a small Creek passive pre which is basically an Alps motorized pot.I also have an expensive DACT stepped att in the Aikido pre.But I have not directly compared the two attenuators.
Most people have preferred the TVC over simple passive resistive vol controls quite clearly although as I mentioned they would cost as much as a very good DIY active tube pre.
 
Hi Protos,
I have a TVC based on Stevens&Billington's TX, bought at Bentaudio.com . I think Django lacks the excellent OCC copper wiring, wich is a very detailed wire. You could build a TVC yourself for around US$650, I believe. I bought mine before (copper) price went up.

My "My-ref" is integrated with a Kookaburra and it is very detailed and controled ;) it costs ten times less than the TVC!
For a direct comparison between TVC and Kookaburra, I am waiting for my new "my-ref" monoblocks.

Cheers,
M
 
i was just wondering if anyone had swapped out the sockets and soldered the opamps straight in. Anyone done this and was there any aprieciable difference.

By the way the amp is burning in nicely. Been running it for a week staright and it is a different amp to when i first powered it on. Never had an amp change as much. Odd.
 
filholder said:
i was just wondering if anyone had swapped out the sockets and soldered the opamps straight in. Anyone done this and was there any aprieciable difference.

By the way the amp is burning in nicely. Been running it for a week staright and it is a different amp to when i first powered it on. Never had an amp change as much. Odd.


Yes, it is best to solder the LM318 directly in. The sockets are there just in case a chip goes bad if someone makes a mistake to make replacement easier.
 
So how is the Oak case coming along?

My father helped me build my first car stereo woofer box over 20 years ago.. He 45'ed all the joints and made a wonderful box that he could be proud of.. It stayed in my car a week before I realized I made the wrong calculations and had another made..

To this day I STILL tell him thank you for taking the time to help me make my first box even though he thought it was "silly". He STILL thinks my stereo and electronics are silly :D .

I hope that Oak case means as much to you as my box does for me. No mater how it looks or sounds in time you will still remember the box long after the amp is gone.

Sorry to be all sentimental, but to me THAT is perfect DIY.. When you can work with dad / son..
 
Hi,

My parents have been in Canada for the last few weeks staying with some family friends (someone my great uncle servered with in the RAF). Also in live in London and my parents live in gloucestershire so i don't get to see then as much as i would like. Probably go there next week and find out what Dads been up to. He's just retired but has been a chippy for about 50 years. Should be a walk in the park for him but it will mean a lot to me.

Like you said that sort of thing is great cos it means you can spend time working with your old man and have somethign really nice at the end of it.

Phil
 
derekyu said:
Hey, Russ.
I finished the Ref A two months ago and am planning to build for the Ref C.

As i am going to buy a pcb not a kit, would u tell me the psu cap (10,000) how big it is in the pcb? 35mm or 30mm?

I find a 30mm cap is not so easy.

It is 30mm.

There are many 50V 10,000uf caps in that size.

We supply Panasonic caps in the kits.
 
Hi guys,

I've just built a plain vanilla Gainclone...from the reports on this circuit's performance, I wish I read this first!

I assume everyone's in agreement that it's much better than the old GC (in whatever iteration)?

Anyway, I've never etched a board before and would like to have a go. There was a post way back with a link to the RevC board artwork, but the link is gone.

Russ, only if you don't mind - can I get a copy of that please?

I'm buying a kit anyway as you guys can source the components for about the same price I can but I don't have the hassle!
;)

Ed, love the Mongo talk. I've been laughing all the way through this massive thread.

Cheers

Stuey
 
Stuey said:
Hi guys,

Russ, only if you don't mind - can I get a copy of that please?

Cheers

Stuey


Sure,

Here are the original single side PDFs.

Keep in mind this was one of the first PCBs I ever worked on. So its not very polished. But it is functional. :)

Cheers!
Russ

Here is the copper side PDF.
 

Attachments

  • monobloc_rev_c_copper.pdf
    94.1 KB · Views: 383
I'd love to get my hands on one of Mauro's new versions of the REV C, sadly he is not selling PCB's outside of Italy, and the assembled boards he is willing to sell start at EUR130... looks real nice... but a bit pricey after imprt duties etc for a pensioner...

EDIT stupid me... Might as copy and paste some of what he e-mailed me.

Evolution is a "my personal signature improvement" of My_ref revC project, plus upsurge of the possibilities range of use, like a "dubble low impedance power".

The limits of the "preliminary" project "My_ref revC" is:
- Sound depended of 220uF elcap (NFB net) versus 1uF input caps (the omitted tech. is very critical and not decisive...)
- Little "compression sensation" at very large sound level and low impedance speakers
- Clipping conditions not more "elegant" and more depended to LM318 and power PSU, (voltage and power)
- Not definitive and complete PCB layout. My original PCB is a good approach of stereo system, but with little problems of PSU induction. I not have tested the Russ design, but a "monoblock" solution not is "my conceptions" of "the best stereo audio amp". If a monoblock it's assembled in a single box, with single trafo, single input and single output, the situation is OK (obvious, if a PCB layout is OK). But if you assemble one stereo amp with 2 monoblock and 1 trafo, the situation it's very problematic....
- some time there are problems of stability. This problem depended of a quality of components, the layout (PCB and cabling) and tollerance of LM318N , LM3886 and net caps, and ground (earth) loop and shielding

The evolution "base":
- resolving all previous list problem (not a few...), with a "driven" assembling, cabling and shielding (PCB structure) and high quality SMD critical components
- add a new "no ELcaps NFB net". One "esclusive" design (active "tuning" DC tracking) performing an total control of ultra low freq. and DC signal. The audio bass performance is simply a "state of art".
- add a new LM318 PSU regolator, with a "linear phase" and "linear impedance" in all audio band.
- performing a new "reverse driven allineaments", more linear

Conclusion: The good base sound of My_ref + more audio dinamic, very impressive bass and more "spatial sound stage", detail and high freq. performances. Another my very problem, more more "repetitive and stable" technical situation...

The evolution "full":
Identical PCB board, only more Traditional components, increase the total current capability.
This solution, like a "LM3886 parallel config.", have a little more THD (third) of "base" version, but dubble current and low impedance load capable.
Like My_ref, this problem is little meaningful, but my idea it's a possibility of developed (in future) an "monoblock" "bridge" or "balanced" solution, for power increase or simple a "good balance approach" (what for the time being doesn't interest me...)
This version have a dubble DF (damping factor) of a "base" version (and My_ref revC version) and a little degrade of a "reverse driven" response . It's increase a bass and mid bass control, but not always is an amelioration of high frequency.
The power headsinks increase, and the case and the thermal cupled has to be more accurate.
I suggestion to passing to the "full" version after having tested the "base" version. in a lot of cases it doesn't feel the necessity to passing to " full " solutions...

--------------------------------------------------------------------------
Shipping situations:
I have a very "problems" of International shipping. Recently I have ship an package to US, and the shipping time has been of one monthes (with a shipping cost about 25euro....) !!
My principal problem is that I don't dispose of a "commercial abutment" to sell the boards, because my initiative is private and for he has many boards....

For this reason, I sell normally the boards only in Italy.
In any case, this are the costs and pay method (no commercial situation) of the new Boards (The cost is too high compared of a "simple" My_ref kits...):

The method of payment that is the "banking international credit transfer", with Euro currency and without banking direct expenses to my load, or the international postal money order.

Cost of shipping of: 1 PCB board, (SMD pre-solder + 2 custom headsinks)= 130 Euro
Cost of shipping of: 1 Kit Evolution "base" (PCBcomplete + "base" BOM)= 220 Euro
Cost of shipping of: 1 Evolution "base" mounted and tested (PCBcomplete + "base" BOM)= 270 Euro

Shipping cost:15 euro Italy, 30euro (hypothesis) other States.