Goldmund Mimesis 3 Clone, World's Best Amplifier!!!!

Status
Not open for further replies.
Hello DIYAudio members! Everyone who has been following me on these two threads, http://www.diyaudio.com/forums/solid-state/174468-very-best-amplifier-i-have-ever-heard.html , and http://www.diyaudio.com/forums/group-buys/175296-pcb-order-goldmun-clone.html , I will from now on will only post here. To everyone else, please enjoy the ride.

I've been in the audio industry for 10+ years and have heard everything from Levinson, Bryston, Pass, Ayre, Rotel, Boulder, Cary, and 90% of all other audiophile manufacturers. I've never heard anything to even belong on the same league as the Goldmund Mimesis 3 amplifier. The design is state of the art, the sound is world class. Everything else doesn't matter.

The goal of this project is simple: To clone the Goldmund Mimesis 3 amplifier 100%. I will not take any suggestions, or upgrades on the circuit. Not negotiable. Only on the PCB layout and construction of the amplifier.

Here's where this project stands and where it will be heading in the next week:

1.) Please see the original, 100%, accurate schematic of the Goldmund Mimesis 3 and the protection circuit below.

2.) The PCB layout is nearly completed and I will be posting it early next week. The basic layout remains as Alex has created, with the now correct parts to reflect the Mimesis 3 amplifier, correct part sizes, correct lead diameter holes, etc. I will be posting the complete part list early next week as well.

3.) I will order a pair of the PCB boards from a high quality US manufacturer. The order will be placed the following Monday after Thanksgiving Day.

4.) I will build this amplifier step by step on this thread, with pictures and explanations. The PCB boards will be tested for quality, functionality, fit of parts, etc. The final amplifier will be tested on a Tektronix.

5.) Goldmund Mimesis 3 has the following power rating specs: 100 watts (2-8 ohms) and 200 watts at 3 ohms. The rest can be found here: GOLDMUND MIMESIS 3 STEREO POWER AMPLIFIER - SPECIFICATIONS

6.) NOTE: All electrolytic capacitors in the schematic have the + shown as an empty rectangle and a - as either a black filled rectangle, or a line.

This is an amazing project that rarely comes about. I hope everyone enjoys it :)

Here is the 100% accurate schematic of the Goldmund Mimesis 3 Amplifier:

An externally hosted image should be here but it was not working when we last tested it.


Here is the Goldmund Mimesis 3 power supply (100% exact):

An externally hosted image should be here but it was not working when we last tested it.


Here is the Goldmund Mimesis 3 speaker protection circuit (100% exact):

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.


Goldmund Mimesis 3 specifications:

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Agisthos - I've never said that the Goldmund Mimesis 9.2 amplifier was the best I've heard. In that thread I basically said that Goldmund in general was the best amplifier I've ever heard and Mimesis 9.2 schematic just happened to be on hand at the time so I posted it.

But out of all the Goldmund amplifiers, Mimesis 3 is by far the best sounding amplifier. And it also happens to be more practical and easier to build. So this will be my project, if you would like to follow it.
 
Glad to hear that this didn't die! I was starting to lose hope that this would never get going.

Nagys, what are the main difference between 9.2 and 3? I am assuming that 3 is even older than 9.2. Why does it sound better? In 9.2, did Goldmund cut corners?

Nagys' I thank you for going with the build first, now hopefully all the other folk will be pleased and there will no longer be any gridlock and attacks to this thread.
 
its got lower voltage rails, some folk here say that this makes amps sound better. Intersting though that I recall reading that the high voltage in the 9.2 might have been for sonic reasons as the output voltage was way high for the power output. On second looks, this has lower front end rails than the output. This amp was made uptill 93 I think. The 9.2 was made from 92 till 2002, thats an eternity in this business.
 
Luke - The schematics for Mimesis 3 and Mimesis 9.2 are 95% identical. Mimesis 9.2 features a few more bypass/filter capacitors, one extra pair of MOSFETS, higher rail voltages, and a few other extremely minor differences.

Here is an updated Goldmund Mimesis 3 power supply schematic, showing more detail:

An externally hosted image should be here but it was not working when we last tested it.


And an updated Goldmund Mimesis 3 protection circuit schematic (NOTE: The series inductor on the output is 18AWG):

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.
 
Last edited:
Just to prevent any arguments...

1.) T5 and D1 are connected correctly.

2.) D2, D3, D4, and D5 are connected correctly.

3.) C11 and C21 are connected correctly.

4.) D6, D7, D8, and D9 are connected correctly.

This is the way Goldmund does it and that is all that needs to be known. Please no endless arguments on what YOU think is the proper way of connection. This dead horse has been beaten over and over again in my previous threads and it's getting rather old.

That's how the Goldmund circuit is and that's how the PCB board will be laid out and manufactured. End of story.

Furthermore, the original Goldmund Mimesis 3 amplifier has a logic circuit board behind the front plate and another one behind the rear plate. These boards have multiple relays, chips, etc. It has a delayed turn on, thermal shut down, line filtering and other useless gadgets. These will not be shown in any schematic, nor will be a part of the PCB board, or in any way a part of the cloned amplifier.
 
Last edited:
Just to prevent any arguments...

2.) D2, D3, D4, and D5 are connected correctly.

.

These diodes can be connected this way, but when conducting,
they will be in serial with R19 , thus being less effective to limit
the relevant voltages...

R5 should rather be 30K and not 3K as in the amps 100% accurate
schematic..

Other than that, this amp as the 9.2 , is quite fast at 60V/uS SR,
but its linearity is of another time, that is, poor compared to current
decent standards...
 
Excellent news, Nagys :up:

I have a few questions about the layout in the photo:

There appear to be three "little black boxes" per channel. I assume one is the input module and one is the relay, but what is the third?

The little blue thing on the heatsink looks like a thermistor, but I don't see any on the circuit diagram? Is it maybe something to do with biasing, or the protection circuit?

I'm also curious as to what part of the circuitry is in the separate compartment at the back of the amp. Is that just the protection circuitry?

Cheers - Godfrey
 

Attachments

  • gm.JPG
    gm.JPG
    38.1 KB · Views: 4,053
dadod,

You are correct because if we follow the schematic T7 and T9 are no longer a current mirror.

Well it is good news that Nagys is going to build this, so that he can correct such errors before the group buy begins.

Contrary to what Nagys might believe I want this to be a successful project but built correcly and consideration given to members that don't have the ability to debug the design if problems should arise.

Jam
 
Last edited:
Hi Nagys,
There is an error on mimesis3 schematic, base of the T9 should be connectet to collector via capacitor (probably 10pF) not directly.
It's OK, there's a compensation cap across T8. Nagys did mention that different models were slightly different. Then again, it might be smart to allow for a cap there on the PCB - just as an option for tweaking.
... T7 and T9 are no longer a current mirror...
Yes, the connection between the base and collector of T9 needs to be cut to fix that.

Another small detail: maybe C11, C14, C18 and C21 could be a higher voltage rating, just for reliability.

Otherwise, it's looking good. :)
 

Attachments

  • gm2.GIF
    gm2.GIF
    8.6 KB · Views: 3,521
Wahab and Dado - Thank you for catching the mistakes. I've been working late last night and can't always be 100% perfect. R5 should be 33K. The Base and Collector of T9 should not be connected. For the PCB, I will leave mounting holes in that area so that if for some reason the amp is not stable, a 10pF capacitor can be installed there. The Goldmund Mimesis 3 does not have a capacitor there, but the Goldmund Mimesis 9.2 does. It's probably all dependent on PCB layout. Please see the corrected Mimesis 3 schematic below:

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Godfrey - The back circuit board has part of the protection and line filters. The front circuit board has the logic circuits, delayed turn on, thermal shutdown, etc. The first block is the A1 module. The second block is the S1 module. S1 houses part of the protection circuit. The third block is the relay. Yes, the blue thing is a thermistor.

The only thing I will incorporate is the protection circuit. I will not incorporate any other gadgets, nor will I post their schematics. Please do not ask. I don't have the time to take the amp apart and trace them. Some will even be impossible to trace, like the mains line filter, as it's potted.
 
Status
Not open for further replies.