Dario, Thanks. I tried to fit the improved LM317 CCS + improved shunt, but the space was tight and requires moving many things. So I left that for a future revision (if this layout turns out to be defective, I'll include the regulator changes along with the corrections).
C1, C2 could be 330 or 470uF - optimal value is yet to be determined. If you look at the trace lengths, it is about as close as it is on V1.4. The AC current through the rails to each LM3886 is highly correlated (nearly identical), so it just approximately doubles the ripple current at C1, C2. One could use 4 caps, or just higher-valued caps with sufficient ripple-current capability like Elna RSH/RJH/RJJ or Panasonic FM or Nichicon Muse KG/KZ/FG/FW or Black Gate STD.
You're welcome! 🙂
I know your goal is to use only through hole parts but if you use SMDs just for the two pumps you can gain a lot of space and improve pump performance too. 😉
I'm sure you did your math about it but by instinct I would have preferred 4 caps nearer to the opamps.
Also, for a double pump setup it would be better to go higher with smoothing caps, say 15000 to 22000 uF but since you used 35mm caps it shouldn't be a problem.
Yup, no problem. Actually, my testing prototypes will use 12,000uF/63V, but kit-builders can easily upgrade to higher values of their choice - Panasonic TS-UP 22,000uF/50V will fit, along with a bunch of Nichicon KG Type I.
I'm experiencing something strange with my amplifier and is not the first time, a had the same problem with a gainclone.
The problem is the following: without signal when i move the potentiometre to the maximum i can hear a hum, the protection relay clicks and i can see my woofer is making a long movement first inside and as im moving the potentiometre back to minimum goes outside.
Also the LED is dimmed a little bit whem that hum appears on the speaker.
I'm i doing something wrong as i had the same problem also with the previous gainclone ?
The problem is the following: without signal when i move the potentiometre to the maximum i can hear a hum, the protection relay clicks and i can see my woofer is making a long movement first inside and as im moving the potentiometre back to minimum goes outside.
Also the LED is dimmed a little bit whem that hum appears on the speaker.
I'm i doing something wrong as i had the same problem also with the previous gainclone ?
this does not seem correct.
Disconnect the speaker and measure the output offset as you change the vol pot rotation.
Disconnect the speaker and measure the output offset as you change the vol pot rotation.
That's great Siva !! Several months ago I asked if a mirrored dual LM3886 configuration was possible. The answer was a resounding group NO!
This appears to be a big step beyond that with all the advantages of a single board.
I'm very eager to try it in a new integrated build with a JC-2 pre and a DIY DAC all in a single box. 😀
This appears to be a big step beyond that with all the advantages of a single board.
I'm very eager to try it in a new integrated build with a JC-2 pre and a DIY DAC all in a single box. 😀
Do we have Mauro's permission to show details of Evo?
Even though the PCB is Linux' creation, it is based on Mauro's schematic, which as far as I know he has not given permission to publish.
Even though the PCB is Linux' creation, it is based on Mauro's schematic, which as far as I know he has not given permission to publish.
Mauro gave us the My Ref in it's various forms for a single chipamp implementation.
He went on to develop a dual chipamp version, parallel I believe, and named this Evo.
Again as far as I know Mauro has resisted all calls from Forum Members to release Evo details for open use/reproduction/cloning.
Post3817 confirms this early release PCB (not yet double checked) is based on Evo.
He went on to develop a dual chipamp version, parallel I believe, and named this Evo.
Again as far as I know Mauro has resisted all calls from Forum Members to release Evo details for open use/reproduction/cloning.
Post3817 confirms this early release PCB (not yet double checked) is based on Evo.
Last edited:
Siva it's not the first one that proposed a double pump My_Ref (apart Mauro):
http://www.diyaudio.com/forums/chip-amps/54571-my-audiophile-lm3886-approach-54.html#post1139692
While Mauro's My_Evo (Evolution) uses a double pump I'm not sure it's enough to say that Siva's project is derived from the My_Evo.
I do agree that the interim name of Siva's project it's not the best one and may be misleading.
http://www.diyaudio.com/forums/chip-amps/54571-my-audiophile-lm3886-approach-54.html#post1139692
While Mauro's My_Evo (Evolution) uses a double pump I'm not sure it's enough to say that Siva's project is derived from the My_Evo.
I do agree that the interim name of Siva's project it's not the best one and may be misleading.
Lukio does not claim it is derived from Mauro's Evo.
But Linux wrote
That brings Linux's comment to read
"Dual Howland Evo2 monoblock board, dimensions 127mm x 100mm"
Linux is clearly stating that the PCB is based on Evo2
Your
But Linux wrote
All sentences with brackets must read and make sense and give the same message with the contents of the bracket removed.
That brings Linux's comment to read
"Dual Howland Evo2 monoblock board, dimensions 127mm x 100mm"
Linux is clearly stating that the PCB is based on Evo2
Your
cannot be right.
Last edited:
You're right Andrew but I can't see either where Siva claims that...
On the PCB we can read 'Derived from MyRev RevC'
The unique possible referral is the unhappy choice of the name, isn't it?
Or I've misssed something?
What I understand from that sentence is that Evo2 is the name Siva gave the project.
But I could be wrong, I suppose we should wait Siva's clarification on it.
Last edited:
Yup, this is actually closely derived from the MyRef Rev C, not Mauro's proprietary Evolution design (of which the only publicly available schematic is an incomplete one in an Italian audio magazine, AFAIK). The placement of the large Elcos has a superficial similarity to the those in published pics of the Evolution, but the resemblance ends there - the layout is entirely from scratch, though portions of it are similar to the Twisted Pear V1.2, my V1.3 and V1.4 boards (which were already close to optimal for a single-Howland layout).
I lifted the name Evo2 from a Mercedes-Benz performance car from the '80s, which was derived from a stock 190E baby Benz (one of my favourite cars). Similarly, this dual Howland is simply a hot-rodded version of the stock MyRef Rev C/E, no more. I posted the schematic on the forums last year, but didn't get around to completing the layout and getting it sent for fabrication until this month. A glance at that schematic will confirm that it's really a MyRef derivative.
I can change the name from Evo2 to MyRef2, if that brings more clarity regarding its origins - though it still has the potential to ruffle some feathers, like my previous derivative design, the MiniRef. I'm open to suggestions.
Last edited:
I like "MyRef_V.X2" or "MyRef_X2V". Even sounds like a fancy car - or a super fast computer😎
Last edited:
hehe V.X looks like an emoticon. does sound good, though.
myref_redux?
otoh, my myref's about to go full exotica with some vcap teflon caps that i'm seriously considering from partsconnexion. oh they also have polystyrene caps in all the values i need. what do you know.. 😀
myref_redux?
otoh, my myref's about to go full exotica with some vcap teflon caps that i'm seriously considering from partsconnexion. oh they also have polystyrene caps in all the values i need. what do you know.. 😀
I suggest you to follow the same guidelines Mauro gave me for the Fremen Edition.
Use a name that makes clear that your amp is not an official My_Ref (or worst MY_Evo) revision.
Something like My_Ref Indian Twins will do 😉
What about moving completely away from the reference and start using "LM3886" something or other - maybe LM3886_X2e or the like. That would simplify much of the issue. A possible fork in the road.
Last edited:
