You're welcome Andrew.
I know that my enthusiasm on component selection, materials, and so on can sound silly/naive/irritating to more technically prepared people like you but, although in a different field (computer science), I'm a tech too with a technical diploma.
Maybe that enthusiasm can be, sometimes, misleading.
Hi Andrew, as noted - Dario hit all the significant technical points. My purchase path/history matches the progression of the improvements due to revisions and developments for the LM3886 family.
Since the introduction of each concept/version/flavor, developers and experimenters have contributed upgrade paths that have proven to be quite useful and effective. The BrianGT stands somewhat separated, as it has no reference to anything "MyRef". I have made no changes to the original parts on the kit other than Uriah's Resistor Replacer. That device significantly cleans and warms the signal to the point I would recommend it as a "Must" for that amp. I also built two BPA-150s for use with 4 ohm speakers but that project is sitting on the shelf. Those and the BrianGTs could well be all that's desired for those wanting excellent sound and have not been sucked into the quest for "Audiophile Nirvana"
For the MyRef family there are a few universal items that produce dramatic improvements:
1. Mundorf AG caps for C3 & C8 - everything from the mids down gains power, depth, punch and control.
2. C9 - Black Gates I believe are the acknowledged best choice, but are difficult to find and the price can be dumb high. In any case that position is critical for clarity and stage.
3. C13 - same as above in importance but can be more easily chosen to match a user's preferences and system.
The singular standout for all MyRefs, up to the FE and even for stock builds, is Siva's LF01 module. When it is added a whole new amp appears - warm, clean, smooth and inviting.
Though I thought I might be spinning my wheels a bit, I purposely decided to build the Fremen Edition from the stock Mouser BOM Dario provided. In truth (a little confession is good for the soul) I held back in reporting how good the FE sounded because I didn't want to sound hyperbolic. It's ironic that you ask this question today - cause yesterday was the first time I chose to use the phrase that came to mind when I first powered up the amps on really good speakers - "a quantum step". I did, in fact use that phrase in a PM to Tom E. several months ago.
I never thought I would be doing things like using sockets for swapping resistors and caps, or going so far as reversing polarities for components - but as I follow the designer's lead, the benefits have been born out. I am confident that when a few more people build the FE and report what they hear, the idea of a big step forward in top end DIY audio will be confirmed.
Since the introduction of each concept/version/flavor, developers and experimenters have contributed upgrade paths that have proven to be quite useful and effective. The BrianGT stands somewhat separated, as it has no reference to anything "MyRef". I have made no changes to the original parts on the kit other than Uriah's Resistor Replacer. That device significantly cleans and warms the signal to the point I would recommend it as a "Must" for that amp. I also built two BPA-150s for use with 4 ohm speakers but that project is sitting on the shelf. Those and the BrianGTs could well be all that's desired for those wanting excellent sound and have not been sucked into the quest for "Audiophile Nirvana"
For the MyRef family there are a few universal items that produce dramatic improvements:
1. Mundorf AG caps for C3 & C8 - everything from the mids down gains power, depth, punch and control.
2. C9 - Black Gates I believe are the acknowledged best choice, but are difficult to find and the price can be dumb high. In any case that position is critical for clarity and stage.
3. C13 - same as above in importance but can be more easily chosen to match a user's preferences and system.
The singular standout for all MyRefs, up to the FE and even for stock builds, is Siva's LF01 module. When it is added a whole new amp appears - warm, clean, smooth and inviting.
Though I thought I might be spinning my wheels a bit, I purposely decided to build the Fremen Edition from the stock Mouser BOM Dario provided. In truth (a little confession is good for the soul) I held back in reporting how good the FE sounded because I didn't want to sound hyperbolic. It's ironic that you ask this question today - cause yesterday was the first time I chose to use the phrase that came to mind when I first powered up the amps on really good speakers - "a quantum step". I did, in fact use that phrase in a PM to Tom E. several months ago.
I never thought I would be doing things like using sockets for swapping resistors and caps, or going so far as reversing polarities for components - but as I follow the designer's lead, the benefits have been born out. I am confident that when a few more people build the FE and report what they hear, the idea of a big step forward in top end DIY audio will be confirmed.
Last edited:
Hi Hurtz,
sadly no measures, I don't have the necessary equiment.
But subjectively the differenceis is obvious, we're not talking about nuances.
Hmm too bad, would really interest me.
Although I'm a little bit suprised that so few people have/use measurement equipment, considering the amount of interchangeable parts I wouldn't trust my judgement after the third "revision".
Could you just do some simple rmaa measurements? I'd really like to know how it compares to my lm3886 concerning IMD and noise.
cheers!
hurtz
Although I'm a little bit suprised that so few people have/use measurement equipment, considering the amount of interchangeable parts I wouldn't trust my judgement after the third "revision".
Could you just do some simple rmaa measurements? I'd really like to know how it compares to my lm3886 concerning IMD and noise.
cheers!
hurtz
Same here hurtz,
All I own is a Radio Shack sound level meter that just gives general SPL readings. However, in the list of options I gave please not the indication of "dramatic" relating to those four. They are not just my opinions but reflect a compilation of reports/agreements across several LM3886/MyRef threads on the forum.
There are a number of things like wire composition, connector styles/brand, interconnect length and so forth that I would consider "tweaks". I tried to avoid those and give a summary of what are honestly considered to be upgrades.
If you can describe the measurements you are seeking, we might be able to point you to some relevant posts already on the threads.
I am actively searching for an affordable PC based diagnostics tool that will give usable information. If you you are aware of such please pass along the info. That would be both helpful and a lot of fun.
All I own is a Radio Shack sound level meter that just gives general SPL readings. However, in the list of options I gave please not the indication of "dramatic" relating to those four. They are not just my opinions but reflect a compilation of reports/agreements across several LM3886/MyRef threads on the forum.
There are a number of things like wire composition, connector styles/brand, interconnect length and so forth that I would consider "tweaks". I tried to avoid those and give a summary of what are honestly considered to be upgrades.
If you can describe the measurements you are seeking, we might be able to point you to some relevant posts already on the threads.
I am actively searching for an affordable PC based diagnostics tool that will give usable information. If you you are aware of such please pass along the info. That would be both helpful and a lot of fun.
Last edited:
Hi bcmbob!
Since sound quality is so highly subjective I try to concentrate on a technically perfect build, whatever improvements in listening quality may come with it ;-)
So I am naturally keen on measuring those differences, to not at last, get a better price/value ratio.
I would think of rightmark audio analyser at first, http://audio.rightmark.org/ its a free tool to measure THD, noise, IMD, crosstalk and some more using a pc sound card. Some time ago I used an asus xonar d2x which has excellent noise and dynamics rating. It goes for around 130€.
A cheaper and most probably still sufficient alternative would be the asus xonar DX around 65€.
I also heard that the auzentech soundcards perform equally well and are a bit cheaper at that, but I never had one of those.
The method of testing is quite simple, rmaa will play a series of different test tones, record the resulting wave-file and afterward analyse it.
Now to measure you connect the line-out from the soundcard to the line-in of the amp, and the power-out via a voltage divider (or much better/safer opamp) to the line-in of the soundcard. Take good care when using a simple voltage divider and triple check that all cables are what they pretend to be... I killed one of my line-in channels of my d2x...
If you google "asus xonar d2x rmaa" the second link will give you a pdf with a good guide on how to do the loop-back test. For any further questions, don't hesitate to ask!
cheers,
hurtz!
Since sound quality is so highly subjective I try to concentrate on a technically perfect build, whatever improvements in listening quality may come with it ;-)
So I am naturally keen on measuring those differences, to not at last, get a better price/value ratio.
I would think of rightmark audio analyser at first, http://audio.rightmark.org/ its a free tool to measure THD, noise, IMD, crosstalk and some more using a pc sound card. Some time ago I used an asus xonar d2x which has excellent noise and dynamics rating. It goes for around 130€.
A cheaper and most probably still sufficient alternative would be the asus xonar DX around 65€.
I also heard that the auzentech soundcards perform equally well and are a bit cheaper at that, but I never had one of those.
The method of testing is quite simple, rmaa will play a series of different test tones, record the resulting wave-file and afterward analyse it.
Now to measure you connect the line-out from the soundcard to the line-in of the amp, and the power-out via a voltage divider (or much better/safer opamp) to the line-in of the soundcard. Take good care when using a simple voltage divider and triple check that all cables are what they pretend to be... I killed one of my line-in channels of my d2x...
If you google "asus xonar d2x rmaa" the second link will give you a pdf with a good guide on how to do the loop-back test. For any further questions, don't hesitate to ask!
cheers,
hurtz!
Thanks hurtz,
I've been flirting with the xonar cards since they came out. Unfortunately, all my extra cash has been diverted to amp building. I do have a M-Audio Delta 66 and a Delta 1010 that might be usable with the rightmark program. I downloaded it a couple weeks ago, but now that I have someone to guide me, I'll start trying to get my head around it.
Thanks again.
I've been flirting with the xonar cards since they came out. Unfortunately, all my extra cash has been diverted to amp building. I do have a M-Audio Delta 66 and a Delta 1010 that might be usable with the rightmark program. I downloaded it a couple weeks ago, but now that I have someone to guide me, I'll start trying to get my head around it.
Thanks again.
Hi Hurtz,
I realize now that maybe you want to see some measure of the orginal My_Ref to compare it to a Gainclone... Am I right?
In the original My_Ref thread there are several measurements made by Mauro Penasa (the designer) and some of Soongsc.
Read the attachment and from post #90 of the linked thread.
From the attached document:
Bandwith (tipical): 2Hz-70Khz
Output power(±37V): 40W/8ohm – 56W/4ohm
Damping Factor (8ohm): >200 (frequency and compensation depended)
S/N 600ohm: >90 dB (compensation depended)
THD e IMD (tipical), 20Hz-20Khz, 1-40W/8ohm: <0.05%
Thanks!
That was pretty much what I was looking for. Sadly though I can only test my builds using rmaa and have no fancy test-equipment for phase shift and such. Maybe when I finish my studies I'll have the budget to afford such things
In the meantime rmaa all the way ;-)
Also I think there a quite a few microphones out there to measure output of the speakers but they are much more expensive than a simple sound-card.
cheers!
hurtz
That was pretty much what I was looking for. Sadly though I can only test my builds using rmaa and have no fancy test-equipment for phase shift and such. Maybe when I finish my studies I'll have the budget to afford such things
In the meantime rmaa all the way ;-)
Also I think there a quite a few microphones out there to measure output of the speakers but they are much more expensive than a simple sound-card.
cheers!
hurtz
Hey hurtz,
I may be in good shape. I see that one of the default devices in RMAA is the M-Box 2. I got a great deal on eBay for one of those several months ago. I don't know if the signal is as clean as the best xonar cards, but it puts me in the ballpark. I'll start setting up a test bed sometime next week.
I may be in good shape. I see that one of the default devices in RMAA is the M-Box 2. I got a great deal on eBay for one of those several months ago. I don't know if the signal is as clean as the best xonar cards, but it puts me in the ballpark. I'll start setting up a test bed sometime next week.
Sadly, I'm going to have to push it off till Monday. I installed some Caddock resistors in a few positions a couple days ago and also want to change out two caps - C7 & C30 - to some top choice samples from Dario. It's also suggested to use an IsoTek burn-in process to get each cap (actually the entire system at each incremental change) close to optimum performance quickly and also determine which orientation sounds best. I decided it was worth the delay to get these last few adjustments completed.
Trust me, I'm as curious about the results/comparisons as anyone.
Trust me, I'm as curious about the results/comparisons as anyone.
Last edited:
That sounds good! Make sure to post the resulting .sav of your loop-back test
cheers!
hurtz
I think the MyRef can beat good Class A designs. I have compared against at least one, I certainly prefer the MyRef. General LM3886 design similar to the application note can be very good too. In my case, I have one of each that are under competition against each other. While taking turns to mod one to beat the other, I don't think I have a conclusion yet, but the MyRef may eventually win. The key issue how you get the power supply to work optimally with the circuit, which is not a small task.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.