Actually, it's in post 2561 here or a ways back in the other Mod thread:
It also happens a number of the changes in the Mod 2.0 were addressing deviations from the original design I'd suggested against before they got laid out in 1.0.
Tom's rightly quick to point out he has a ton of validation and characterization invested (among other things I ain't got no APx-525) but, if one's interested in historical trivia, the most interesting topology difference between the original and commercial layouts is (IMO) a single compensation capacitor due to the two designs hitting slightly different points on the efficient frontier. As this thread attests, the handoff's been beneficial to the DIY community. Which is good as that was the point since I didn't have the time to make it happen on my own.To be clear in one sense, most of what's novel about the Mod is the application of amplifier improvements developed for class D to class AB. Much of the class D side of that appeared in Eva's posts here on DIY Audio before 2007 and more visibly, in a commercial sense, with Bruno Putzey shipping UcD modules through Hypex. A lot of it can also be found in IRF application notes dating back to at least 2007 and in content posted to DIY Audio from circa 2010 by myself and a couple others. From an amp design perspective it's probably most accurate to think of the Mod as a simplification from class D enabled by the 3886 being a power stage more easy to interface to than what's available in switchmode. Whether the associated decrease in efficiency is worthwhile is a complicated question that's a bit OT here.
To be clear in another sense, I've only vaguely tangential involvement with the HP-1 as the IP embodied in the Mod is flexible enough to apply. Pretty much the same for supplies and tubes. For example, my "contribution" to the SMPS-86 was tossing Tom a heads up when Meanwell launched the series used with the SMPS-86 board a year or so prior.
The composite amplifier topology has been around since the 1960ies, if not earlier. You're not old enough to claim credit for that.
Your circuit showed a negative phase margin in simulation. I.e. if built, it would have oscillated. Guaranteed! You also failed to deal with the various quirks of the LM3886, which I dealt with in Rev. 2.0. I was more inspired by Scott Wurcer's circuit than by yours so if anybody deserves credit it would be him and his employer at the time. The reference to his original article is on my website.
I recall spending quite a few nights and long weekends in the lab validating the grounding schemes as well. These schemes were variations on what AndrewT has been pushing here for years. Who should get the credit? You? Me? Andrew?
"Only one capacitor different". Please! I'm sure Hyundai and BMW are only different on the hubcaps too.
Now you want to claim credit for the HP-1?! Get over yourself! You have no knowledge of the circuit within the HP-1 as I purposely didn't share that information with you. All you have is the list of chips used: OPA1611, OPA1612, LME49600. Getting the OPA1611 to play nicely with the LME49600 involves some creativity, knowledge, and work. Good luck with that in the simulator! I've used the LME49600 since 2009.
You have at NO point provided ANY financial contribution or investment in Neurochrome. You have taken no risk.
I am the one who has accepted all the risk and paid all the bills - including the rather large one for the APx525.
I'm not sure what your intent was with the post above. Perhaps you would elaborate on that?
If you are looking to establish a reputation as being a competent circuit designer, I suggest participating regularly on various forum threads. If you consistently provide good advice in a language that people, including the beginners, can understand, your skills will speak for themselves.
Tom
Your circuit showed a negative phase margin in simulation. I.e. if built, it would have oscillated. Guaranteed! You also failed to deal with the various quirks of the LM3886, which I dealt with in Rev. 2.0. I was more inspired by Scott Wurcer's circuit than by yours so if anybody deserves credit it would be him and his employer at the time. The reference to his original article is on my website.
I recall spending quite a few nights and long weekends in the lab validating the grounding schemes as well. These schemes were variations on what AndrewT has been pushing here for years. Who should get the credit? You? Me? Andrew?
"Only one capacitor different". Please! I'm sure Hyundai and BMW are only different on the hubcaps too.
Now you want to claim credit for the HP-1?! Get over yourself! You have no knowledge of the circuit within the HP-1 as I purposely didn't share that information with you. All you have is the list of chips used: OPA1611, OPA1612, LME49600. Getting the OPA1611 to play nicely with the LME49600 involves some creativity, knowledge, and work. Good luck with that in the simulator! I've used the LME49600 since 2009.
You have at NO point provided ANY financial contribution or investment in Neurochrome. You have taken no risk.
I am the one who has accepted all the risk and paid all the bills - including the rather large one for the APx525.
I'm not sure what your intent was with the post above. Perhaps you would elaborate on that?
If you are looking to establish a reputation as being a competent circuit designer, I suggest participating regularly on various forum threads. If you consistently provide good advice in a language that people, including the beginners, can understand, your skills will speak for themselves.
Tom
I finally got my Modulus-86 pair up and running. I'm driving Klipschorns with the amp (around 105dB/1W/1M efficiency), and with the inputs open I can *just barely* hear hiss with my ear up to the tweeter. No noise whatsoever at normal listening distance. I realize the proper way is probably to shunt the inputs, but open is how I did it for a test. To me, that's just incredible.
I've been listening for about 30 minutes per day for a week or so and am sold. This amp has some serious punch on high-dynamic-range audio. Sounds fantastic on orchestral pieces, prog-rock, you name it.
I'm very pleased with the design, the great instructions, everything. Thanks Tom, two thumbs up!
I've been listening for about 30 minutes per day for a week or so and am sold. This amp has some serious punch on high-dynamic-range audio. Sounds fantastic on orchestral pieces, prog-rock, you name it.
I'm very pleased with the design, the great instructions, everything. Thanks Tom, two thumbs up!
Tom,
Your post is an excellent, well thought, out response. You freely give credit where deserved. Your honesty and skill should be strongly recognized and appreciated.
FWIW, I have no connection with Tom or Neurochrome other than being a very pleased owner of a Modulus86, and can't wait to finish the Parallel86. That will complete my systems upgrades and I can't imagine buying or wanting another amp. I'm 70 years old, this should be it.
I've mentioned that I've listened to very many Hi-End amps at some of the top line retailers like Sound By Singer, including a few chip and class D types at a few other "respected" retailers.
In fact one of my amps was rated class A by Stereophile (big deal) and another may have also been rated the same. After listening to your Mod86 for almost a year, the other two amps are sitting in storage and are for sale. The Mod86 blew them away and I expect the Parallel86 will do the same.
henrylrjr
Your post is an excellent, well thought, out response. You freely give credit where deserved. Your honesty and skill should be strongly recognized and appreciated.
FWIW, I have no connection with Tom or Neurochrome other than being a very pleased owner of a Modulus86, and can't wait to finish the Parallel86. That will complete my systems upgrades and I can't imagine buying or wanting another amp. I'm 70 years old, this should be it.
I've mentioned that I've listened to very many Hi-End amps at some of the top line retailers like Sound By Singer, including a few chip and class D types at a few other "respected" retailers.
In fact one of my amps was rated class A by Stereophile (big deal) and another may have also been rated the same. After listening to your Mod86 for almost a year, the other two amps are sitting in storage and are for sale. The Mod86 blew them away and I expect the Parallel86 will do the same.
henrylrjr
I just assembled Modulus 86. Previously I assembled lm3886dr as well. Before that I used Naim 5si. LM3886DR was worse than Naim in terms of accuracy, details, however better in therms of "naturality" - more pleasant to listen to. Modulus 86 beats Naim in those two aspects. It is not night and day difference, but Modulus 86 is bit more accurate and definitely more natural than Naim.
Thanks Tom! Fantastic work and great design.
Thanks Tom! Fantastic work and great design.
My Aragon 8008BB, bought new in 2001, had been my favorite for 15 years until I bought a second hand McCormack Deluxe Edition DNA 0.5. For me It surpasses the Aragon in speed, timbral accuracy and just plain listening pleasure. The speed relates to the sounds of percussion, cymbals, piano and guitars sounding more natural. I believe the McCormack is an exceptional amplifier. However, the modulus86 brought those characteristics to a more satisfying level for me.
I'm not going back. That's why I bought the parallel86.
henrylrjr
I'm not going back. That's why I bought the parallel86.
henrylrjr
You're quite welcome. Thank you for your feedback. It's good to hear that the MOD86 works well with high-efficiency speakers. Given the 33 µV RMS (A weighted) output noise, I'd certainly expected it to work well but it's always nice with an actual data point from the real world.
You're welcome. I'm glad you like it.
The LM3886DR is a good implementation of the LM3886 and will deliver the data sheet performance of the LM3886. It doesn't apply any special tricks but is a good all-round LM3886 amp. I'm not surprised a good amp like the Naim beats it. The Modulus-86, on the other hand, uses a precision opamp to provide error correction on an LM3886 output stage, thus achieves much better performance. The only commercial amp I've come across (thanks Anand) that approaches the performance of the Modulus-86 is the Benchmark ABH-2, which will set you back about 10x the cost of a nice Modulus-86 build.
Tom
Thanks Henry. I really appreciate it.
It's good to hear that the Modulus-86 is still going strong and that you continue to enjoy it. Thanks for sharing.
Tom
Does anyone know of any downsides to positioning an Antek AS-3224 on edge such that the mounting bolt is horizontal? I would use an L bracket to secure it.
I'd like the smallest enclosure in width and depth possible. That got me to thinking a smaller foot print, with deeper internal depth might work. It would allow, 6.5" long (tall) heatsinks, and plenty of room to for the 5" Dia. transformer on edge.
I'm thinking of getting a Modushop case with 165mm (6.496") internal depth. I think everything would easily fit in an 11"W x 8"D x 6.5" internal depth case. My only concern is magnetic field from the trans. Is the Antek also shielded on the top and bottom surfaces?
Thanks,
henrylrjr
I'd like the smallest enclosure in width and depth possible. That got me to thinking a smaller foot print, with deeper internal depth might work. It would allow, 6.5" long (tall) heatsinks, and plenty of room to for the 5" Dia. transformer on edge.
I'm thinking of getting a Modushop case with 165mm (6.496") internal depth. I think everything would easily fit in an 11"W x 8"D x 6.5" internal depth case. My only concern is magnetic field from the trans. Is the Antek also shielded on the top and bottom surfaces?
Thanks,
henrylrjr
A toroid doesn't have a large leakage field, so I'm not too worried. Looking at this article, I'm not sure orientation matters greatly as long as you don't put the sensitive circuit right in the doughnut hole of the toroid.
I friend of mine has his amps set up with a large 1 kW Plitron toroid on the front panel, mounted vertically, and the amp on the rear panel. Granted the chassis is probably 35-40 cm deep, so there's good separation, but still. The doughnut hole is aimed right at the amp PCB. I didn't measure anything on that amp which would indicate this to be an issue. His amps are a 150 W discrete design by Randy Sloane.
Tom
I friend of mine has his amps set up with a large 1 kW Plitron toroid on the front panel, mounted vertically, and the amp on the rear panel. Granted the chassis is probably 35-40 cm deep, so there's good separation, but still. The doughnut hole is aimed right at the amp PCB. I didn't measure anything on that amp which would indicate this to be an issue. His amps are a 150 W discrete design by Randy Sloane.
Tom
If you afraid of magnetic radiation try Audio Supreme v2 transformer from toroidy.pl They are encapsulated into nice silver look metal case. Such tranny could be actually put on top of the case like in tube amps.
+1. That's what I use, but admittedly I did it for looks and a clean appearance. They are steel covers.
Best,
Anand.
I have no plans for one. For me to start such a project there needs to be a good business reason. I have yet to find one.
I have enough projects to cover the next few years. Even if I do decide to design a current output amp, the business reason(s) need(s) to be really, really, REALLY good for me to suspend my current projects.
Tom
No comment.
Okay. One comment: Higher output power. The Modulus-286 is known to the public already.
I hope to have the MOD286 out by Christmas.
Tom
Okay. One comment: Higher output power. The Modulus-286 is known to the public already.
I hope to have the MOD286 out by Christmas.
Tom
No comment.
Okay. One comment: Higher output power. The Modulus-286 is known to the public already.
I hope to have the MOD286 out by Christmas.
Tom
Nice amp
Tom, sonce you already made nice headphone amplifier, simple nice sounding dac/preamp/headphone amp would be also good idea.