I think this is quite an interesting update. How does it sound compared with the earlier versions?
Hi Tom,
Are you going to do a major design update to the rest of the Modulus family, namely the Modulus-286?
Are you going to do a major design update to the rest of the Modulus family, namely the Modulus-286?
Last edited:
Jesus Tom, that imd is spectacular. We'll done.
I was thinking that too. Looks like the AP hit the end stops at lower frequencies.
Thank you. Yeah. The IMD is basically the noise floor of the system as far as I can tell.Jesus Tom, that imd is spectacular. We'll done.
Hi Tom,
Are you going to do a major design update to the rest of the Modulus family, namely the Modulus-286?
The only planned changes to the Modulus family is that the Modulus-186 will likely retire once I've sold out. I have 45 of them left.
I have no planned changes to the Modulus-286 or Modulus-686 other than remeasuring their performance with the new AP.
Tom
Nice one Tom
But where to buy the LM3886.............
There doesn't seem to be alotta stock available out there.
t
But where to buy the LM3886.............
There doesn't seem to be alotta stock available out there.
t
They have a few less now. Now have enough for the Mod686. 😉
Hi Tom. Do you have any explanation of the negative rail oscillation when clip the signal at composite amp?
Can you fix this problem with any procedure or just reduce the issue to an acceptable level?
Can you fix this problem with any procedure or just reduce the issue to an acceptable level?
Last edited by a moderator:
The LM3886 itself and any composite amplifier with the LM3886 will have some amount of rail sticking and occasionally a short-lived parasitic oscillation where it exits clipping. The clipping behaviour of the Modulus-86 is actually very clean for a composite amp. That's not just my opinion. Others have verified this and commented in this thread and/or the Modulus-86 Build Thread.
I've made the Modulus-86 as well-behaved as I can. It's stable even with 2.2 uF in parallel with 8 Ω and marginally stable with 4.7 uF || 8 Ω load.
Tom
I've made the Modulus-86 as well-behaved as I can. It's stable even with 2.2 uF in parallel with 8 Ω and marginally stable with 4.7 uF || 8 Ω load.
Tom
This is a typical example of the large-signal transient response of the Modulus-86 Rev. 3.0. 10 kHz, square wave input. 8 Ω load. No signs of oscillation or instability here.
Tom
Tom
Thank you the answer.This is a typical example of the large-signal transient response of the Modulus-86 Rev. 3.0. 10 kHz, square wave input. 8 Ω load. No signs of oscillation or instability here.
View attachment 1025259
Tom
Ok, but the square wave not equal the clipping behaviour. Can you show me the modulus clipping level on a scope?
I don't have the plot handy for Rev. 3.0, but here's Rev. 2.1:

Note that the rail buzz is very low in amplitude, short-lived, and only show up with hard clipping. I'm hoping my builders have the savvy to turn the music down once it sounds distorted.
I would obviously rather be without the rail buzz and I spent considerable design effort reducing it. I'm trading off performance below clipping for performance during clipping. I hope we can agree that it's desirable to have an amp that has good performance below clipping.
Unfortunately I don't have access to the internal nodes in the LM3886 so I can't turn its protection circuits off. I suspect the SPiKe protection is what's causing the rail buzz. Either that or saturation in the VAS. I don't have access to that node either. 🙂
Tom

Note that the rail buzz is very low in amplitude, short-lived, and only show up with hard clipping. I'm hoping my builders have the savvy to turn the music down once it sounds distorted.
I would obviously rather be without the rail buzz and I spent considerable design effort reducing it. I'm trading off performance below clipping for performance during clipping. I hope we can agree that it's desirable to have an amp that has good performance below clipping.
Unfortunately I don't have access to the internal nodes in the LM3886 so I can't turn its protection circuits off. I suspect the SPiKe protection is what's causing the rail buzz. Either that or saturation in the VAS. I don't have access to that node either. 🙂
Tom
And now the Modulus-86 build video:
In this video I show you how to build a full stereo amp using the Neurochrome Modulus-86, Power-86, Guardian-86, and Intelligent Soft Start along with a 2x22 VAC, 200 VA power transformer and a Neurochrome Modulus Chassis.
You can find the build videos for the Modulus-86 and Power-86 here:
In this video I show you how to build a full stereo amp using the Neurochrome Modulus-86, Power-86, Guardian-86, and Intelligent Soft Start along with a 2x22 VAC, 200 VA power transformer and a Neurochrome Modulus Chassis.
You can find the build videos for the Modulus-86 and Power-86 here:
- Modulus-86:
- Power-86:
- Home
- Vendor's Bazaar
- Modulus-86: Composite amplifier achieving <0.0004 % THD+N.