Hi all, a while ago I built a stereo amplifier with 2 Modulus-86 boars that I've been using for a few months. They both work well, or at least that’s what I thought… After upgrading the heatsink I found out that one of the units has a flaw that I can't exactly explain.
It works nominally up until 10W RMS (measured output on a 8ohm resistor load) , however if push it more than that, or simply if I plug/unplug the audio input (using unbalanced RCA) connector, which typically generate a loud 60Hz hum, the chip suddenly start draining a ton of current (and generate lots of heat!), power consumption (measured at the PSU) jumps from 30W to 100W and I saw some very high frequencies (> 50kHz on my measurement bench)
I've tried swapping all 3 OpAmp/THAT chips with the working board but the behavior stays the same.
Any ideas ?
It works nominally up until 10W RMS (measured output on a 8ohm resistor load) , however if push it more than that, or simply if I plug/unplug the audio input (using unbalanced RCA) connector, which typically generate a loud 60Hz hum, the chip suddenly start draining a ton of current (and generate lots of heat!), power consumption (measured at the PSU) jumps from 30W to 100W and I saw some very high frequencies (> 50kHz on my measurement bench)
I've tried swapping all 3 OpAmp/THAT chips with the working board but the behavior stays the same.
Any ideas ?
That's odd. My Modulus-86 is dead quiet even if I leave the differential input disconnected. If I use an XLR-RCA adapter to convert the input to single-ended, I do get a faint mains hum. Obviously, if I touch the input with a finger, I get tons of hum. That's normal.
While plugging/unplugging the amp while it's on is not 'nice', it should not result in oscillation or any kind of lock-up/latchup or excessive power draw.
I would check the compensation caps: C6, C8, C16, and C18. Make sure you have the correct values installed. Check the markings on the caps. It does occasionally happen that Mouser ships the wrong parts. It's very rare, but I've had it happen more than once.
The fact that one of two boards misbehave while the other is fine indicates to me that you're likely dealing with an assembly error, probably around the compensation caps. Either a stuffing error (wrong part populated) or a cold solder joint.
Toss me an email. I'm happy to help.
Tom
While plugging/unplugging the amp while it's on is not 'nice', it should not result in oscillation or any kind of lock-up/latchup or excessive power draw.
I would check the compensation caps: C6, C8, C16, and C18. Make sure you have the correct values installed. Check the markings on the caps. It does occasionally happen that Mouser ships the wrong parts. It's very rare, but I've had it happen more than once.
The fact that one of two boards misbehave while the other is fine indicates to me that you're likely dealing with an assembly error, probably around the compensation caps. Either a stuffing error (wrong part populated) or a cold solder joint.
Toss me an email. I'm happy to help.
Tom
Last edited:
Thanks! I will double check everything tomorrow with lots of daylight and keep you posted!
oh... I'm in trouble of my build. I hear some waive noise from my speakers. It is negligible but in anyway I'm able to hear it if no any music playing and my ear is close to the speaker.
The pre-story... Unfortunately, during the build I forgot to connect the ground (from mods-86 power input) to a chassis, and after I plug pseudo-diff xlr cable one track from R3 to R11 has disappeared, after some manipulations (my attempt to find what is wrong) the same unhappy destiny happens for the track J2(1) to C11(2). I've restored them, everything working well, except of that noise... As I remember I didn't hear it when I power amps first time (without XLR signal input). May be that problem is not related to my "pre-story" )). Any advice, please.
The build set-up.
Power supply: AS-2222 -> Power-686 -> Mods-86
Signal: RCA -> UniBuffer -> XLR
Thanks
The pre-story... Unfortunately, during the build I forgot to connect the ground (from mods-86 power input) to a chassis, and after I plug pseudo-diff xlr cable one track from R3 to R11 has disappeared, after some manipulations (my attempt to find what is wrong) the same unhappy destiny happens for the track J2(1) to C11(2). I've restored them, everything working well, except of that noise... As I remember I didn't hear it when I power amps first time (without XLR signal input). May be that problem is not related to my "pre-story" )). Any advice, please.
The build set-up.
Power supply: AS-2222 -> Power-686 -> Mods-86
Signal: RCA -> UniBuffer -> XLR
Thanks
Sounds like you have a severe grounding issue in your setup. Have you checked your outlets with an outlet tester (see attached)?
The traces are pretty wide. It would take a few tens of ampere (if not more!) to cause them to vaporize. This points to a serious wiring error and/or some "serious negotiations" between pieces of equipment of exactly what constitutes "ground".
R3 does not connect to R11, so if you have them connected that way, that could be part of the problem. The ground trace goes from J2-1 to C11 to R11.
If correcting the wiring issue doesn't fix the noise, I'd replace the THAT1200.
Tom
Attachments
I was lucky to buy one of the 286 kits and am over the moon with it.
I have also made an 86 based amplifier as mentioned earlier.
I have inserted both amplifiers into very good systems using valve monoblock amplifiers. In both cases the image width shrank, sort of to between the speakers rather than room filling width.
I was puzzled by this technically as the modulus boards have such good PSRR.
To investigate I converted the 86 amplifier into a true dual mono build and hey presto the image width returned.
86 internals:
As the result was so encouraging I decided to convert the 286 to dual mono as well. I bought a second power supply board to install, however once I had the top off the 286 and measured properly I realised there was a problem.
The internal depth is 200mm, the power supply boards are 100x100mm.
There needs to be a 5mm gap all around the pcb for safety, looks like I need to find another 15mm or so, or convert the case into a Tardis ???
A bit of thinking and increasing the depth of the case seemed the best approach. I have spent all morning cutting and bending an aluminium extrusion, then having to make cut outs where the major dimension fouled the sockets and fixing screws, so quite fiddly.
After making a new plate to hold the two pcbs the amplifier looks like this:
The extension frame was covered in some 'carbon fibre' sticky back plastic to look the part.
The result is the only dual mono Neurochrome 286 amp in existence (unless you know better). The performance now has taken on a holographic presentation, most certainly worth the expense and effort to achieve.
I have also made an 86 based amplifier as mentioned earlier.
I have inserted both amplifiers into very good systems using valve monoblock amplifiers. In both cases the image width shrank, sort of to between the speakers rather than room filling width.
I was puzzled by this technically as the modulus boards have such good PSRR.
To investigate I converted the 86 amplifier into a true dual mono build and hey presto the image width returned.
86 internals:
As the result was so encouraging I decided to convert the 286 to dual mono as well. I bought a second power supply board to install, however once I had the top off the 286 and measured properly I realised there was a problem.
The internal depth is 200mm, the power supply boards are 100x100mm.
There needs to be a 5mm gap all around the pcb for safety, looks like I need to find another 15mm or so, or convert the case into a Tardis ???
A bit of thinking and increasing the depth of the case seemed the best approach. I have spent all morning cutting and bending an aluminium extrusion, then having to make cut outs where the major dimension fouled the sockets and fixing screws, so quite fiddly.
After making a new plate to hold the two pcbs the amplifier looks like this:
The extension frame was covered in some 'carbon fibre' sticky back plastic to look the part.
The result is the only dual mono Neurochrome 286 amp in existence (unless you know better). The performance now has taken on a holographic presentation, most certainly worth the expense and effort to achieve.
I'm certainly not aware of anyone crazy enough (or with enough dexterity) to attempt a dual mono build of the MOD286 Kit LE.
I love that carbon fibre chassis extension. Nice work! Interesting with the image width. A few days ago I received a 4-conductor, 50' SpeakON cable (GLS brand via Amazon). My plan is to cut it in half and use it to bi-amp my KEF R700 using the Modulus-686. If I experience any changes in imaging, I'll let you know.
Tom
Hi, I'd like to ask for some opinions.
I'm listening in a near-field setup on a desktop with speakers 1 meter away from me. The speaker is a Harbeth P3ESR @ 83 dB/W.
I've done the diyaudio / Archimago power test (Archimago's Musings: MUSINGS: How much amplifier power do you really "need"? Why not test for yourself?) and at quite unpleasant levels (with earplugs) I recorded 4.5 Volts at speaker posts, which should be 25 Watt at 8 Ohm if I count correctly.
Most importantly, I usually listen at 60 dB, definitely not anything like 85 dB commonly heard in audio circles.
Currently I am listening to a ICEPower 125ASX2 module, which I'm not entirely satisifed with, so I'm thinking maybe I'd be happier with something from the Neurochrome Modulus line, say Modulus 86 for a start. The alternative would be an NCore 125NC.
Do you think a Modulus 86 would handle my power needs well, without clipping? If yes, what power supply would you recommend for it, I'm thinking either a SMPS300RE or a Power 86 + transformers. Are there any audible difference between them? If there is nothing wrong with the SMPS300RE, I might pick it from Audiophonics which offer warranty on it. Any reason to go with the Power 86 instead?
I'm listening in a near-field setup on a desktop with speakers 1 meter away from me. The speaker is a Harbeth P3ESR @ 83 dB/W.
I've done the diyaudio / Archimago power test (Archimago's Musings: MUSINGS: How much amplifier power do you really "need"? Why not test for yourself?) and at quite unpleasant levels (with earplugs) I recorded 4.5 Volts at speaker posts, which should be 25 Watt at 8 Ohm if I count correctly.
Most importantly, I usually listen at 60 dB, definitely not anything like 85 dB commonly heard in audio circles.
Currently I am listening to a ICEPower 125ASX2 module, which I'm not entirely satisifed with, so I'm thinking maybe I'd be happier with something from the Neurochrome Modulus line, say Modulus 86 for a start. The alternative would be an NCore 125NC.
Do you think a Modulus 86 would handle my power needs well, without clipping? If yes, what power supply would you recommend for it, I'm thinking either a SMPS300RE or a Power 86 + transformers. Are there any audible difference between them? If there is nothing wrong with the SMPS300RE, I might pick it from Audiophonics which offer warranty on it. Any reason to go with the Power 86 instead?
4.5V at 8 ohms is 2.5W, not 25W. You should be totally fine with the output of the LM3886. It sounds to me like you're not listening to heavy metal in an auditorium. As for the other question, I can't really answer that, will have to defer to someone who tried both.
This blog post definitely tells me that I should be squaring 4.5 * 4.5 = 20. Why do you recommend 2.5?
Archimago's Musings: MUSINGS: How much amplifier power do you really "need"? Why not test for yourself?
Yes, I listen only to acoustic music, usually with only a few instruments at quiet levels.
Archimago's Musings: MUSINGS: How much amplifier power do you really "need"? Why not test for yourself?
Yes, I listen only to acoustic music, usually with only a few instruments at quiet levels.
I don't recommend anything, just by ohm's law, if you're reading 4.5V with an 8 ohm speaker, then P = (V^2)/R = (4.5x4.5)/8 = 2.53W
It has to be a particularity of the test tones:
Since the maximum RMS power for a sine wave is -3dB, and the test tone is -12dB, that means we need to gain +9dB to get the maximum amplitude. This represents a voltage ratio gain of 2.818383.
Therefore, to estimate peak power:
P = [(measured V at -12dB) x 2.818383]^2 / R
or
P = (measured V at -12dB)^2 x 7.94328 / 8-ohms impedance
In other words, estimated peak RMS wattage across an 8-ohm load = (measured V at -12dB)^2.
It sounds like they're trying to compensate for the crest factor of music. That's actually an interesting approach.
I seem to recall that the perception of loudness is related to the average (or RMS) SPL rather than the peak SPL. The trouble with music is that the peaks of the signal may be 6-20 dB higher than the average. This ratio of peak-to-average power is referred to as the crest factor. The industry appears to have settled on 10 dB as a "standard crest factor", though that may have more to do with power supply and heat sink design.
With digital sources, one could do the following:
- Crank the music loud
- Play a test tone at -10 dBFS (assuming 10 dB CF)
- Measure the resulting output power
- Multiply this number by 10 (or 10^(CF/10)) if you're using a different CF
- Design the amp to be able to provide this power
Tom
Did something similar but with a full swing test tone at regular listening, and from there in 20 steps to levels far beyond what I'd ever listen to. My conclusion was that a unity gain power amp should be enough for my daily listening. Rarely I'd need more than a watt for peak signals.
I'm confident you'll have plenty power with the mod-86. Don't think the power supply matters much. I have the audiophinics one. And I love this amp... . (Went from main nap 250dr to nc400 to mod-86 and I'm satisfied where I am...)
I'm confident you'll have plenty power with the mod-86. Don't think the power supply matters much. I have the audiophinics one. And I love this amp... . (Went from main nap 250dr to nc400 to mod-86 and I'm satisfied where I am...)
OK, I'm more and more confident that for my near field setup the Modulus 86 will have plenty of power. The only thing I'm totally lost about is the power supplies. The Power86 + transformers would be 2x as much as a new SMPS300RE for me, is there any difference between the two either audiably or in usage?
Does the amplifier make any pop/click noise at power on with either SMPS or Power 86?
Also, how is the 2.2 version of the Mod86 compared to 2.4? I could purchase a second hand 2.2 unit, were there any substantial improvements since then?
Does the amplifier make any pop/click noise at power on with either SMPS or Power 86?
Also, how is the 2.2 version of the Mod86 compared to 2.4? I could purchase a second hand 2.2 unit, were there any substantial improvements since then?