Modulus-86 build thread

Jump to Latest
#5,821
potstip said:
Hello Tom ;

I am using Toroidal transformer + Power - 86 for my mod-86 amplifier atm. I am planing to replaced Power - 86 with APEX regulated psu. (Pls see attached file)

I guess psu noise will be lower. Can i expect to get better THD+N?
Click to expand...

I quote, from one of Tom’s original Modulus 86 design manuals:

The power supply rejection ratio of the Modulus-86 amplifier is truly excellent. Hence, the Modulus-86 performs as well on an unregulated supply as it does on a regulated supply. This is one of the benefits of the composite amplifier topology. The only advantage of using a regulated supply with the Modulus-86 is that the peak output power will be a few watt higher, as the regulated supply will experience less voltage droop as the load current is increased.
Click to expand...

This is DIY, and of course you are allowed to experiment. But given the design topology of the Modulus series of amplifiers, I would spend time streaming more music and understanding that music has a crest factor and we as humans don’t typically listen to sine waves ;-)

Best,
Anand.
 
#5,822
Thanks Anand for fast response;

I read that part before. As you know power-86 is only bulk C capacitors but also in the power 686 section of the manual mentioned as here below;

"Option for C-only and CRC snubbers included for the advanced builders." I want to be advanced 🙂 I can use CRC or CRC+ CAP MULT or regulator etc.. If it make sense for sure in the means of distortion performance.

I am asking this to Tom because i do not have the measurement equipments to evaluate. Real world performances of these variations.

Things can change by the time;

In the same section;
He mentioned about RC snubber in power 686
"RC snubber to eliminate coupling of diode switching noise into sensitive circuits."

But in contrast;
https://neurochrome.com/pages/rectification-snubbers#Snubbers
"As expected, the snubbers made no difference in the output ripple and noise of the power supply. Thus, it is rather irrational to expect it to change the output of the amplifier through the power supply"

It seems snubbers are not effects amplifier as more sensitive DAC circuits.
So i wish we can have a chance to see the different power supply performance as we saw for the sunubbers.
 
#5,823
The Modulus amps are power supply agnostic. They really don't care if the supply is regulated or not. What you will get from the APEX supply when used with the Modulus amps is:
  • Increased heat dissipation
  • Increased cost
  • Increased circuit complexity -> lower reliability
  • Lower max output power (due to the drop-out voltage of the regulator)
The THD+N or output noise of the amp won't budge.

If you're fine with these drawbacks there's no harm in trying.

Note, though, that LM3886 amps without the Modulus error correction would likely benefit from a regulated supply. This includes the LM3886DR.

Tom
 
  • Like
Reactions: invaderzim, radiotron and timgrindell
#5,824
Hello Tom;

Mostly on discrate design i used to connect speaker ground to end of psu pcb zero volt point. Thats why i use my modulus in this way and its works like a charm at the end all goes to chasis star point but at the same time i wonder if "gnd spk out" on modulus pcb has less noisier than my current setup. Or you just put it for easy installation?


Thanks in advance....
 
#5,827
tomchr said:
This would likely help quite a few people:

It's just a "short" (11 minutes) video on basic amplifier testing using a newly built Modulus-86 as an example.

Tom
Click to expand...
That's great. Such a helpful and informative video - explained to me why my multimeter AC measurement are so off the wall and also shows how to use the multimeter without the need for the bench AC voltmeter for just some quick and cheerful measurements. Just out of interest @tomchr - that DC offset is really low. Does that amp have DC servo or some such?
 
#5,830
@tomchr would you care to talk some more about your experience with vertical biamping and this feature in Neurochrome builds you cover in your videos?

I currently bi-wire, which q-acoustics present some backing evidence for here. Not sure if there’s any advantage in my case though as I only bi-wire from the speaker terminals on the back of the chassis, I only use single cable for + and - from the amplifier outputs to the terminals on the back of the chassis. So the benefits of bi-wiring might be lost in my configuration?

Moving on from the topic of bi-wiring, what benefits are achieved through vertical biamping with Neurochrome amplifiers? Or is it simply a better solution (or just different outcome) to purchase a higher end model, e.g 686 which will have a better ability to drive complex loads and demanding loads (e.g pair of drivers through a passive crossover).

Thanks!
 
#5,831
Just to clarify: Bi-wiring is when you connect one amplifier output with two cables in parallel to the tweeter and woofer terminals on the speaker. Bi-amping is when you use two separate amplifier channels - one for the woofer and another for the tweeter.

The advantage of bi-amping is that you can use an active crossover operating at line level. That makes it possible to use active filters, which can have steeper filter slopes than the passive crossovers found in speakers. Also, the distortion in a line-level filter can be much smaller than that of a passive filter operating at the output voltage of the amplifier.

Tom
 
  • Like
Reactions: troystg
#5,832
Ok, thanks @tomchr . I understand the design intent now.

Are you able to comment on bi-wiring if you only bi-wire from the chassis and not all of the way from the Modulus amplifier outputs?

Any benefit to vertical biamp with passive crossover with amplifiers like Modulus, Hypex and Purifi which are very ‘well behaved’ in their performance? Or is your suggestion for vertical biamp only to support active crossovers (for which I clearly understand the benefits).
 
#5,833
I'm not sure what you're asking. I don't see much of a point of bi-wiring in general. If there is any benefit to bi-wiring what will matter (most) is the longer cable run to the speakers and not so much the 15 cm of wire inside the chassis.

If you're going to 'bi' anything, I'd go for bi-amp.

The point of bi-amping in my opinion is to allow you to use active (or DSP-based) crossovers. I don't see much of a point of bi-amping if you're using the passive XO in the speaker.

With a well-behaved amp (i.e., one with a low output impedance) I'm also not convinced that you'll gain much (if any) benefit from bi-amping - and certainly not from bi-wiring.

Tom
 
#5,835
There is also the type of 'bi-amping' that is similar to 'bi-wiring' except you have two separate power amp channels feeding the two speaker cables. Each power amp channel is still fed the full-range signal, but drives only the 'woofer' or 'tweeter' speaker terminals on the speaker. The idea is that each amp only needs to supply current over part of the frequency range. to me it seems like a way to sell both twice the number of speaker cables AND twice as many power amps.
 
#5,837
Bi-Wiring IS an improvement but has many variables for that improvement.
Definitely active XO and bi-amping is the way to go, however if funds are short bi-wiring is a modest, low('ish) cost improvement.

Mostly depends on the amplifiers handling of the reactive components of the passive XO. Putting the common tie point at the amp terminals does help "slightly". Both amp damping and the added resistance between XO branches does have an effect. Not nearly as much as bi-amping though, which is a drastic increase.
 
#5,838
potstip said:
In the same section;
He mentioned about RC snubber in power 686
"RC snubber to eliminate coupling of diode switching noise into sensitive circuits."

But in contrast;
https://neurochrome.com/pages/rectification-snubbers#Snubbers
"As expected, the snubbers made no difference in the output ripple and noise of the power supply. Thus, it is rather irrational to expect it to change the output of the amplifier through the power supply"

It seems snubbers are not effects amplifier as more sensitive DAC circuits.
So i wish we can have a chance to see the different power supply performance as we saw for the sunubbers.
Click to expand...
Circling back to this. I think you're conflating two issues here:

1) Snubbers eliminate ringing in the secondary of the power transformer. Snubbers make no difference in the output voltage of the power supply as this ringing occurs when the rectifier diodes are off. Therefore, any effect of the snubbers cannot be from conduction of noise through the power supply. It's impossible to conduct current through a diode when it's not conducting.

2) Switching noise can be radiated, however. It can then couple inductively or capacitively. Inductive coupling can be minimized by minimizing the loop area of the "transmit" coil and of the "receive" coil. The "transmit" coil in this case would be the secondary wiring of the transformer. Minimize the loop area there. That's why I recommend that you zip tie those together in a tight bundle. Same with the primary wiring. The "receive" coil would be the PCB layout of the amplifier. If you use my boards, I took care of that for you. If the coupling is capacitive a grounded shield can help. This can be accomplished on the PCB layout as well. However, as either Ott or Morrison note in their book(s) on grounding, "separation beats shielding on cost any day of the week". You could simply place the boards such that the input of the is away from the rectifier. You'll notice I do that in my Modulus-86 and Modulus-286 build videos.

On the Modulus-86 product page you'll find a measurement of the amp's noise floor with its input shorted. It's powered by a Power-86 (no snubber) and a toroidal power transformer. You'll notice that the mains hum is barely peeking up above the noise floor. This is with a circuit board that sits 'naked' on the lab bench. You can play with snubbers until the cows come home. It won't do a lick of difference. Not with the Modulus-86 anyway.

The only reason I offer options for snubbers on the Power-86 and Power-686 is that some perceive a snubber-less supply (or a supply with a simple C-only snubber) as inferior even though the snubber makes no difference in the output of the amp. Market wants snubbers so I deliver snubbers. Not every decision is rooted in science and engineering ... or reality. 🙂

Tom
 
#5,839
m10chalk said:
That's great. Such a helpful and informative video - explained to me why my multimeter AC measurement are so off the wall and also shows how to use the multimeter without the need for the bench AC voltmeter for just some quick and cheerful measurements. Just out of interest @tomchr - that DC offset is really low. Does that amp have DC servo or some such?
Click to expand...
Tom: This question has probably been asked before, will you setup and deliver this Modulus-86 Audio Amplifier as a complete kit? and may be also with Modulus-286 inside as an option?
 
#5,840
I have no plans to put together a kit. I don't enjoy spending my time putting itty-bitty parts into itty-bitty bags. Mouser and Digikey do that much more efficiently and affordably than I'd be able to.

I don't understand your last question. I posted a Modulus-286 build video in early/mid May. Just search for Modulus-286 on YouTube and you'll find it.

Tom
 
You must log in or register to reply here.
Top Bottom