Class D

[Steve williams]

Thanks

I misunderstood

 

[treitz3]

You guys have got me curious. I just ordered the CD off Amazon. The beginning of that song [Duende] sounded like I was listening to "Friday Night in San Francisco", which is a great performance IMO. Thank you, gentlemen.

 

[amirm]

Music isn't clipped but levels are within a hair of max SPL given the "loudness wars." On some equipment it may clip but as rule, it should not do that.

 

[Orb]

Music isn't clipped but levels are within a hair of max SPL given the "loudness wars." On some equipment it may clip but as rule, it should not do that.

I am with Amir on this, also it depends upon the whole system in how it reproduces the digital digital square wave (meaning certain audio components may behave differently, such as DACs and not just at the amp).
A square wave signal does not mean the amp has clipped and does not mean it is going to affect speakers, what matters in those regards is the strength of that signal; so how much it is amplified and in doing so creates-demands excessive energy and distortion for amps and speaker drivers.
So playing a square wave signal-sound at low levels would not cause clipping from an amp perspective and also should not put the speaker driver under fail stress.
The only time I think an input square wave signal is dangerous at any signal level are for true digital PWM amps as this can cause instabality and such amps will have complex protection algorithms.

Cheers
Orb

 

[fas42]

Music isn't clipped but levels are within a hair of max SPL given the "loudness wars." On some equipment it may clip but as rule, it should not do that.

Specifically, on the Duende track, there is no obvious square wave, like on the famous Iggy Pop remastering, but the maximum level is hit over and over again, and you can see that the peak of the waveform at that point has been squashed by a certain amount. Specifically, in Audacity, you can flag, view the clipping points and the track turns into a sea of red.

Frank

 

[dallasjustice]

@Gary

Gary,
I am interested in the Ncore moduels. In fact, I ordered a pair last week. One thing I noticed is that your GR360 amp sports a very powerful linear power supply. I also noticed that Bruno basically says that there's no advantage to using any power supply other than his custom designed SMPS with the new Ncore. I don't know whether you've had an opportunity to work with the new Ncore module, but I would love to hear your thoughts in general about why a linear power supply could have more potential for great sound with switching amps than a high performance SMPS.

 

[garylkoh]

Gary,
I am interested in the Ncore moduels. In fact, I ordered a pair last week. One thing I noticed is that your GR360 amp sports a very powerful linear power supply. I also noticed that Bruno basically says that there's no advantage to using any power supply other than his custom designed SMPS with the new Ncore. I don't know whether you've had an opportunity to work with the new Ncore module, but I would love to hear your thoughts in general about why a linear power supply could have more potential for great sound with switching amps than a high performance SMPS.

If I'm selling a high power linear power supply, then I might also say that there is no advantage to using any power supply other than my custom designed linear power supply. There are certainly advantages to using a SMPS. For one, I wouldn't have to visit a masseur every couple of weeks because of a bad back due to lifting the 37lb power supply needed to power my GR360. I tried many a SMPS and did not like the sound. I have NOT tried Bruno's latest Ncore module with the latest SMPS. I did try the Hypex SMPS a long time ago with the UcD400 but preferred the linear power supply. Unfortunately, YMMV. I couldn't measure the difference in power bandwidth or distortion. I just listened longer and more enjoyably with the linear power supply.

With the Ncore modules - that's a whole development cycle that I don't have time at this point to get into. I modify the UcD400 for the GR360, and everytime they change the UcD, I have to go through the listening/development again.

 

[kevinh]

Ncore SMPS

I would note that the Ncore module and it's matching SMPS 600 is for the DIY version.

The OEM version is a different (higher powered) module the NC 1200. AFAIK Bruno
has not designed a SMPS for the OEM Ncore 1200.

 

[dallasjustice]

So, maybe Bruno just doesn't want the headache of DIY'ers attempting to build their own power supplies and all of the customer support needed to deal with it. I can totally understand why a well designed linear supply could sound better than a switcher, especially with switching amps.

Gary, maybe it wouldn't be so bad if you went to a masseuse instead of a masseur.

 

[amirm]

I have not dug into their designs lately. But at high level, switchmode amplifiers are highly dependent on the accuracy of their supply voltages because that is exactly what they output. A pulse goes out the amp that switches between the power supply voltages. The solution seems to be copious amount of negative feedback to compensate. So the argument becomes whether that is a good thing or not.

The monster Mark Levinson 53 is that big also because of its linear power supply (borrowed from Mark Levinson reference 53). At 135 pounds for one channel, it also is a back breaker. That is why I let my crew move it around.

 

[kevinh]

Designing a SMPS is harder than designing a Linear PS, same with a Calss D amp, board layout, rfi all sort of issues that designing say a tube amp that designers don't hace to deal with. The CAE tools required are beyond the means of many small audi manufacturers.

 

[garylkoh]

So, maybe Bruno just doesn't want the headache of DIY'ers attempting to build their own power supplies and all of the customer support needed to deal with it. I can totally understand why a well designed linear supply could sound better than a switcher, especially with switching amps.

Gary, maybe it wouldn't be so bad if you went to a masseuse instead of a masseur.

I like a good, strong massage, and the masseuse just can't generate the force. The place I go to is just a couple of doors away from Amir's store

Dealing with the DIY crowd is a headache. I'm sure you're right about the customer support needed to deal with all of them.

 

[Vincent Kars]

FYI: http://www.acousticimagery.com/page2.htm

 

[dallasjustice]

And now for something a little different. This is a build Ryan Mintz is doing for my Ncore modules. He's almost done with it. I haven't heard them. I know Ryan can build a pretty great linear psu for a switching device.

The power supply is a linear unregulated power supply. The images do not include additional AC filters, but they do have the Mundorf Silver/Gold/Oil capacitors, which are part of the AC filter. There are also some additional filters hidden underneath the PS boards.

The power supply for each channel is comprised of nearly 250 parts, 6 different types of capacitors, 11 different capacitor values, and 6 different types of filter. The AC filter is comprised of Mundorf Silver/Gold/Oil and a proprietary filter that he's working on. The grounding and filtering is proprietary and Ryan says he's working on the patents for it now. The goal was to create a very fast, very high bandwidth linear power supply that will run stably under any load. He says that he treats voltage ripple, current ripple, ultra high frequency noise (into the ghz range), heat, and 1/f noise in the audible range.

Playing music loudly the voltages stay at a stable +/- 66.063V without wavering in either direction.

The entire chassis is CNCed aluminum. The shelf where the amplifier resides doubles as a heatsink, which then heatsinks to the rest of the chassis. This reduces heat-born noise and improves stability. The aluminum also serves to shield the power supply from the amplifier module.

The amplifier uses XLR balanced inputs and the system is wired with 15awg vari-strand wire from Kimber. The module has some capacitor part swaps. Small parts were used for their speed and higher bandwidth.

I've been told that Bruno's SMPS can never be improved upon, although Bruno never made such a claim. To test this assertion, I also have his SMPS and I've asked Ryan to mount them next the module on each chasis. As you can see, there's plenty of room on the shelf to fit the SMPS. Once I have everything up and running, I will start a new thread if there's enough interest from others in what these boogers sound like. I will also compare Ryan's PSU against Bruno's SMPS.

 

[garylkoh]

VERY nice! That is the way to go for the Hypex modules, also the UcD. I do a similar power supply for my Reference amps, but I use 23 different capacitor values, but only two of each value, whereas it looks like you've got more than two of each value.

One additional trick I use is that I use a LC resonator tuned to the sampling frequency of the module. This adds a little of a "jitter flywheel" effect to make things even smoother.

Let us know what the amp sounds like. I'm guessing that it will be quite spectacular.

 

[dallasjustice]

VERY nice!
Let us know what the amp sounds like. I'm guessing that it will be quite spectacular.

Thanks Gary. I should have them by next week. WILCO on the reveiw. I have the BMC monoblocks to compare them against. I believe the BMC monoblocks are top performers, so it should be fun.

 

[Odyno]

Just completed my NCore amp with its SMPS.
Just to let you know that the NCore outperform the UcD in every area and specially the bottom end.
I previously used for power amp a Pass Labs Aleph 5 then went for an amp with UcD400HG then to a pair of UcDHG HxR and from there to the NCore.
The performance was always upward but wasn’t expecting such a gap from the UcDHG HxR to the NCore.

 

[barrows]

Dallas: I look forward to hearing your impressions of the nCores. You probably have one of the more refined systems to evaluate these amps in, compared to those who have reported on sonics so far. I am soon going to be building a pair for a friend, and will be able to compare with my Pass. I am actually hoping I like the nCores, as I would love to trade to them and get the benefit of less power use.

 

[AudioExplorations]

barrows, welcome to WBF! I'm hoping you are the same 'barrows' that hangs out over at computeraudiophile... I learnt A LOT from your posts and discussions there (I was jdebonth on CA and later changed my nick to AudioExplorations in case you remember). You will love it over here, a lot of very smart people in this community.

You should check out the preamp/no preamp thread over here, took a bit of a different turn to the various times this came up on CA...

 

[barrows]

Yes, same "barrows" here, and thanks for the welcome. I will check out the preamp thread, although I already know what works best in my system (no preamp).