I Have Tried...
I have built an nCore 400.
A class D audio cda-120 with a standard toroid power supply
A Texas Instruments TA55630 based amp with a Mean Well switching supply and a couple of single ended triodes, one a 2A3.
I listen to Magnepan III's, Spica TC-60's and Large Advents.
I have settled on a classic AB amp with high bias point (Van Alstine) for the Magnepans. The nCore was just too relentlessly revealing... Very impressive and great as a tool to expose a recording or signal chain but too much for relaxing.
The cda-120 is very clean and pretty sounding in the highs but not enough power for the Maggies... Fine with the others but a little lean in the bass on all of them.
The TI based amp with the SMPS was fine with all but not a stand out in any area... That one has moved on. The nCore matches beautifully with the Advents where it's power slams the bass and the phenomenal detail pushes through the old school tweeters.
I have built an nCore 400.
A class D audio cda-120 with a standard toroid power supply
A Texas Instruments TA55630 based amp with a Mean Well switching supply and a couple of single ended triodes, one a 2A3.
I listen to Magnepan III's, Spica TC-60's and Large Advents.
I have settled on a classic AB amp with high bias point (Van Alstine) for the Magnepans. The nCore was just too relentlessly revealing... Very impressive and great as a tool to expose a recording or signal chain but too much for relaxing.
The cda-120 is very clean and pretty sounding in the highs but not enough power for the Maggies... Fine with the others but a little lean in the bass on all of them.
The TI based amp with the SMPS was fine with all but not a stand out in any area... That one has moved on. The nCore matches beautifully with the Advents where it's power slams the bass and the phenomenal detail pushes through the old school tweeters.
The pumping effect is nothing you can do about it, as every buck do pump up, except very big caps or a wel regulated smps can limit it, but if ypu like house music, not only ypu do pump up ut also the class d.
Best is using a full bridge, has advances al the way.
regards
Last edited:
You wrote, it's not clear. With such a high efficiency of the cascade in class D as 80-90%, the positive effect of the recovery can not be neglected so as to maximize its use. It would be good to quote the source of the book or article. As for example, this is indicated in the patent Рисунки патента 1490704 - Усилитель мощности класса д
It states that the goal of recuperation is to reduce the energy losses in the primary winding of the transformer and incoming wires due to the difference between the current consumed and the current returned.
It states that the goal of recuperation is to reduce the energy losses in the primary winding of the transformer and incoming wires due to the difference between the current consumed and the current returned.
Last edited:
Hi All
I have done now some schematic work and sims, combine different ones.
I do however not understand the 3th harmonic who is high and the rest is very low, even under -100dB afcourse in sim, not real, these version is 3 level, however I get not get that nice 3 level output, however, the signals are shifted, can the models as lossles switches what I use the reason that it do not work, maybe #grizlek can explane. the only way I get the three level as it has to be on output when use two triangles one inverted with offset, so there is one above and one below zero, that way comparators can do there job better, but in books, a shifted output puls train has to do the job just with two audio signals and one inverted no offset.
I have to say I do not see the three level singals on the output but see that the carrier is effective doubled to 600 Khz so it does work in some extense..
I have done now some schematic work and sims, combine different ones.
I do however not understand the 3th harmonic who is high and the rest is very low, even under -100dB afcourse in sim, not real, these version is 3 level, however I get not get that nice 3 level output, however, the signals are shifted, can the models as lossles switches what I use the reason that it do not work, maybe #grizlek can explane. the only way I get the three level as it has to be on output when use two triangles one inverted with offset, so there is one above and one below zero, that way comparators can do there job better, but in books, a shifted output puls train has to do the job just with two audio signals and one inverted no offset.
I have to say I do not see the three level singals on the output but see that the carrier is effective doubled to 600 Khz so it does work in some extense..
Attachments
Last edited:
Have you tried to find out what is better in sounding the standard toroid power supply or switching supply? Whether there are any interferences, because of the different frequency of the operation of the switching devices.
No I haven't.
The TI module and the cda module required completely different supplies.
I will say the sure 32512 run off a toroid supply is sounding really really fine.
Connected to Magnepan III's yesterday and it is still there. No complaints yet.
The Hypex sound really fine and they are switching supply.
The TI module and the cda module required completely different supplies.
I will say the sure 32512 run off a toroid supply is sounding really really fine.
Connected to Magnepan III's yesterday and it is still there. No complaints yet.
The Hypex sound really fine and they are switching supply.
Last edited:
It is not needed perse that the low pas in in feedback, technologies as sliding control of comparators can give psrr of -70 dB as in papers. and use that with air coils who are liniair and free of saturation can sound well.
Yes I need to go pcb-ing some ideas. the class I be the way is not only from Crown I did came through a site who show one who costs 25.000 dollar for one channel. also here the use of air coils.
Mark Levinson No.53 Reference monoblock power amplifier | Stereophile.com
regards
Yes I need to go pcb-ing some ideas. the class I be the way is not only from Crown I did came through a site who show one who costs 25.000 dollar for one channel. also here the use of air coils.
Mark Levinson No.53 Reference monoblock power amplifier | Stereophile.com
regards
Optimized work I get this HD with only prepost feedback, it use a triangle carrier, it is not ucd version. Pic 1 is 3 amp output, and pic2 is 16 amp output 8 ohms. full bridge 3 level shifted switches,
You see it can also be low in hd, see also the last pic, where the shaper is removed and the difference.
class d is quite a challenging technology with quite good outcomes these days.
regards
You see it can also be low in hd, see also the last pic, where the shaper is removed and the difference.
class d is quite a challenging technology with quite good outcomes these days.
regards
Attachments
Among the ones I have tried, I agree with you.
NC400 is very impressive. Very strong and it controls the speakers like crazy. I doubt there is any speaker that it cannot control. However, it gets tiring for me after 30 minutes or so.
TI based amps are nice too. Though their highs gets irritating kinda. Hard to explain but I think it could not really do well on the HFs for me.
The only amp that I still enjoy to listen is based on Tripath TA2022 chip. Although I am not a bass freak, I will say the bass is not their strongest point BUT the vocals/midrange is amazing IMO. It is very smooth (a little slow too maybe) and romantic sounding. I can listen to it without any problems for hours. They are not going to rock your speakers and probably will have a hard time driving the Maggies but if you listen to your music at low volumes, you may be OK with them.
Too bad Tripath is not around anymore but you can still buy their boards/chips.
I do not care too much about the measurements. I am sure NC400 will measure way better than the Tripath but I like the sound of my Topping TP60 better than NC400. And it is a lot cheaper.
Maybe a moderator can move this to the thread here so it is less a mess. to many tread I have who is not active anymore, the nabure radiation problems can also be deleted.
Class D investigation.
For what I do with class D this tread is better fitted on thread from link.
thanks.
Class D investigation.
For what I do with class D this tread is better fitted on thread from link.
thanks.
Yes, back to the original question. Which DIY Class D amp sounds best? Been reading through tons of threads in the Class D section... it's all quite mind-numbing to be honest.
The main issue is not everyone agrees on what sounds good. Most seem to say the units they built sound good, but the "reviews" and descriptions are limited. The most common problems being listener fatigue or lack of bass. Neither is appealing.
The more I think about it, I suppose I'm searching for is a DIY alternative to the NAD M27. I was really impressed with the M27's sound. Paired with the M17 pre/pro, it was dead silent, powerful, huge soundstage, accurate, and not harsh or fatiguing in the least. Dare I say a bit sweet overall?
I thought perhaps the NC400 modules might be similar to the M27. Apparently not based on jencelo's and bassnkeys' assessment of the NC400's they built. Fatigue being the main culprit. Oh well, the NC400 modules are expensive and wouldn't have saved me much over the M27 anyways.
Can anyone share a comparison between the NC400 and TI TPA3255/3251?
Listening fatigue on the Nc400? Sounds like you have issues with your source. I have never experienced this with my Ncore amps...
I have the fxaudio version of the tpa3250 and it is no match for the Ncore which are more refined. However, there are probably better implementations of those TPA amps than the fxaudio.
I have the fxaudio version of the tpa3250 and it is no match for the Ncore which are more refined. However, there are probably better implementations of those TPA amps than the fxaudio.
Maybe fatigue wasn't the right word. Nc400 is NOT bright at all, they are actually on the warmer side. However, after 30 minutes or so, I would need to turn the volume down. With TP60, I can listen to music without any change for hours. (Yes same volume levels). Though I am not bashing the nc400s or TIs. They are a great amps. I just like the Tripath chips better.
I have exactly the opposite experience with tpa3116 amps I have tried...
The tpa3250 is not as bright as those, but definitely not as transparent as the hypex.
Something was not right with the TPA3116's HF I had. It was the Sure board. Some high frequencies was sounding off for me. Maybe it was the faulty board but I doubt it.
Yes, I have used a variety of sure modules and came to the same conclusion...
Stuff about build, diy design, technologies and self edicated solutions for class D is now only on Class D investigation thread. no talks about the excisting modules but diy modules where we can talk about what is the best path to lower HD and class of warmth in the sound as well, including diy builds and the resulting sound, also openloop with digital generated triangles or comparator generated ones.. there are much kinds of tryouts and patents we can discuss here and post simulations..
We can hier going on with commercial modules.