Is there a concensus on what is currently "the best" class D amplifier boards? Specifically "best" for low noise. And availability from reputable sources would be a big plus for me!
I built some active speakers featuring pro audio coaxial drivers, and the horns are extremely sensitive, which means noisy. The horns are rated at 106dB, and they currently have a massive pad on them to cut the noise to a tolerable level. But it is still there.
So I would like to build a class D amp that is extremely quiet. At a minimum I'd like to build a 2 channel 30+ watt amp in a 1RU form factor with balanced inputs.
I would be interested in buying a ready made unit, but I haven't found one. I really do want the 1RU form factor and balanced inputs. And I want as much of a guarantee as possible that the amp will be genuinely quiet, suitable for high sensitivity drivers.
Depending on cost and availability, I may want to make it a 4 way amp, with 2x 30watt + 2x ~100 watts. (My mids are 98dB sensitivity, so they are also a bit noisy and require a pad to tame them...)
My current amp is a multichannel emptiva xpa gen 3 (class H).
I built some active speakers featuring pro audio coaxial drivers, and the horns are extremely sensitive, which means noisy. The horns are rated at 106dB, and they currently have a massive pad on them to cut the noise to a tolerable level. But it is still there.
So I would like to build a class D amp that is extremely quiet. At a minimum I'd like to build a 2 channel 30+ watt amp in a 1RU form factor with balanced inputs.
I would be interested in buying a ready made unit, but I haven't found one. I really do want the 1RU form factor and balanced inputs. And I want as much of a guarantee as possible that the amp will be genuinely quiet, suitable for high sensitivity drivers.
Depending on cost and availability, I may want to make it a 4 way amp, with 2x 30watt + 2x ~100 watts. (My mids are 98dB sensitivity, so they are also a bit noisy and require a pad to tame them...)
My current amp is a multichannel emptiva xpa gen 3 (class H).
According to the internet, there was a Topping RA3, which is a 1RU form factor and balanced inputs. But despite fairly recent reviews of the unit, it does not seem to exist?
But I would prefer something made in the States or from a company that has a presence in the states. I generally avoid buying from places like Aliexpress.
But I would prefer something made in the States or from a company that has a presence in the states. I generally avoid buying from places like Aliexpress.
What specification did you read? For what product? I read the data sheet for every ICEpower amp that Parts express sells, and what did I find? Well, I'll show you!
The fact that the 100AS1/ AS2 are given a rating means that they were designed to be better? Missing ratings always mean the ratings are not good enough to brag about.
The "<77" means I should expect it to be right around 77....but how does this compare to all the other amps? I don't have any ratings for other amps, and unfortunately Amir at Audio Science Review does not include this output noise in his testing, so that number means next to nothing to me right now.
So I'm hoping someone can provide info on what measurement value is actually good? Is rating of 77 particularly good? Or is it more "meh"?
Amir has a review of the 125ASX2, which is $100 more than the 100AS2. The 125ASX2 landed in the top half of his amplifier ratings, but it was not exceptional. So it seems unlikely that the 100AS2 is a quieter amp at a lower price point. But it has a noise rating that none of the other amps have...
Ok, figuring this out a little. I found µV measurements on the ICEpower website, under specifications tab. But µV measurement is missing from their data sheets? I typically expect data sheets to have the most complete information, so I've been slightly confused.
Would there be any meaningful difference between amps with 20 and 25µV noise?
Choices. Please comment if you think I'm on the wrong track here?
Horns: 106dB sensitivity, 8ohm. These needs the lowest noise, so probably the 50ASX3BTL.
Mids: 98dB, 8ohms. I'm thinking the 250ASX2 with 125watt x2 is plenty.
Woofers: 89dB, 4ohms. Noise probably doesn't matter now, so maybe a 300AS1 +300A1
Subwoofer: 86dB, 4ohm. I'm leaning toward the 500ASP + 500A.
Would there be any meaningful difference between amps with 20 and 25µV noise?
Choices. Please comment if you think I'm on the wrong track here?
Horns: 106dB sensitivity, 8ohm. These needs the lowest noise, so probably the 50ASX3BTL.
Mids: 98dB, 8ohms. I'm thinking the 250ASX2 with 125watt x2 is plenty.
Woofers: 89dB, 4ohms. Noise probably doesn't matter now, so maybe a 300AS1 +300A1
Subwoofer: 86dB, 4ohm. I'm leaning toward the 500ASP + 500A.
https://reference-audio-analyzer.pro/en/v-dbv-dbu.php#gsc.tab=0
20->25; Approx 2dB.. barely detectable difference I would say...
//
20->25; Approx 2dB.. barely detectable difference I would say...
//
Best ones currently available would probably be something like an ncx500 (7uv unwtd@11.5dB gain), or 1et400a (11.5uv awtd@12.8dB), or nilai500 (9uv unwtd@11.9dB), or nc500(9uv unwtd@12.4dB), they all have fairly low gain levels but since you're only doing like 30 watts it wouldn't be a problem, these are indeed quite expensive though, that 3e audio board is a fairly good value but i guess it's not available in your area? Bummer. There's also a 1et9040ba around the corner, which is gonna be even more expensive, but is 4.5uv awtd@14.4dB gain, definitely the lowest noise class d out these, period. If there isn't an output noise rating you can use snr to calculate it, noise=(10^-(snr/20))*sqrt(impedance*power that it's tested at). if the test power is not specified then just use the rated power for the highest impedance rating and calculate at that impedance, for example 8ohms*380watts for the ncx500
While it's a technological and value marvel its performance is best under your bandwidth of interest.
Another option may be a Tannoy-style autotransformer or speaker transformer. Sounds a bit perverse but it greatly increases your field of suitable amplification. A recent Joseph Crowe blog post measured an Intact Audio version.
https://intactaudio.com/Speakerformer.html
Attachments
Thanks. You made me do math...
I see how you got the -2dB relative difference, but it is referencing 1v. But the sensitivity measurement of the driver is based on 2.83v
So I made this spreadsheet. Let me know if I got it right? This is from first principles as I understand them, where half the power (watts) = -3dB. So I'm calculating what the voltage and amperate needs to be for each successive power level. My results seem pretty similar to the results from the online calculator, so I'm hoping it is right?
If these numbers are roughly correct, then the noisiest ICEpower amp would produce less than 16dB at 1 meter noise when connected to my horns. The ambient noise is well above that, so it seems output noise is a non-issue with these amps?
Or am I missing something?
I found a good data point for me, atleast. The Genelec 8040A, which I used to own, has a self-noise rating of <= 10dB at 1 meter. I remember hissing being slightly audible when I was close to the speakers.
So it was a mistake to say 16dB noise is inaudible.
Based on my numbers, 20-30µV noise on my horns would be the same or less than the self-noise of the Genelec. That's good enough for me. But is my estimate right? Not sure....
So it was a mistake to say 16dB noise is inaudible.
Based on my numbers, 20-30µV noise on my horns would be the same or less than the self-noise of the Genelec. That's good enough for me. But is my estimate right? Not sure....
This is something I’ve battled with through out my horn journey too. I listen near field with big horns.
Below are ones that I’ve tried that are pretty much dead quiet on >110db horns.
class d: ncore nc500
class a: accuphase, first watt
Class ab: topping la90
Have tried countless tube amps and have yet to find one that is dead quiet through the horns.
Would appreciate to hear if others have any other recommendations. Especially class d.
Cheers
Below are ones that I’ve tried that are pretty much dead quiet on >110db horns.
class d: ncore nc500
class a: accuphase, first watt
Class ab: topping la90
Have tried countless tube amps and have yet to find one that is dead quiet through the horns.
Would appreciate to hear if others have any other recommendations. Especially class d.
Cheers
Just as another point of reference, I'm using a topping PA3s amp on HF108 drivers. The amp spec says noise is <70uv rms (A weighted), the 8ohm drivers are given as 109db 1w at 1m.
At 1m I can clearly hear some background noise when no music is playing, I find it loud enough to be irritating, so will need to try padding the drivers. By 3m (from the naked driver, no horn/waveguide) it isn't intrusive, and I'd be OK with it, but it is audible in quiet conditions; I also notice the silence after switching the amp off.
The noise is a constant level (independent of the volume knob position and unrelated to source/inputs). But that level is 'much' quieter than the commercial class AB hifi amps I've tried previously. It is also less intrusive in tone, more like a gentle swirling wind noise than a hum or sharp hiss for example.
Edit: that is using active crossovers so direct wired amp to driver. If I was to buy again I'd probably consider the 16ohm version of the drivers to reduce voltage sensitivity, if the amps woukd work ok at that load. I think (for my living room environment) there would still be more than enough SPL headroom even allowing for flattening their response on a controlled directivity waveguide.
At 1m I can clearly hear some background noise when no music is playing, I find it loud enough to be irritating, so will need to try padding the drivers. By 3m (from the naked driver, no horn/waveguide) it isn't intrusive, and I'd be OK with it, but it is audible in quiet conditions; I also notice the silence after switching the amp off.
The noise is a constant level (independent of the volume knob position and unrelated to source/inputs). But that level is 'much' quieter than the commercial class AB hifi amps I've tried previously. It is also less intrusive in tone, more like a gentle swirling wind noise than a hum or sharp hiss for example.
Edit: that is using active crossovers so direct wired amp to driver. If I was to buy again I'd probably consider the 16ohm version of the drivers to reduce voltage sensitivity, if the amps woukd work ok at that load. I think (for my living room environment) there would still be more than enough SPL headroom even allowing for flattening their response on a controlled directivity waveguide.
Last edited:
You must take the distance to the source into account.
https://www.wkcgroup.com/tools-room/inverse-square-law-sound-calculator/
//
Post in thread 'Suggested affordable quiet amp for high efficiency speakers' https://www.diyaudio.com/community/...-high-efficiency-speakers.399146/post-7347003
I also have active crossovers, but I have an L-Pad circuit and capacitor in line with the driver. I think it is a -20dB pad, but hissing is still evident. I have a 30ohm resistor in series and a 6ohm in parallel? I don't remember exactly, and I don't want to remove a driver to check... Each horn is on it's own 65watt channel, and the L-Pad calculators I used said the horn is only seeing 3 watts of the available 65 watts. I feel I've gone as far as I can go with an L-Pad!
The one thing I can say about the Emotiva amp (Class H) is the noise is the least objectionable noise I've heard from an A/B amp. It's just steady white noise.
If I measure AC voltage on my current amplifier speaker output in mV, will that give an accurate representation of this noise? I'm usinga Fluke 115, so I expect the number is fairly accurate. I am getting 15mV. On my chart that translates to -45dB, so that should translate to 61dB @ 1 meter. I know it was 70dB with the SPL meter inside the horn, which I did to eliminate the influence of ambient noise...
I'm just trying to understand how bad my amp is, and how much better the new amp will be.
I'm just trying to understand how bad my amp is, and how much better the new amp will be.