Choosing several Class-D amplifiers for my active 4-way loudspeaker

[brandonhook1]

Hey Everyone,

I'm having a hard time making a choice between all the available options for amplification out there. I have a 4-way dipole design (kind of similar to LX521), with the following arrangement:

23-110Hz, Dipole sub:
This is using dual 15" drivers, each measuring 3.4 ohms at the lowest point in the sweep in my freq range. I'm using an XLS2500 with drivers in parallel, red LED periodically lights up on really big bass hits. measured amplifier output level to the drivers is regularly reaching 650W with my music collection.

110-550Hz, open baffle mid-bass:
requires about 200W peak, dual WO24TX-8's

550Hz-2.3kHz, Coaxial midrange:
no surprise this only requires about 10-15W max

>2.3kHz, tweeter:
less than 5 watts normally.

I'm using a DSP for the crossover and compensation equalizations.

With the amplifiers I currently use, I can hear hiss and other noise from all frequency ranges. I'd like to upgrade. I can certainly find adequate power in the Hypex lineup for the subwoofers and mid-bass ranges, but I'm not sure what is the best option for the mid/tweet coax.

I could pick up 2 stereo amplifiers from Buckeye for example, the NC502 (each channel feeding a single 15" driver @ 3.4 ohms), and the NC252 (one channel feeding parallel WO24TX-8's [2.9 ohm], the other feeding the coax with passive crossover at 2.3kHz [3.6 - 6.0 Ohms]), but in this arrangement so much headroom is available for the coax driver. It feels suboptimal, as I lose the DSP benefit for mid/tweeter crossover.... and I worry about my coaxial's death should something bad happen unexpectedly.

Also, I feel it's not normal to have 250W available for a midrange/tweeter.... seems like using a semi truck to pickup groceries.

What should I do? I look forward to any recommendations or suggestions you have to offer .

 

[EERecordist]

https://www.google.com/search?q=4+way+audio+speakler+plate+amp DSP crossover built in. Many are ICE. It's hard to imagine today the noise is in the power amp. Keep us posted on your solution which worked. Maybe consider setting up at home with a calibrated microphone and Room EQ Wizard.

 

[kemmler3D]

https://www.google.com/search?q=4+way+audio+speakler+plate+amp DSP crossover built in. Many are ICE. It's hard to imagine today the noise is in the power amp. Keep us posted on your solution which worked. Maybe consider setting up at home with a calibrated microphone and Room EQ Wizard.

Have you actually seen tests of any of these? I've looked for 4-way plate amps that are even remotely comparable to the Hypex Fusion stuff and found exactly nothing sold to consumers, (Pascal amps could be an option, but are OEM-only) and a lot of questionable stuff similar to what comes up in that search.

@brandonhook1 my suggestion for moderate power, (mids & tweet) low noise and decent cost would be Fosi stuff. If you don't care about cost, Topping has some extremely clean, moderate-power amps but they're pricier per channel/watt.

For the sub I'm surprised you can hear any hiss with the current amp. If you need more than 600 super-clean watts, there aren't a ton of options off the top of my head.

And welcome to ASR!

 

[EERecordist]

Have you actually seen tests of any of these? I've looked for 4-way plate amps that are even remotely comparable to the Hypex Fusion stuff and found exactly nothing sold to consumers, (Pascal amps could be an option, but are OEM-only) and a lot of questionable stuff similar to what comes up in that search.

I have not used or tested any dsp plate amps. They are an OEM business which ASR-respected brands design. It would be fun to have them tested by a sensitive Audio Precision.

 

[brandonhook1]

Have you actually seen tests of any of these? I've looked for 4-way plate amps that are even remotely comparable to the Hypex Fusion stuff and found exactly nothing sold to consumers, (Pascal amps could be an option, but are OEM-only) and a lot of questionable stuff similar to what comes up in that search.

@brandonhook1 my suggestion for moderate power, (mids & tweet) low noise and decent cost would be Fosi stuff. If you don't care about cost, Topping has some extremely clean, moderate-power amps but they're pricier per channel/watt.

For the sub I'm surprised you can hear any hiss with the current amp. If you need more than 600 super-clean watts, there aren't a ton of options off the top of my head.

And welcome to ASR!

It certainly might be 2x4HD creating the noise floor, to be honest, I haven't focused any effort to investigate that. It's very faint, but certainly not difficult to hear when you're standing near the drivers. At the listening position, it's silent, so maybe the noise is actually a non-issue practically speaking.

I really wish I could get by with the fusion plate amps. I initially bought the FA253 amps last year in preparation for this build, but my ripole math was a bit premature, and so, they had to go. I would absolutely love to make my own DSP+HYPEX 4-way plate amp. I'm up to the task engineering wise, but lacking the motivation to do more layouts and board builds.

I'll check out Fosi and Topping. I did notice that Topping "Tops" Amir's SINAD list.

Is there any real issue driving the mid/tweeter with a NC252MP channel?

 

[kemmler3D]

Is there any real issue driving the mid/tweeter with a NC252MP channel?

I think that would be fine... The SINAD numbers on those are not SOTA in 2025, but are more than decent. They are a bit overkill in terms of power but otherwise I have no objections.

 

[Sokel]

Get an icepower 1200as2 and hook two 300a2 on it if you can DIY a little,
This way you can either have any number of channels from 3 to 6 to mix and match.

All for under a grand and you can drive trains with it.

 

[brandonhook1]

Get an icepower 1200as2 and hook two 300a2 on it if you can DIY a little,
This way you can either have any number of channels from 3 to 6 to mix and match.

All for under a grand and you can drive trains with it.

I actually really like this idea. I have read thru all the ICE data sheets, and I see no real limitations for me, the noise floor is certainly still below audibility thru my drivers, and the SINAD is unlikely to be noticeable in my application. Are there any real-world practical issues using the ICE modules versus the Hypex/Purifi? (other than heat?)

 

[FERCSA]

I had a 3-way active setup with three Fosi V3 stereo. You can't find a better deal. If you have more money get the PA5II.
The "one box" solution would be a hypex plate amp and it includes DSP as well.

One more note if you have a small/medium room I can't image a TPA based chip amp (mentioned above) wouldn't be enough.

 

[JohnnyNG]

I don't have any suggestions but would love more details on your speaker.

 

[Sokel]

Are there any real-world practical issues using the ICE modules versus the Hypex/Purifi? (other than heat?)

None that I have encounter 5 years that I'm running them.
I have a 1200as2+300a2 as monos using each one channel for lows - mid/highs for my semi-active setup (the price difference with mono modules is tiny so why not not have some channels as reserve? )

There's two things that might not suit you.
One is that the 1200as2 is a bridged design so any high level connection with a sub for example is a no. No ground for these.

The second one is to exploit the full potential. they have a gazillion of controls and they can report pretty much anything.
They will work as is out of the box just by connecting the mains, inputs and speakers but it would be nice to take advantage of their potential.

A multicolor (common anode) led for example can show you a lot of what they report (clipping, etc), or a second input can come handy for their SignalSense feature (they turn on with signal and off after 13 min without) .

In any case we're here to help if you decide.

 

[brandonhook1]

None that I have encounter 5 years that I'm running them.
I have a 1200as2+300a2 as monos using each one channel for lows - mid/highs for my semi-active setup (the price difference with mono modules is tiny so why not not have some channels as reserve? )

There's two things that might not suit you.
One is that the 1200as2 is a bridged design so any high level connection with a sub for example is a no. No ground for these.

The second one is to exploit the full potential. they have a gazillion of controls and they can report pretty much anything.
They will work as is out of the box just by connecting the mains, inputs and speakers but it would be nice to take advantage of their potential.

A multicolor led for example can show you a lot of what they report (clipping, etc), or a second input can come handy for their SignalSense feature (they turn on with signal and off after 13 min without) .

In any case we're here to help if you decide.

Ok thank you. That's good to know. It seems like there's a lot of creative and useful possibilities here. I'm already planning to make my own DSP + Codec + balanced line driver board, might be good to incorporate some of the ICE module control and status flags, either via small OLED screen, or just some LED's.

I'm wondering, how well do the ICE modules handle lower impedance loads? It's not so clear in the data sheets (performance plots) how the 700AS2 or 1200AS2 behave with <4 ohms. Each channel would be driving a 3.4 Ohm load.

 

[Sokel]

Ok thank you. That's good to know. It seems like there's a lot of creative and useful possibilities here. I'm already planning to make my own DSP + Codec + balanced line driver board, might be good to incorporate some of the ICE module control and status flags, either via small OLED screen, or just some LED's.

I'm wondering, how well do the ICE modules handle lower impedance loads? It's not so clear in the data sheets (performance plots) how the 700AS2, or 1200AS2. Each channel would be driving a 3.4 Ohm load.

1200as2 drives 2 Ohm no problem (2.7 official) , it have been tested here too.
A warning though, 700as2 is not ice Edge technology, is older.
It will upgrade to Conductor technology (latest) soon I guess.

Do a search here, there's couple of threads about it.
From what we have tested the 1200as2 is the sweet spot for lows, 300a2 and 400a2 are near 100dB SINAD, all low noise so choose wisely.

Edit: Since you seem creative, exploit their best feature, their analog voltage/current outputs either with VU meters or a nice screen.

 

[brandonhook1]

Looks like I have some further reading to do. Thanks Sokel. The hardest part for me is getting started. I really appreciate the steering and suggestions .

 

[Valentino]

Very interesting project, a bit like mine which has been evolving/been refined for 16 years now. It operates below the dipole peak from 45 to 6000 Hz. I use a NC252MP+100HF for each 3-way dipole and a Icepower125asx2 for each closed box sub.There’s a block diagram in my profile.

 

[brandonhook1]

Very interesting project, a bit like mine which has been evolving/been refined for 16 years now. It operates below the dipole peak from 45 to 6000 Hz. I use a NC252MP+100HF for each 3-way dipole and a Icepower125asx2 for each closed box sub.There’s a block diagram in my profile.

That's pretty impressive. I wasn't able to meet all my polar characteristics requirements for this build, My biggest issue was being below the dipole peak past 2.5kHz without going nude. If I went nude, I had to dramatically sacrifice my overall max SPL. In the end, I chose headroom over radiation characteristic. I'd love to see your design .

 

[brandonhook1]

1200as2 drives 2 Ohm no problem (2.7 official) , it have been tested here too.
A warning though, 700as2 is not ice Edge technology, is older.
It will upgrade to Conductor technology (latest) soon I guess.

Do a search here, there's couple of threads about it.
From what we have tested the 1200as2 is the sweet spot for lows, 300a2 and 400a2 are near 100dB SINAD, all low noise so choose wisely.

Edit: Since you seem creative, exploit their best feature, their analog voltage/current outputs either with VU meters or a nice screen.

Is the ICEpower 1200AS2 enough, or do I need to move to the 2000AS2 HV?

I'm worried that the 1200AS2 may be underpowered in my application once I hang a 300A2 module off it. I'm currently running a pair of high-excursion 15" drivers in a ripole sub configuration, driven by a 1200W/channel amplifier (Crown XLS2500), and I occasionally see clipping on high-bass moments in movie playback. That load would be different with the ICE amps. It would see a 3-Ohm 500-600W load per channel, during the same playback peaks.

Simultaneously, I'd be hanging a 300A2 off the 1200AS2. Here's what it would be driving:

• Channel 1: Parallel WO24TX-8s (2.9Ω) — up to 250W peak
• Channel 2: Coaxial mid/tweet crossed >500Hz — 20W absolute max

From the datasheet:
“The continuous combined output power is the sum of the power of the driven channels. The available output power can be distributed evenly or asymmetrically…”

That seems to confirm what I’m worried about: during dynamic peaks, I may run out of rail or thermal headroom when both the sub section and the 300A2 are pushing. My question:

Do these ICE modules have sufficient local energy storage (bulk caps) to handle short-term peaks without robbing other channels, or does power draw on one channel directly constrain the others in real-time?

I’m considering the 2000AS2 HV as an upgrade if needed. It's overkill on paper, but the additional headroom might ensure stability under simultaneous sub/midbass bursts. Also, seems like the thermal load would be better managed with the bigger amp board.

For reference, I'm measuring ~103dB SPL at 2.5 meters at peak moments during movie playback in my current setup (mono tower), and I want to maintain that standard — not interested in losing output capacity.

Would love to hear from anyone who has pushed the 1200AS2 hard while hanging a 300A2, or from anyone who’s deployed the 2000AS2 HV in a serious multi-way active system.

At these levels,

Thanks!

 

[Holmz]

Hey Everyone,

I'm having a hard time making a choice between all the available options for amplification out there. I have a 4-way dipole design (kind of similar to LX521), with the following arrangement:

23-110Hz, Dipole sub:
This is using dual 15" drivers, each measuring 3.4 ohms at the lowest point in the sweep in my freq range. I'm using an XLS2500 with drivers in parallel, red LED periodically lights up on really big bass hits. measured amplifier output level to the drivers is regularly reaching 650W with my music collection.

110-550Hz, open baffle mid-bass:
requires about 200W peak, dual WO24TX-8's

550Hz-2.3kHz, Coaxial midrange:
no surprise this only requires about 10-15W max

^these^ do not REQUIRE 10-15W, that is max.

>2.3kHz, tweeter:
less than 5 watts normally.

Max I would think.

I'm using a DSP for the crossover and compensation equalizations.

With the amplifiers I currently use, I can hear hiss and other noise from all frequency ranges. I'd like to upgrade. I can certainly find adequate power in the Hypex lineup for the subwoofers and mid-bass ranges, but I'm not sure what is the best option for the mid/tweet coax.

I could pick up 2 stereo amplifiers from Buckeye for example, the NC502 (each channel feeding a single 15" driver @ 3.4 ohms), and the NC252 (one channel feeding parallel WO24TX-8's [2.9 ohm], the other feeding the coax with passive crossover at 2.3kHz [3.6 - 6.0 Ohms]), but in this arrangement so much headroom is available for the coax driver. It feels suboptimal, as I lose the DSP benefit for mid/tweeter crossover.... and I worry about my coaxial's death should something bad happen unexpectedly.

Also, I feel it's not normal to have 250W available for a midrange/tweeter.... seems like using a semi truck to pickup groceries.

What should I do? I look forward to any recommendations or suggestions you have to offer .

I would just put 2x AIYAMA A70 amps, and use that for the R&D part. They like $70 on Amazon, or they were.
You would need 8 channels, but you can test a single speaker that way, and do all the EQ stuff.

Hypex do a 3 channel amp, but you would still likely want a another 2 channels for the 21-110 Hz.

 

[Sokel]

Is the ICEpower 1200AS2 enough, or do I need to move to the 2000AS2 HV?

I'm worried that the 1200AS2 may be underpowered in my application once I hang a 300A2 module off it. I'm currently running a pair of high-excursion 15" drivers in a ripole sub configuration, driven by a 1200W/channel amplifier (Crown XLS2500), and I occasionally see clipping on high-bass moments in movie playback. That load would be different with the ICE amps. It would see a 3-Ohm 500-600W load per channel, during the same playback peaks.

Simultaneously, I'd be hanging a 300A2 off the 1200AS2. Here's what it would be driving:

• Channel 1: Parallel WO24TX-8s (2.9Ω) — up to 250W peak
• Channel 2: Coaxial mid/tweet crossed >500Hz — 20W absolute max

From the datasheet:
“The continuous combined output power is the sum of the power of the driven channels. The available output power can be distributed evenly or asymmetrically…”

That seems to confirm what I’m worried about: during dynamic peaks, I may run out of rail or thermal headroom when both the sub section and the 300A2 are pushing. My question:

Do these ICE modules have sufficient local energy storage (bulk caps) to handle short-term peaks without robbing other channels, or does power draw on one channel directly constrain the others in real-time?

I’m considering the 2000AS2 HV as an upgrade if needed. It's overkill on paper, but the additional headroom might ensure stability under simultaneous sub/midbass bursts. Also, seems like the thermal load would be better managed with the bigger amp board.

For reference, I'm measuring ~103dB SPL at 2.5 meters at peak moments during movie playback in my current setup (mono tower), and I want to maintain that standard — not interested in losing output capacity.

Would love to hear from anyone who has pushed the 1200AS2 hard while hanging a 300A2, or from anyone who’s deployed the 2000AS2 HV in a serious multi-way active system.

At these levels,

Thanks!

Have a look at this thread, it will give you all the answers (or most) about power and comparison with 2000as2.

Power amps: continuous output power measurements

When amps are tested here at ASR as well as other places, output power is one of the things measured. I can't however find any information about the duration of such a measurement. They are also conducted at 1 kHz only, if I'm not mistaking? Now - consider we want our amps to drive home theater...

www.audiosciencereview.com

To cut the story short, 1200as2 will be more than enough for your needs and most importantly is steady as a rock.

Have a look at real time test, away from datasheets (taken from the thread) :

 

[brandonhook1]

Thank you Sokel, I most likely would never have stumbled across that thread! .

One more question: What is the consensus out there about heat sinking or airflow requirements for the 1200AS2? What's the real-world temp rise experienced on that module when mounted in a non-vented or limited vented enclosure? I will be building my own "plate amp" of sorts for these modules, together with the DSP I/O board I'm making (balanced input-->DSP-->4 channel balanced output board, diagnostics/protection). I just want to make sure I have a sense from the community about these module's temp rise.