Sure 3x 2x100W Amp Question

[zach_t]

Hello everyone. I am thinking about building my own 6x100W (into 4-ohms) amp using 3 Sure 2x100W boards and I have a couple questions. Ultimate goal is to use in a 3-Way miniDSP setup (hence the 6 channels).

I am looking at Sure AA-AB32971 which requires 15V-30V input and 8-A. (https://www.parts-express.com/sure-e...ogy)--320-3340)

Q1: Will the Mean Well 24VDC, 14.6A power supply work? Can I connect all three board to this one power supply (looks like I can from the picture)? Is 14.6A enough available current draw? I do plan on driving a 4-ohm load.

Q2: My CD has an adjustable 2V (maximum) line out. Is 2V enough to drive the miniDSP and the 2x100W amp board? How does this math work? What is the maximum output of the amp boards with 2V in?

Thanks!

P.S.- Is there a way to post two links in the same thread?

[Millstonemike]

Using 3 separate digital amps in proximity with one another has a good chance to cause problems. Since they are separate, each digital amp chip's switching frequency will be independent of the others. This has caused issues, interference (i.e., noise) with other model amps in past implementations. The only way to know if that will be an issue is to try them.

Alternatively, I would use the Sure 6x100 amp board. The switching frequencies for multiple chips on this board are locked together avoiding the above issue.

https://www.parts-express.com/sure-e...board--320-307

I presume you intend to Tri-amp two speakers using the miniDSP and the 6 channels of the amp(s). As such, you would never really have high wattage going through all 6 channels. Just like a passive XO, power will be split among the drivers for any given frequency.

Lastly, at 24 V, you will get a nominal 38 W rms into 8 ohm speakers (double that for 4 ohm drivers). So for two 8 ohm speakers 4 A should be sufficient. With a 36 V, you can achieve ~80 W rms into 8 ohm speakers requiring 4.5 A PS. Having said that, having a PS with a few extra amps never hurts.

You can paste multiple links in several ways. One way is this:

(a) Copy the URL from the address bar in the page you want to link.
(b) Position the cursor in your post where you want the link to appear.
(c) Type CTRL-V to paste the link at that position.
(d) Repeat as necessary.

[xmax]

Should work fine, those power supplies can do a real 350 watts continuos, the amps will not draw that.

[Kornbread]

FYI, the power rating for every Sure amp I've bought so far has been grossly overrated. If it says 100 watts, expect maybe a good clean 40 ... but ... IMO ... for really cheap amplification those T-amps are hard to beat for good sound on a cheap beer budget. Not champagne, more like paying the price for beer that just says beer on the can, but it tastes like Michelob.

I've got a 2x400 (32313?) on mids, and a 4x100 (it seems to have disappeared from PE's site) driving tweets on my not a khanspire ... build. No giant killer, but darn good for the money.

I'd recommend using separate power supplies if using separate amps, sometimes these amps like to talk to one another.

Also, if your considering these cheap amps I assume you are also considering the basic $85 minidsp 2x4 board. For a pair of 3-ways it will take two of them, and be mindful they are a bit sensitive in reguards to input and output levels. If you're good with computers, look Neil up, he's been working with the audrino? boards (IIRC the same board found in the basic mini2x4), the price is about 1/4 that of the minidsp.

[a4eaudio]

Kornbread, can you clarify the statement that they are grossly overrated?

For example, if one of the boards is listed as 2x100w (@ 6 ohms, 36v, 10% distortion) are you saying it cannot do that? Or is your comment that the 2x100w is deceiving because, although it can do that it sounds like crap, and that to sound good and clean (maybe using let than the maximum power supply and might only want distortion at 0.5%) you can only use 2x40w of the 2x100w that is possible.

[Kornbread]

Deceptive advertising. They should rate these boards more in line with reputable manufacturers with ratings @4or 8 ohm (toneburst or otherwise) and reasonable distortion figures. Level the playing field so to speak. My guess is not all boards are created equal, there may be some out there that surpass their ratings, into real loads, and real distortion figures ... not these.

I don't think anyone that cares about their music wants 10% distortion added to their favorite musical piece.

It's like that bag of tator' chips that says in big bold letters, "25% more" ... more what?

[mattk]

Deceptive advertising. They should rate these boards more in line with reputable manufacturers with ratings @4or 8 ohm (toneburst or otherwise) and reasonable distortion figures. Level the playing field so to speak. My guess is not all boards are created equal, there may be some out there that surpass their ratings, into real loads, and real distortion figures ... not these.

I don't think anyone that cares about their music wants 10% distortion added to their favorite musical piece.

It's like that bag of tator' chips that says in big bold letters, "25% more" ... more what?

Yep, 10% distortion is a ridiculous amount to use as a spec - when Sound & Vision measures amps/receivers, they measure output at 0.1% and 1%.

[Kornbread]

One thing I always liked about Stereophile, they took a barrage of measurements to validate, or sometimes dispute, their listening impressions. Their upper limit for thd vs output power was 1% into any load. Their tests were consistent and repeatable putting everyone on equal footing. There were a few exceptions ...

[xmax]

1%THD has long been considered clipping, the sure amps are rated poorly but do work surprisingly well for the money,
I see no problem with putting all those amps on the one PSU (I have done similar with no problems) If you want state
of the art Class-D amplifiers look at the Ncore by Hypex. They sound as good or better than most class A-B amps.

[Kornbread]

1%THD has long been considered clipping, the sure amps are rated poorly but do work surprisingly well for the money,
I see no problem with putting all those amps on the one PSU (I have done similar with no problems) If you want state
of the art Class-D amplifiers look at the Ncore by Hypex. They sound as good or better than most class A-B amps.

There just a ... little ... more expensive. Yes, the NC line is very nice. Their ucd line is no slouch either.

[Millstonemike]

Kornbread, can you clarify the statement that they are grossly overrated?

For example, if one of the boards is listed as 2x100w (@ 6 ohms, 36v, 10% distortion) are you saying it cannot do that? Or is your comment that the 2x100w is deceiving because, although it can do that it sounds like crap, and that to sound good and clean (maybe using let than the maximum power supply and might only want distortion at 0.5%) you can only use 2x40w of the 2x100w that is possible.

Kornbread's estimation of "40 clean watts" is a keen observation. I calculated ~38 W rms for 24 V and 8 ohm speakers in my original post. it is possible to get 100 W out of the amp but you need and drive a 3 ohm speaker - it's all marketing.

"Clean' W rms is based on the PS V+ and the speaker impedance. Those two factors, V+ and speaker impedance, impose a hard limit on W rms" prior to clipping dictated by physics. Before clipping, THD arises from the inherent design of the chip; typically a good deal below 1%.

Below is the distortion graph for the TDA7498E chip used in the Sure "X x 100" boards. The chart is for 4 ohms. You can see that clipping just begins after 78 W rms where THD is held to under 0.1%. At that same 24 V, 8 ohm drivers will get half that power before clipping: 39 W. Using 8 ohm speakers at 36 V yields a clean 78 W rms before clipping.

[zach_t]

Thank you everyone! Great conversation.

1. Good point in regards to having multiple switching amp next to each other - I had not thought of that.
2. I chose 3x 2x100W because the 6x100W option had a minimum impedance of 6-ohm. I have a pre-existing midrange I want to use and it's minimum impedance is 4-ohm (actual) @ 200-Hz. It will stay under 6-ohm between 150-Hz and 1,000-Hz.
3. The 2x100W amp are rated at 78W at 1% into 4-ohms.
4. My midrange can handle 18-V (81-watts), hence selecting the 78W @ 1% amp (to get to maximum excursion - in room, approx. 103-dB).
5. I thought I had seen a reasonably priced 2x8 miniDSP amp, but there is only a 2x10, which means I may want to re-think this a little.

OPTIONS
1.) 2x4 miniDSP + 4x100W Sure amp (https://www.parts-express.com/sure-e...08-(t--320-335) [64W into 4-ohm @ 1% THD] + subwoofer with plate amp.
2.) 2x 2x4 miniDSP + 3 separate amps, 2x100W ea (or less for the tweeter if needed).

[neildavis]

Having multiple independent class D amps sharing the same ground or even being close to each other is sometimes a problem. For the older Tripath-based designs, the two stereo channels were typically "tuned" to be greater than 40KHz apart in frequency, This ensured that the difference frequency between the channels was out of the audio range. This is what is says in the TC2000 data sheet, which is the modulator used for these amps:

CFB:  Feedback delay capacitor that both lowers the idle switching frequency and filters very high frequency noise from the feedback signal, which improves amplifier performance. The value of CFB should be offset between channel 1 and channel 2 so that the idle switching difference is greater than 40kHz.  Please refer to the Application / Test Circuit.

When you have 6 independent channels, it is difficult to maintain this 40KHz spacing between the switching frequencies. As a result, there might be two different types of problems: 1) audible tones at the difference in the switching frequencies due to shared currents in the ground wiring, or 2) magnetic coupling between the inductors. With careful grounding and adequate spacing you can avoid these problems, but various people have posted about these problems on this forum, so it can certainly happen.

The typical solution for multiple amps is to run them from the same clock. The Sure six-channel TDA7498E uses a synchronized clock for the three chips, and most of the multichannel class D amps use a common clock. You won't necessarily have grounding or coupling problems, but it's good practice to avoid using multiple class D amps on the same power supply.

[donradick]

I certainly do not have the experience or knowledge some of our esteemed colleagues.

But PE sells a Sure 4x100W board based on the TK2050.
Looks like it will deliver more clean(ish) power than the TK7498 boards.

https://www.parts-express.com/sure-electronics-aa-ab33182-4x100w-at-4-ohm-class-d-digital-audio-amplifier-board-sta508-(t--320-335


http://datasheetcatalog.com/datashee...0/TK2050.shtml

[Ron_E]

I built an amp around the 320-325 Sure 4x100W board paired with a 28V PS. I measured 33 watts into 8 ohms before obvious clipping and 55 watts into 4 ohms. For a 4 ohm load that's less than a 3db reduction in output level over 100 watts out. Since your running an active system 55 watts will get you 17db above the drivers efficiency before the amp clips. I run the mids and tweeters on my Cherry Pi speakers with 22 watts/channel.

Ron