AudioEngine A2 clone - possible?

[ememm]

My friend has an Audioengine A2 and we all love its sound. It may not be the ultimate audiophile speakers out there but for the size and price, I feel it has immense value.

Lately I've been thinking of a diy project, something as close to A2 in size and SQ, but at lower cost. I have done a basic search in this forum, but could not find any thread with similar objective. (Please point me to any, if you know)

So basically I am looking for:
- a small 2-way vented design, close to 1.5 litre(0.05 cuft) volume - size is a big concern but could go a bit bigger than A2s.
- budget ~200$ - assume woodwork and amp separate
- clean, punchy bass (bass extension is not a priority)
- balanced sound (if not, prefer warm rather than bright)
- not really bothered about the soundstage, but resolution between instruments is a must.
- planning to use a topping amp to drive them.

I would like to have your expert opinion on such a project.
Is it even possible? I'm sure somebody must have already attempted something like this. What is the smallest hi-fi diy speakers possbile(in this budget)?

I am a noob and this will be my first attempt at DIY - so please go easy on me. I am from India, so driver availability might be a problem, but don't let that stop you from suggesting. Also, I am confident that I'll be able to understand and use cabinet design softwares if required, with some guidance.

Any kind of constructive comments are welcome.

Thanks in Advance,
EmEmm

 

[Phenoholic Anonymus]

Hi EmEmm,

The trick of the audioengine speaker maybe the active crossover / amplifier powering it. This makes it easy to extend the bass, x-over real smooth and buckle out some nasties of the drivers.

So in my opinion, the question would be, is working with active crossovers and amplifiers fit as your first project?

 

[ememm]

That would complicate things a bit, I was hoping to focus on the driver and cabinet part. But assuming that I am ready to take up the amp/active crossover as the second phase of the project, can you suggest suitable drivers/cabinet design?

 

[Phenoholic Anonymus]

I am of the passive x-over persuasion. So i cannot help you with driver selection for an active. Highly speculative: you seem to like the clean sound of the woven carbon / glassfiber woofer. Take a look at the tymphany peerless glassfiber woofers and maybe a wavecor or sb acoustics large dome tweeter. But this is still thinking in passive x-overs, keeping your x-over point options as wide open as possible, so you can use less parts in the x-over. But this is strictly my personal taste or belief.

- edit - Wait a minute, will look for drivers close or at least the size of Audio engine 2. tweeter Wavecor TW 022WA02, Hi-Vi T20-8, Vifa NE19VTS-04
Woofer peerless TC9FD10-08

Take a look at mini dsp that would be a good place to get info about active x-over.

 

[ememm]

Will do. Thanks. I'm also open to both active and passive options, within the budget ofcourse. I want to get as much opinions as possible before I start getting my hands dirty

 

[Lojzek]

Look at this atomic ant!

Tang Band W3 1876s 3 Mini Subwoofer 264 909 | eBay

Tang Band W3-1876S 3" Mini Subwoofer

Parts Express Ebay store does not ship to India.

 

[xrk971]

You can do it for $200 with drivers, active DSP XO, class D bi- amps. As an example, look at what I did here:
http://www.diyaudio.com/forums/full-range/265915-hyperfast-hypercube-based-2-way-4.html#post4147908

You could use just a rectangular prism shaped box but drivers and electronics would be similar. Going with a Vifa TC9FD as a tweeter XO low at circa 300Hz has huge advantages from imaging and phase coherence. A 2-way with a XO at 1kH cannot match it for phase and time alignment.

Cost breakdown

MinDSP $80
Plugin 2x4 advanced XO for miniDSP $10
TPa3116d2 class D amps $30 for 4 channels (40 watts ea)
$12 for two 19v 4.6a SMPS
5in Tang Band W5-876SE woofers $20x2=$40 (or equivalent Peerless woofer)
Vifa TC9FD $12x2=$24
$196 total.

http://www.diyaudio.com/forums/full-range/265915-hyperfast-hypercube-based-2-way-4.html#post4147908

If you go sealed with Linkwitz transform in miniDSP the sound will be superior to bass reflex for group delay and transient dynamics.

Impulse response:

Group Delay:

 

[Phenoholic Anonymus]

XRK's idea is very good. But i reckon passive version is cheaper 🙂 Both x -over options will work i think, and it's much easier to get an x-over right in the 300 hz region

 

[dumptruck]

Multiple W4-1720 designs come in just a little overbudget, and can be built in an enclosure just a hair larger than the A2, or add 1-4 liters and port for loads more bass. At these sizes, the port eats up a lot of volume, so I admit this is really not so close to the request. Example:
file:///C:/Users/John/Documents/Audio%20Stuff/Napoleons.pdf

Anyhoo, A2 measured amplifier response from Stereophile:

and speaker response with its amp (IIRC, they usually get the port summing a bit wrong, but we get the idea and it's in line with some other measurements I found - the resonant dip in the mids seems to be a real thing):

If you added a 100Hz port to the small sealed W4-1720 cabinet you'd get a better version of the same low end response without EQ, although you'd be losing ~4dB clean output with average tunes - doing it actively to the sealed one with a high pass would make more sense, and I'd skip the bass hump and shoot for a little more extension (or forget it and use a sub) regardless. Anyway, my main point was to post the measurements. This speaker should not be that difficult to beat for $200 without amps, I don't think. That's a pretty healthy budget.

 

[xrk971]

This?

Or this?

Hmmm...

 

[dumptruck]

That's graph scale abuse - yours only looks marginally better because of the midrange anomaly on the A2, and the voicing above 1k looks less pleasant (yes I know it's measured too close, but it's too hot around 3.5k, and there's no reason for that hump-shaped lower mid response either). I'd overlay them to prove the point, but it's a moot point since you could do better with the DSP.

 

[xrk971]

You are probably right if they were overlaid - the variation could be flattened but I think it is fairly flat and perfectly suitable for most listening. I am surprised that the A2 has a 6dB dip at 1khz. Almost looks like the XO point phase is not aligned (if XO is there). But the bass response and group delay on my sealed speaker is actually better.

 

[ttan98]

Hi,

A2 is so cheap there is no point cloning provided you want to achieve these objectives:

1. Design a pair better than A2 for the same price, very difficult to achieve.
2. Learn how to design a pair comparable in sound quality or better than A2.
3. Experience the sound from A2, I believe it is cheaper to buy a pair.

I designed a pair of LXMini "clone" using totally different drivers so that I could experience the sound reproduction. The cost of my pair ended up to be cheaper than that if I buy the LXMini drivers and plans separately.

 

[dumptruck]

You are probably right if they were overlaid - the variation could be flattened but I think it is fairly flat and perfectly suitable for most listening.

I'm sure it's fine, DSP just kinda begs for nitpicking.

I am surprised that the A2 has a 6dB dip at 1khz. Almost looks like the XO point phase is not aligned (if XO is there). But the bass response and group delay on my sealed speaker is actually better.

You know what? I said "resonant" because that's what Stereophile thought, but I changed my mind. The baffle step down is right there, disappearing shortly below where it starts because of the amp boost. I don't know why they didn't just have the amp fix that. The crossover looks like it's probably asymmetric, acoustic point around 3.5k, looks like it could be better.

(btw, I actually like Audioengine, just picking at 'em because that's the idea of the thread - doing better)

 

[Brisso57]

You could always take the tack of using a minidsp to design your passive XO, then sell the minidsp to recoup some funds.

 

[xrk971]

Anyone who uses a miniDSP and active XO will see how easy and fast it is to make a good sounding speaker and how powerful it is that they probably won't sell it. I certainly won't trade mine for passives.

 

[ememm]

Looks like this discussion quickly went towards the active XO - miniDSP route.
I would take that route only if similar results are impossible to achieve using passive xo.

Can someone please suggest some good proven drivers which can be used for such a project?

 

[Phenoholic Anonymus]

I forgot to mention the cheap 3/4 inch dayton neo dome. Measures well, easy to handle. You are right DSP is all the craze now. People are forgetting DSP cost + 2 amps and all the extra's....

 

[Mr. dB]

Hi EmEmm,

The trick of the audioengine speaker maybe the active crossover / amplifier powering it. This makes it easy to extend the bass, x-over real smooth and buckle out some nasties of the drivers.

So in my opinion, the question would be, is working with active crossovers and amplifiers fit as your first project?

The A2 does not have an active crossover. There is a stereo power amp in the Left speaker, the Right speaker is connected to the Left with a length of 2-conductor zip cord.

 

[xrk971]

I forgot to mention the cheap 3/4 inch dayton neo dome. Measures well, easy to handle. You are right DSP is all the craze now. People are forgetting DSP cost + 2 amps and all the extra's....

When I was looking at the Manzanita speaker (2 way passive OB), the parts for the XO were more expensive than a miniDSP and two class D amps ($120). It depends on the XO parts quality I suppose, but the Manzanita was not using anything particularly "boutique". If you include the ability to re-use the miniDSP and bi-amps indefinitely for new projects in the future, the cost benefits will surely be realized quickly. Just my two cents...