• The Vendor's Bazaar forum is for commercial offers and transactions. Only unmoderated members can post here.

    diyAudio provides this forum for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members. Use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

ZOUDIO AIO4CH: 4-channel amplifier with DSP and Bluetooth

just wanted to make a few small suggestions about this product:

the plate accessory should be made of thicker aluminum and/or have more screw holes.
the plate bends when it is screwed in with a gasket.

it's annoying to have to make a separate port through the speaker cabinet for USB config connection
would be nice to have this on the plate

config tool is awkward to work with when developing a filter network.
a mute function for each output channel would be great, for starters.
right now you can only disable both channels of one chip together.
(trying to work around this by putting a -99db gain on one channel)
phase inversion by channel would be nice too

I do want to say this is a nice product at a good price point, thank you.
 
anyone have ideas?
Not without knowing the system architecture, i.e. what amps are connected to what speaker drivers...

An unloaded (open) driven output can ring the output filter at ~40k; unsure if you can just disconnect the other drivers than the one you're trying to measure at the driver terminals, without some ill effect. TI said a class D amp could shutdown sensing something amiss on the outputs with ringing output filters.

I'm sure that depends on how hard you need to drive the outputs to make the measurement. I'd bet the amp wouldnt care at 2.83V = 1W = 88db @ 1M...
 
I have the two outputs of one module (dual mode) powering a CD tweeter and a mid.
The other modules channels (dual mode) are powering a DVC woofer. This is in a unity horn, btw.

I think there is something buggy with the disable function. I can disable the mid/high and just play the woofer section.
When I disable the woofer section to measure the mid/high section, I get no output.
Also, the -99db gain mute kludge works on the woofer section but not the mid/high section.
So only one module responds to these functions properly.

I do have another amp unit that I can swap into the cabinet and test.
Also it's been impossible to swap the speaker output wires because the Phoenix connectors are facing the plate.
(but it just occurred to me that I could install the amp board facing away from the plate...)
 
I just finished my AIO438 and tested it quick with bluetooth, all works.
Already installed configuration tool to office pc and eyed user interface: Maximum delay is only 0,33 ms?
 

Attachments

  • zoudio.jpg
    zoudio.jpg
    259.6 KB · Views: 267
@yairf
Point 6 is now incorporated in the last firmware version.
With command 'default_on enable' you can set the default state after power-on.

@tommus
Thanks for the feedback, I will see what I can incorporate in the future.

So you are saying that if you disable channel C/D, only those channels are disabled,
but if you disable channel C/D, all channels go silent?

The amp is indeed meant to be mounted with its terminals facing away from the plate.
This gives the easiest accessibility.

@Oceanw
Maximum delay is indeed 0.33ms, limited by the DSP hardware.

@AudioResearch
Dynamic EQ works and is dependant on actual volume measured by the amplifier, not phone or volume knob level.
 
The delay functionality is not working at all for me in the AIO438.

I have configured channels A and B to feed bass and tweeter respectively in a 2-way speaker setup. Setting the delay on A channel produces no difference in my measurements. In the following figures are measurements of the system with delays of 0.1ms and 0.02 ms.

interface.PNG

0.1ms.jpg
0.02ms.jpg


There should be a noticeable difference according to my simulations. Additionally, i have tried other values for the delay, setting some delay for all channels, 2.01 and 2.21 firmwares, and different versions of hte configurator software as well. None have produced any results. I suspect there is some delay applied still, since the the mesurements dont seem to match my simulations with 0 delay, but I cannot be certain.

The crossovers seem to work otherwise.
 
Yes that is true. I noticed this a few weeks ago, but have not published this behaviour yet, my apologies.
Since configtool V2.0.0 the amp runs a DSP flow which supports individual input mixers for A/B and C/D, but apparently this flow does not support delays.
If you are not using these individual mixers, which you do not seem to be using, there is a work-around available.
I will send you a PM.
 
I would think the max volume setting is a convenience, giving a "soft" protection from turning the volume level too high. After all, the volume control is just a free spinning knob with no mechanical stop.

Now, if the question would be with a recording having the highest possible PCM encoding value and that limit set to 24 and the volume spun up all the way, does this produce 100% PWM at the peak? I have no idea.
 
Any updates regarding the new config tool and progress with TWS? 🙂
The configtool with the delay issue patched will be released end of this month.
TWS is still pending, but at the moment all developers that are capable of enabling it are quite expensive.

I have the "old" AIO4CH.

I understand that the Max volume is 24 dB.

Is there any risc for digital clipping or is this just for protection of very high effency speakers?

Normally this high gain is not needed, but if your source material is of decent quality it should be no problem.
 
Analogue out from iphone/ipad/iPod is a little low
Why not connect via this amp's prominent feature; BT? You do realize the analog in on the AIO4CH is only 12 bit? Sometimes, I go in that way from a Behringer mixer, with live bass, guitar, vocals and effects - and it sound just fine to me. Which makes me wonder...how can that be. Why am I not hearing this gigantic SQ difference because it's "only" 12 bit or whatever?