3e Audio SY-DAPx002 TPA3250/TPA3255 DSP Amplifier

[somebodyelse]

If you think Royal Mail are bad wait until you see the alternatives. All the couriers I've looked at have a higher minimum fee than Royal Mail's flat rate, and for higher values they'll take a percentage.

There are a number of options for programming the DSP. FreeDSP list a couple of boards in the 'Programmers' section of the board list - these are compatible with AD's programmer board, and probably very similar to what 3E will start selling. If you've got a Raspberry Pi (or another linux board with SPI/I2S) you can hook that up and run sigma_tcp which SigmaStudio will talk to over the network. Beyond that there are software options from HifiBerry (for the Pi) and Auverdion (for ESP) that could be modified to provide online control.

For the STM32 I'd check the unpopulated header to the left of the USBi box header. If we're lucky it'll connect to STM32 pins for one of its programming methods.

 

[Toku]

Hi Somebodyelse.

My initial guess before I received the amplifier was that this would be Rotary Encoder as this would make for a much more versatile solution as you rightly say. I have not measured the pot but I have a message from 3e-Audio confirming it is a traditional Pot with a push switch as this excludes the possibility of adding a remote control for the Volume which is a great shame as the rest of the design seems ready for such an option.

I would love to know what spare inputs are still available on the DSP chip as it would be really nice to include an optical input given that most of the hardware is already present !? This would then bypass the initial A to D conversation.

Clearly the DSP chip can be reprogrammed through the USBi interface with the right interface board. I would also like to know if the Main Processor code can be updated ?!

I can imagine a "plus" version of this amplifier with options such as Remote Volume (and input) Control, Optical IN and USB IN. This would require the Volume "Pot" to be upgraded to a Rotary Encoder.

I already hesitated to order this unit because the room was already full of Chinese digital amplifiers. But after all, I pressed the order button.
If the results are good, I want to buy the version of TPA3255, which will be released in two months. Thank you for your advice.

 

[nodigital]

Where did you purchase the USBi interface and at what price ?

I bought from AliExpress - https://www.aliexpress.com/item/1907907422.html?spm=a2g0s.9042311.0.0.118c4c4dfTtrka
£2.60 + £1.21 delivery. If you select AliExpress standard shipping for an additional £1.40 postage, it arrives in around 11 days

Cable is custom made to comply with the recommendation of 2.2K pull-up resistors between the SDA/SDC pins and the VCC. Once I know things are working I want to try out an ESP8266 for a wireless connection. Its just a darn shame the socket is hidden away and its not too obvious what the correct disassembly is to get to it - They could have put a hole in the case top with screw down plate and simple PCB extension that would have opened up a whole load of DIY extension capability and access to the less technical.

The 3e board may probably be a good choice as it'll be ready to just plug in and is an easier all-at-once purchase.

Have just received mine. On first inspection externally, still looking like QC issues - cracked plastic binding post:

Binding posts feel very cheap and shouldn't have been built to the lowest $ cost, rather it should have been built to differentiate with the masses of other amps out there.

Sounds acceptable though running it all on the desktop: M-DAC to amp but playing through some Mission MX-1's

Its a really wrong setup with muddy middle range so EQ'ing these will see how well the correction works. I've a NAD T778 and with Monitor Audio GR60's in the other room to compare with, which while not really a fair comparison should let me try and adjust to see how close I can get the target sound.

And its very loud so doesn't need much input from the DAC - PC volume at 27% with just two dots lit is plenty loud enough at less than a meter from the speakers

 

[nodigital]

Just paired an Amazon Echo 1st Gen. Firstly, you have to quickly click the volume then hold it for 2s, with the time between the first click and the second hold being very quick. Any delay and it interprets the second press-hold as a power-off

The volume gain when BT connected is massively different. So, driving from the pc, mega loud with everything backed off, lots of gain. Now, over BT the Echo is at 100% vol setting and the amp is at 90% volume to get somewhere close to the line in level I was listening at. With this config, the vol pot would be at max with the amp actually running at minimal power levels - disappointing. Can't imaging they haven't tested various BT devices so wonder if its the BT profile, or something spec related from this old device that's limiting the volume? Will dig out another echo to see if its any better with a newer model. (edit) nope, still the same with 3rd gen Echo.

Hmm, also wonder if its a setting within the DSP for gain levels from that input source?

 

[Bushellj]

As a reference for UK buyers be aware of the total cost of purchase:

• SY-DAP1002 - £66.33
• Delivery - £7.51
• VAT incurred due to customs declared value - £3.19
• Royal Mail handling fee - £8.00 (ouch)

View attachment 72640
So its shipped with unfortunately a $20 USD declaration - This is just over the £15 free limit, but better than paying the 14.70 actual VAT that should be charged...

I've prepped a CY7C68013A USBi board and now awaiting arrival of the amp and example SigmaStudio project file (Available hopefully by end of the week on the 3e Audio site)
View attachment 72569

Do you know if you can still purchase the FreeDSP USBi Adapter board that just plugs into the CY7C68013A board ? Looks as if it would be much easier to use and also includes the LED indicators.

 

[Djsedaw]

Hi, where can you buy the FreeDSP USBi Adapter board, it seems to be sold out?

 

[Toku]

Can I use this USBI adapter?

https://ja.aliexpress.com/item/3280...-4bb8d4f8d4b8-9d9-bf9d9-bf9d4b8-9d9b9d4b8-9d9 searchweb0_0,searchweb201602_,searchweb201603_

 

[nodigital]

Tried a PC BT audio connection and the gain levels from that are at the same as the line in, so my problems with BT are isolated to the connection with Amazon Echos which is very strange. Would be interested if anyone else gets the same problem and if not, what the device is (dot, studio, etc)

Done some further digging and reading the documentation for CY7C68013A states that "These signals must be pulled up to 3.3 V, even if no I2C device is connected". The board I bought has no details on its circuitry but found another supplier of what looks like the identical board that does have schematics: https://www.hotmcu.com/ezusb-fx2lp-cy7c68013a-usb-development-board-p-281.html

The schematic shows that it does already have 2.2K pull-ups in place. What this means is that its should be a simple straight wire connection from the board SDA/SDC/GND pins to the amp using female-female dupont wires (Such as https://www.aliexpress.com/item/32390991500.html?spm=a2g0s.9042311.0.0.27424c4dCt7FkK ). I'll verify the connectivity

 

[nodigital]

So, it appears to work with plain wires.

In sigma studio, created a new project, drag on the usbi and adau1701 blocks and link them. Power up the board and SigmaStudio detects the board and the USB block goes from red to green

Connect the cable to the amp, leaving it unconnected from the board - There's mains very close by to that data socket so worth keeping hands well away. Power up the amp then connect the cable to the board.

The cable is made of some cheap female-female dupont wires put into larger female sockets from a pack I already had
https://www.aliexpress.com/item/32626426260.html?spm=a2g0s.9042311.0.0.27424c4dCt7FkK 8 pin for the board and a 10 pin for the amp
(On the dupont cables, you life up the tab on the plastic housing and slide it off to reveal the metal connector, then push the connector into the new socket housing where it clicks and latches)

You can also see on the photo above that the J2 jumper on the board is open, this is important for the initial config and driver setup of the board and should remain open in use.

Back in SigmaStudio click on the bottom IC1 tab for the config. In here we can verify that its connected by reading the registers in the DSP - Note its reading values only, it doesnt know what was programmed. Click Read All and it updates the display - 188ms to read the config and it shows that MP1&MP2 are high (whatever that means...)

That's as far as I want to go at the moment until I've got the sample project and know more about configuring and what not to do...

Trying out the a test setup with a REW output and the autoeq resulted in the filter calculation process hanging - Again, need more understanding and time to experiment.

 

[Bushellj]

Tried a PC BT audio connection and the gain levels from that are at the same as the line in, so my problems with BT are isolated to the connection with Amazon Echos which is very strange. Would be interested if anyone else gets the same problem and if not, what the device is (dot, studio, etc)

I also connected my echo dot to the amplifier via Bluetooth and similar issues to you. Volume levels very low unless you turn the echo up to maximum. If the echo disconnects for any reason it comes back on at full volume - Not a nice experience.

Also the quality of sound from the cho using Bluetooth was dismal !!

From my Mac Mini I have been playing with various ways to connect to the Amplifier - Bluetooth (AAC mode as I can’t get the APTX mode to work on the Mac), connecting through an Airport Express to RCA inputs. The Gain in these circumstances is the same on the Bluetooth Connection and from the Airport Express. You can just switch between the two connections with a single click on the MAC to compare the difference.

The sound quality using the RAC inputs (via the Airpoty express) is much much better than the Bluetooth now I have played with some good quality source material. So the Bluetooth is only really useful for the occasional mobile phone usage

 

[Bushellj]

Hi, where can you buy the FreeDSP USBi Adapter board, it seems to be sold out?

Let's see what 3e-Audio comes out with as an option for their amplifiers. Should give a better idea of the requirements although programming these DSP chips is common practice. Pity the software doesn't run on a MAC !!

The FreeDSP USBi adapter appears to be only released as a set of files so you could have your own made. Clearly there were some made available through various DIY electronics suppliers together with the BOM for the components required.

Maybe we should set up a Group Buy if becomes something that everyone wants.

There are at least four releases of this board - v0.1, v0.2 (as seen above), v0.3 (allowed use of SMD components as well as traditional "Thru Hole" components and then v0.4 with some enhancements but I can't find the specific details !?

Also while on this subject there are at least two major releases of the CY7C68013A board. This most obvious difference is one has two jumpers and the other only has one jumper. The headers on the two versions are different so be careful !!! The one with two jumpers needs modification if you want to use it with the FreeDSP USBi "shield" From memory you have to cut the Ground and VCC pins or you will damage the board.

This is a useful link for instructions and the components required :-

http://wiki.op-electronics.com/installare-e-assemblare-opusbi-new/

 

[Bushellj]

Here is another link to the FreeDSP Documentation.

https://docs.google.com/document/d/1rKkCl3gyePwX15c3sjEIKP6Bb7CO77hgRAXtbYz5hA0/edit

 

[nodigital]

I also connected my echo dot to the amplifier via Bluetooth and similar issues to you. Volume levels very low unless you turn the echo up to maximum.

Reading around it looks like Echo's have problems with some bluetooth speakers, possibly related to a firmware update last year. Could be a problem with certain BT chips or a spec compliance problem, etc. No official acknowledgement nor any consistent fix described. Quality from streamed content via the Echo isnt great to begin with unless you're taking a hd music subscription. Listening to radio streams is fine, albeit quiet.

If you want to use a Free USBi board then buying the CY7C68013A with the single jumper is easier otherwise the pins indicated need cutting. The schematic shows that if not, the VCC and ground will be shorted out Would like to understand if that board is actually needed - Aside from the 10-pin header it has a pair of 2N7000 for the SDA/SCL so maybe presuming its either isolation or driving a longer cable (I2C could do 2.5m length) and at the moment its working ok without.

Presuming a board from 3e will be similar combo of CY7C68013A and 10pin plug as an all-in-one as an easier solution

 

[Bushellj]

Reading around it looks like Echo's have problems with some bluetooth speakers, possibly related to a firmware update last year. Could be a problem with certain BT chips or a spec compliance problem, etc. No official acknowledgement nor any consistent fix described. Quality from streamed content via the Echo isnt great to begin with unless you're taking a hd music subscription. Listening to radio streams is fine, albeit quiet.

If you want to use a Free USBi board then buying the CY7C68013A with the single jumper is easier otherwise the pins indicated need cutting. The schematic shows that if not, the VCC and ground will be shorted out Would like to understand if that board is actually needed - Aside from the 10-pin header it has a pair of 2N7000 for the SDA/SCL so maybe presuming its either isolation or driving a longer cable (I2C could do 2.5m length) and at the moment its working ok without.

Presuming a board from 3e will be similar combo of CY7C68013A and 10pin plug as an all-in-one as an easier solution

I think the basic problem with the Echo is the bluetooth is very basic so regardless of the quality of the music source the Bluetooth stream will be highly compressed and very lossy !! The Echo is built down to a price so I guess the Bluetooth chip will be very basic ??

On the USBi Programmer - The later revisions (v0.2 onwards) of the FreeUSBi Shield work on either versions of CY7C68013A boards as they ignore these pins.

Or you could opt for this from Sushi Bits which is basically the Programmer and FreeUSBi board and in one box :-

https://www.tindie.com/products/maxtch/sigmalink-usbi/

 

[Bushellj]

This was the USBi Programmer Board that 3e-Audio used to provide with their DSP based Tone control board.

Although several pictures of the board on several sites - nobody seems to sell it anymore !?

Hopefully 3e-Audio will soon publish what they plan to supply for the new Integrated Boxed Amplifier with DSP.

 

[tmortimer]

Got mine a little while ago. The audio is pretty tinny, but I wasn't using the best speakers + cables i had just added banana plugs to in a hurry so i'll hold off on forming an opinion on the sound quality.

However the BT is something that seems to be a problem. My phone seems to pair just fine, "aptx" logo shows up. But I get no audio and then eventually my phone disconnects. Does anyone have any thoughts there?

 

[nodigital]

Todays experiment is a wireless connection from SigmaStudio. Thanks to the binary and help from https://ez.analog.com/dsp/sigmadsp/...rface-for-sigma-studio-using-esp8266-or-esp32 Flashed a Wemos D1 with the image off the page. Reset and the Wemos starts up as an access point. Connect to it and point the browser at http://192.168.0.1/connect?ssid=xxx&password=yyy to get it to connect to my network. It restarts and takes a while, upto 90s to connect. Once connected, you have to discover the IP (Not great as there's no display on it and it doesnt have a host name to reference to it).

Connected up a cable, this one has the 2.2k from 3.3V to the SDA/SCL. The additional wire is to the RESETB of the ADAU1701 which is connected on the amps USBi socket (confirmed by 3e) and needed by the ESP8266 firmware to reset. In some of the USBi schematics it shows a 5V on the socket at pin4. Unfortunately it's not present (3e confirmed the omission) which means its not possible to power an ESP8266 straight off that connector

Connected up and within Sigma Studio connect the TCP component to an ADAU1701

Right click on the channel, set the IP address and open the connection (Note, it doesnt seem to give an indication its connected unlike the USBi)

Then read all the config.

Note there's a bug in SigmaStudio as highlighted at https://ez.analog.com/dsp/sigmadsp/...-tcp-ip-read-from-dsp-with-temporary-solution that doesn't allow the connection of the E2Prom which is essential for persisting

Anyhow, begins to open up some possibilities.

I've also written a small exe to simplify the conversion from REW output into the MLSAA file format needed by the SigmaStudio AutoEQ - The alternative was faffing around with a python install and hacking a code sample. The AutoEQ which uses that MLSAA is interesting as it takes a long while to calculate and looks to work ok on fixing my speaker deficiencies. My concern at the moment is the amount of 'instructions' it'll take and the number of filters per channel. The sample project will give an indication of how many filters can be applied before it runs out of the processors 1024 total...

 

[nodigital]

However the BT is something that seems to be a problem. My phone seems to pair just fine, "aptx" logo shows up. But I get no audio and then eventually my phone disconnects. Does anyone have any thoughts there?

Tried my phone and see exactly the same problem, so is a consistent bug.

What's odd is that its a well known Qualcomm QCC3008 bluetooth chip from 2017, very sophisticated and with lots of features and capability (You can download the datasheet here). So not sure if its a firmware problem or some odd implementation issue. Either way the only BT audio I can get is from the low volume Amazon echo. I'll raise the problem with 3e Audio though don't know how a firmware fix can be applied to the QCC3008 (I think it would have to be using SPI not I2C so maybe cant do it over the USBi need clarification).

The BT module has a strong resemblance to https://www.aliexpress.com/item/1000005388492.html
Wonder if its the binding with multiple bluetooth devices. Short press then press and hold until the full light ring blinks which I presume is a reset but doesnt make a difference, still no audio

 

[timmaio]

Tried my phone and see exactly the same problem, so is a consistent bug.

What's odd is that its a well known Qualcomm QCC3008 bluetooth chip from 2017, very sophisticated and with lots of features and capability (You can download the datasheet here). So not sure if its a firmware problem or some odd implementation issue. Either way the only BT audio I can get is from the low volume Amazon echo. I'll raise the problem with 3e Audio though don't know how a firmware fix can be applied to the QCC3008 (I think it would have to be using SPI not I2C so maybe cant do it over the USBi need clarification).

I have been listening entirely over BT for the last 3 days with my phone primarily(Samsung Note 10) and some(not much) with my laptop. I have had no issues with either. Sounds quality seems fine, it is on my desk so it's hard to evaluate critically from such short listening distance.
One thing I didn't recall seeing in a specs is powersave, after some time with no audio over BT the amp powers off and requires a single short push of the volume knob to turn it back on. This is fine for a desktop amp, would likely be annoying if I had this installed out of the way.

 

[Bushellj]

Tried my phone and see exactly the same problem, so is a consistent bug.

What's odd is that its a well known Qualcomm QCC3008 bluetooth chip from 2017, very sophisticated and with lots of features and capability (You can download the datasheet here). So not sure if its a firmware problem or some odd implementation issue. Either way the only BT audio I can get is from the low volume Amazon echo. I'll raise the problem with 3e Audio though don't know how a firmware fix can be applied to the QCC3008 (I think it would have to be using SPI not I2C so maybe cant do it over the USBi need clarification).

The BT module has a strong resemblance to https://www.aliexpress.com/item/1000005388492.html
Wonder if it's the binding with multiple bluetooth devices. Short press then press and hold until the full light ring blinks which I presume is a reset but doesn't make a difference, still no audio

.


I have been playing with the Bluetooth functionality for some time. First though I need to clarify that I only have Apple Devices and an Amazon Echo that I use as a music sources - All, I believe are using Apple's AAC Codec rather than AptX other than the Amazon Echo Dot which just uses the basic SBC Codec. I have a Mac Mini, iPhone and iPad. I have been unable to get the Mac Mini to use AptX as I can't get Apple's Bluetooth Explorer to work on the current version of the MacOS ??? Apple IOS Devices do not support AptX - Only AAC !!

They all play fine - I think two devices can be connected concurrently although only one device can play at any one time - You do need to disconnect one of the devices if you want to play music from a third / forth device.

The Amazon Echo is very quiet in comparison to the other devices and the Sound Quality is "crap" so I have given up with this as a source which is a pity as it was originally my main reason for purchasing this amplifier with the built in Bluetooth capability. Anyone know if any of the Amazon Echo devices provide a higher quality Bluetooth Output ??

The Bluetooth takes priority over the RCA inputs and switches back neatly to the RCA inputs when there is no signal on the Bluetooth source. That bit has been nicely implemented.

I am not so happy that the Amplifier automatically power down over a short period of time when there is no signal present. Although this might seem a nice feature on paper it is not very practical in the real world ! I tend to leave my amplifiers on 24/7 - Digital amplifiers use very little power when there is no signal !

Once we have access to the DSP and the capability to re-program it - Maybe we can disable the command to the main CPU to put the Amplifier into sleep mode as it is dependant on a signal (or lack of a signal) through the DSP !!