Please use this thread to share information from your HiFiBerry Amp beta test.
Use this installation guide to install the hardware and configure your system:
http://www.hifiberry.com/installing-the-hifiberry-amp/
Date
Votes
118 comments
-
Roman Bosshard Hey Daniel
How can I become a beta tester for the Amp? Is there already an estimated release date for the device?
Regards, Roman -
Kent Hansson Hello,
I just received my Amp module, fast delivery to Sweden (2 business days). I have installed it on a Pi and it booted nicely with 12v power to the barrel jack and ordinary raspian. I am ready to configure and test the sw part but i can not find anything to download yet so i assume it is work in progress. Let me know when drivers/updates is available.
BR/Kent -
HiFiBerry team Hi Ken,
we didn't think delivery will be so fast. Here is the guide how to configure it:
http://www.hifiberry.com/installing-the-hifiberry-amp/
Best regards
Daniel -
Damian Flynn Great,
I also received 4 modules yesterday in Ireland. the 4 headers are soldered, now time to play. -
Antony Hubbard I've installed the Amp today on one of my Pis, in place of a Hifiberry Digi. The Pi has Squeezelite installed and is used as Logitech Media Server client. I followed the installation guide, then connected my Jamo Concert 8 speakers.
It all worked perfectly. I need to get the speakers on some stands and have a proper listen. Will let you know my impressions.
Thanks for an amazing product.
-
Kent Hansson Hi,
Thanks for the instruction Daniel, i followed it. But i also had to remove the lineblacklist i2c-bcm2708
from the blacklist file in order to find the card. Now it is playing nicely and i will try it out.
BR/Kent -
Antony Hubbard How does the setting of the volume control work? Is the setting of 74 chosen for a particular reason? Is that equivalent to 0 dbFs?
Are you planning to publish some audio performance tests of the Amp? -
HiFiBerry team The range of the used chip is just mapped to the ALSA range of 0-100. 74 is 0dB. We will publish performance data, but not very soon as our measurement system has to be upgraded with the right filters first.
Best regards
Daniel -
Aaron Baff I'm having a problem getting the amp to load, getting the following error messages in syslog, and still see the bcm2835 alsa device.Aug 3 11:47:04 pi kernel: snd-hifiberry-amp snd-hifiberry-amp.0: ASoC: CODEC tas5713.1-001b not registered
Aug 3 11:47:04 pi kernel: snd-hifiberry-amp snd-hifiberry-amp.0: snd_soc_register_card() failed: -517
Aug 3 11:47:04 pi kernel: platform snd-hifiberry-amp.0: Driver snd-hifiberry-amp requests probe deferral -
Aaron Baff Oh, and I also didn't like the idea of the update script modifying things such as the kernel image & modules without also backing them up, so I created my own script. It also applies the module blacklist/load and new /etc/asound.conf (with a backup as well).#!/bin/bash
if [[ $EUID -ne 0 ]]; then
echo "Must run this script as root"
exit 1
fi
SCRIPT_LOCATION="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
for DIR in "/lib/modules" "/lib/firmware"
do
if [ !-d $DIR ]; then
echo "$DIR exist, moving to $DIR.BAK"
mv -n "$DIR" "$DIR.BAK"
fi
done
pushd /
tar xvfz $SCRIPT_LOCATION/modules.tar.gz
popd
cp $SCRIPT_LOCATION/kernel.img /boot
for MOD in "blacklist spi-bcm2708" "blacklist i2c-bcm2708"
do
echo "Appending $MOD to raspi-blacklist.conf"
echo "$MOD" >> /etc/modprobe.d/raspi-blacklist.conf
done
for MOD in "snd_soc_bcm2708" "bcm2708_dmaengine" "snd_soc_tas5713" "snd_soc_hifiberry_amp"
do
echo $MOD >> /etc/modules
done
if [ ! -e "/etc/asound.conf" ]; then
echo "/etc/asound.conf exist, moving to /etc/asound.conf.BAK"
mv -n "/etc/asound.conf" "/etc/asound.conf.BAK"
fi
echo 'pcm.!default {' >> /etc/asound.conf
echo ' type hw card 0' >> /etc/asound.conf
echo '}' >> /etc/asound.conf
echo 'ctl.!default {' >> /etc/asound.conf
echo ' type hw card 0' >> /etc/asound.conf
echo '}' >> /etc/asound.conf
echo "SYNCing, please wait"
sync
echo "YOU MUST RESTART NOW" -
HiFiBerry team Hi Aaron,
thank you for your script. However there is a problem with it: According to your script you add a "blacklist i2c-bcm2708" line to the blacklist file. This line has to be removed, not added.
Remove the i2c blacklist from /etc/modprobe.d/raspi-blacklist.conf and try again.
Best regards
Daniel -
Aaron Baff Ah, ok, missed that.
However, now I can't get my RPi to power up with the Amp card on it :( I measured with my volt meter, and there's 12v going in. Don't see the RPi LEDs light up, and when I measure on the 5v pins on the main header, nothing. No volts. Only thing I've done was to unhook it, take it to a friends, then bring it home tonight and plug it in. Just to confirm, in case I somehow mixed it up when plugging it back in, the inner pole is +, and the outer is -, correct?
EDIT: The RPi does power up and boot properly when connected solely to the standard 5V usb power, with or without the Amp physically on the headers of the RPi. -
HiFiBerry team Hi Aaron,
yes, inner pole is +12V, outer pole is ground.
please unplug the Amp from the Raspberry. Connect 12V to the Amp and measure the voltage over C303 (at the corner of the board). If this is not 5V, go on and measure the voltage over one of the electrolytic capacitors. It should be 12V. If this is not 12V, the fuse (F301) might be blown.
Best regards
Daniel -
Aaron Baff So, the voltage across C303 (opposite corner of the board from the power plug) is 0. I measured across each of the capacitors (all look intact with no problems) and I get 0V. When I measure across F301 (right next to the power plug) I get 12V (12.8 to be exact). It has been unusually humid here recently, could the humidity be affecting the caps? Should I try replacing them one at a time? -
Aaron Baff Ok, so if I were to try replacing them, would 35v470uF work? I found http://www.radioshack.com/product/index.jsp?productId=2102506 at a local RadioShack. -
HiFiBerry team Hi Aron,
no, this looks like the fuse is blown, which is not a good sign. Do you have a lab power supply with current limiter?
Best regards
Daniel -
Aaron Baff But I still see the 12V when I measure across the fuse.
No, I don't have a lab power supply. I have it running off power right from a 12V SLA AGM battery, although I do have a 7.5A fuse in the middle, along with other 12V electronics running off of the same battery. Hmm...only thing different that happened was I had the battery on a charger, so perhaps the charger was outputting a bit too much which popped the fuse. I'll be much more careful in the future, and not have it on the charger when having anything plugged into the battery.
As for replacement, would http://www.digikey.com/product-search/en?pv66=6&pv1292=9&FV=fff4000a%2Cfff8003d&k=smd+fuse&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25 work? Or can you point me to another one from digikey, or the correct one I should use on Amazon US, Digikey, or other company that is in the US? -
HiFiBerry team Yes, the 12V over the fuse is a problem. There should be almost 0V across the fuse. The fuses in you link should work as a replacement.
Best regards
Daniel -
Antony Hubbard Hi Daniel
Unfortunately I have the same problem as Aaron, but I have no voltmeter to diagnose. Software I'm ok with. Hardware, not so good.
I am using this power supply : http://www.powersuppliesonline.co.uk/power-adapters/60w-12v-5a-iec320-c14-desktop-power-supply-ea1050as-12.html
Should I replace the fuse?
Thanks
Antony -
HiFiBerry team It seems, that the fuse is a little bit too weak for the peak current when powering on.
Replacing it by a zero-ohm resistor should be a quick fix. If you have this problem, try this resistor:
http://www.digikey.com/product-detail/en/RL0603FR-070R01L/311-.01QCT-ND/3885425
We can also provide these resistors or replace them on the board. However for a replacement, you have to send the board back to us.
If you have this specific problem, get in touch with at by email info@hifiberry.com to see, what is the best fix for you.
Best regards
Daniel -
Aaron Baff Could I put a 0-ohm resistor inline in the power plug instead of on the board? Or are you saying replace the 3-amp fuse with a 0-ohm SMD resistor instead? Or can I do both? Replace the 3-amp fuse, and put the resistor inline with the power plug. -
HiFiBerry team No, just replace the fuse by a resistor. It will also act as a fuse, but don't blow at a small current peak. -
Thorsten Hecht Hi
I got my Hifiberry Amp delivered and also shot the fuse. Just fixed it with another fuse.
I also took a powersupply like a wall plug to try the worst case. Later i will try a battery to get this thing outside :-)
I am also into Rc Helis and need some decent music playing outside while flying. This thing is quite a fun to listen in comparison with these bluetooth thingees...
Have some transmissionline loudspeakers playing and my first impression is very nice and i like it...
When will thwey be available ? I need more...
Regards Thorsten -
HiFiBerry team Hi Thorsten,
thank you for the feedback. We will deliver the next batch of beta test boards (with a changed fuse) beginning of next week. I might have a few available. How many do you need?
Best regards
Daniel -
Paolo Canali Hi Daniel,
I received my Hifiberry Amp today. I downloaded the current Raspbian sd card image and upgraded everything (apg-get update , apt-get upgrade, rpi-update). After that, I followed the installation procedure (update to the kernel, update to the module blacklist and startup modules).
Near the end of the Raspbian boot, I get the following messages:
Loading kernel module snd_soc_hifiberry_amp
snd-hifiberry-amp snd-hifiberry-amp.0: ASoC: CODEC tas5713.1-001b not registered
snd-hifiberry-amp snd-hifiberry-amp.0: snd_soc_register_card() failed: -517
platform snd-hifiberry-amp.0: Driver snd-hifiberry-amp requests probe deferral
aplay does not list the Hifiberry Amp.
Could you point me on the right direction to solve the issue?
I still need to connect the speakers, because my speaker wire is thick and does not fit in the very small screw terminals of the board (I suggest to use a bigger connector on final board). I will arrange something as soon as I will get the software working. -
HiFiBerry team Hi Paolo,
please run "sudo i2cdetect -y 1" and post the output.
To do this you need to add the line
i2c-dev
to /etc/modules and install the i2c tools:
sudo apt-get install i2c-tools
Best regards
Daniel -
Paolo Canali Hi Daniel,
Thanks for the help. I solved the issue by removing the i2c-bcm2708 line from the Raspbian module blacklist file /etc/modprobe.d/raspi-blacklist.conf. I previously removed only the spi-bcm2708 line because I was unaware that i2c is also needed. You may want to update the "Installing the HiFiBerry Amp for beta testers" post (step6). -
Aaron Baff Ok, so, tried replacing the fuse with another fuse (linked previously), however it must have blew again. After replacing it with the previously linked resistor, it's not up and running fine.alsamixersees it just fine, I can play music, but unfortunately can't control the volume. Play music, yes. Access the mixer, nope. Worked fine with the previous setup, and I'm tried a few different device/mixer_device settings. -
HiFiBerry team Hi Aaron,
can you try just using amixer like this:
amixer sset Master 60% or
amixer sset Master 65%
Does this change the volume?
please post your output of "aplay -L"
Best regards
Daniel -
Paolo Canali I just finished my initial sound test, with a 12v bench power supply current limited to 3A, to sidestep the "weak fuse" issue (i will replace the fuse later). The HifiBerry AMP sounds good. I already listened a few single-chip Class-D amplifiers before, and this one is near the top. At first, I was concerned for the lack of computing power of the Rapsperry Pi, but CPU is below 50% even during FLAC reproduction with VLC; mp3 network streaming at 256 Kbps is flawless. Sound stutters if I intentionally starve the CPU by forcing a screen scroll in Midori. If I keep doing this long enough, the sound driver finally crashes and the board remains muted until I reboot the Raspberry. This is hardly a issue in real life, but a graceful restart of the driver after a failure would be a plus.
I have a question about the power supply. I downloaded the IC datasheet from the manufacturer web site, and I found that the audio power output is proportional to the power supply voltage. At 12V the 1% THD output power (over 8 ohms) is 2x7W only, and this explains my impression of somewhat insufficient power. The audio IC manufacturer maximum recommended supply voltage is 26V (with 30V absolute maximum).
A 24V power supply would be ideal; it is easy and inexpensive to get quality 24V power supplies, and it will enable to get the 2x25W full power output from the audio IC.
The 18V suggested in the HiFiBerry AMP FAQ is a nonstandard value that basically will force to select a expensive adjustable supply or a cheaper laptop power brick (with bad ripple and regulation characteristics). 18V is not enough to get even 2x20W at 1% THD from this audio IC, while I was expecting a 2x25W power output.
I would like to try rising the supply voltage to 24V. This value is fully supported by the audio IC, but are there any additional design constraints that forced to specify 18V in the product FAQ? Daniel, can you advise about this issue?
While browsing the audio IC datasheet to check the I2C functionality, I noticed that this chip is a fairly advanced component with programmable Dynamic Range Compression and speaker equalization features. Is it possible to enable these features? Is there some command I can try?
Please sign in to leave a comment.