via Hackaday: Moonbounce Music
![](../themes/icons/grey.gif)
Weβve recently added a new room to the Matrix HAM Radio Space for Digital Voice modes as this was an area of interest that didnβt really fit into any of the other rooms.
The new Digital Voice room has attracted a lot of attention from members, with a lot of the focus being on the AllStarLink system. Michael, DK1MI built an AllStarLink node in the cloud for us all to use for Matrix Nets and so I decided I had to get in on the fun.
The Jumbospot SHARI SA818 Amateur Radio AllStarLink Radio Interface was originally designedΒ byΒ N8ARΒ andΒ implementsΒ aΒ RaspberryPi 2/3/4Β hostedΒ AllStarLinkΒ nodeΒ usingΒ aΒ NiceRFΒ SA818Β embeddedΒ VHF/UHFΒ radioΒ module and sound card.
The two USB connectors on the SHARI device are position such that they plug into two of the available 4 USB ports on the RaspberryPi without the need for cables. This keeps the whole solution together in one neat package.
Before you start you will need to obtain a node number and secret (password) from the AllStarLink Portal. To get this you will need to provide proof to the AllStarLink administrators that you are a licensed Amateur Radio (HAM) operator. This is done by uploading a copy of the first page of your HAM licence to the website for the admin team to check. This can take 24hrs to be completed so make sure you get this all done before trying to build your node. You cannot build a node successfully without a node number and secret.
Of course you will also need a transceiver that can operate on the 438.800Mhz frequency or other frequency of your choice on the 2m or 70cm HAM band.
You will also need to open port 4569 on your internet router and setup port forwarding to the IP Address that you will be using on your RaspberryPi node. Itβs important to use a static IP Address on your RaspberryPi.
There are quite a few different Linux based operating system (O/S) images that are available for the RaspberryPi devices that have been specifically tailored for the AllStarLink node and include all the necessary software and library packages out the box.
I decided to use the Raspbian GNU/Linux 10 (buster) based distribution as it is based on the very stable and reliable Debian Linux distro. You can download the exact version I am using from the Raspbian link above or directly from my website here.
Once downloaded you need to burn the ISO image onto a suitable SD card for your RaspberryPi. I use BalenaEtcher as itβs extremely quick and reliable at burning ISO images to SD cards.
Of course if you are a hardline Linux command line junkie you can always use dd to create the SD card.
Once youβve got your O/S onto your SD card, slot it into your RaspberryPi making sure your SHARI device is connected to the two USB ports and then power it up. Make sure you have a good PSU for the RaspberryPi as the two devices together draw around 3A of current during the transmit cycle. (I use a 3.6A PSU from Amazon).
The default login for the Raspbian O/S is shown below. Login via SSH and configure your RaspberryPi for your local network. Itβs important to use a static IP Address configured either directly on the RaspberryPi or via DHCP in your router.
Login: repeater
Passsword: allstarlink
SSH port: 22
Once you have your RaspberryPi connected to your LAN you are ready to start configuring it for AllStarLink.
The first thing you need to do is login to the raspi via SSH and then become root user using sudo as shown below:
sudo su -
Once you are root user, you need to add the AllStarLink repo to the sources file and update the operating system using the following command:
curl -s http://apt.allstarlink.org/repos/repo_signing.key | apt-key add
apt update --allow-releaseinfo-change
apt dist-upgrade
Copy and paste each line one at a time into your terminal. Once the last command finishes, the system is up to date and can be rebooted as follows:
reboot
Once the raspi has rebooted, login again via SSH as user repeater and then become root user again.
You now need to install a couple of Python components that are required by the system to function. Use the commands below as user root:
apt-get install python3-dev python3-pip
pip3 install pyserial
Next you need to change directory into the asterisk config file directory using the command shown below:
cd /etc/asterisk
In this directory you will find all the default config files that come as part of the distro. For this build weβre not going to use them and so we need to move them out of the way ready for a set of config files that have already been configured correctly.
Using the following commands create a new directory, move into that new directory and then move all the unwanted configuration files into it:
mkdir ORIGINAL-CONF-FILES
cd ./ORIGINAL-CONF-FILES
mv ../*.conf ./
ls -la
cd ../
You should now be back in the /etc/asterisk directory which will now be empty apart from the custom directory which we left in place.
You now need to copy the correctly configured configuration files into the /etc/asterisk directory. Start by downloading the zip file containing the new configuration files
Once downloaded, copy the .zip file into the repeater users home directory (/home/repeater) using either scp on the Linux command line or if using Windows you can use the FileZilla Client in SFTP mode using the login details above.
Once you have the .zip file in the repeater userβs home directory you need to copy the file into the /etc/asterisk directory as user root:
cp /home/repeater/AllStarLink-Config-v3.zip /etc/asterisk/
Next as user root, change directory into the /etc/asterisk directory and unzip the .zip file:
cd /etc/asterisk
unzip ./AllStarLink-Config-v3.zip
Once the file is unzipped you will have a directory called AllStarLink-Config in the /etc/asterisk directory. You now need to cd into the directory, copy all the files out of it into the /etc/asterisk directory leaving a copy in the AllStarLink-Config directory for future reference:
cd /etc/asterisk/AllStarLink-Config
cp ./* /etc/asterisk
cd /etc/asterisk
You now need to move a couple of files into the repeater users home directory using the following commands:
mv ./SA818-running.py /home/repeater
mv ./gpio /home/repeater
Once the files have been moved you need to set the correct ownership and privileges on the files using the following commands:
chown -R root:root /etc/asterisk/*.conf
chown repeater:repeater /home/repeater/gpio
chown repeater:repeater /home/repeater/SA818-running.py
chmod 755 /home/repeater/gpio
chmod 755 /home/repeater/SA818-running.py
The gpio BASH script and configuration details were supplied by Mark, G1INU in the Digital Voice room on the Matrix. It adds the COS light functionality to the setup. The COS light will now light every time the SA818 hears RF on the input.
The next thing you need to do is configure the SA818 radio device in the SHARI. The script I used was originally from https://wiki.fm-funknetz.de/doku.php?id=fm-funknetz:technik:shari-sa818 all Iβve done is change the entries to switch off CTCSS and change the frequency to 438.800Mhz. Configuring the SA818 is done by running the SA818-running.py Python programme that you moved into the repeater user home directory. Making sure you are still user root, run the following commands:
cd /home/repeater
./SA818-running.py
At this point your SHARI SA818 device will be configured to operate on 438.800Mhz and CTCSS will be disabled.
If you want to change the frequency or enable and set a CTCSS tone to access the node you will need to edit the Python programme using your favourite text editor and change the entries accordingly. Once changed rerun the program as shown above and your SHARI will be reconfigured to your new settings.
Next you need to move the allmon.ini.php file into the correct directory so that it enables access to the Allstar Monitor web page on the device so that you can manage connecting/disconnecting nodes. Use the following commands as user root to achieve this:
cd /etc/asterisk
mv ./allmon.ini.php /var/www/html/allmon2/
chown root:root /var/www/html/allmon2/allmon.ini.php
chmod 644 /var/www/html/allmon2/allmon.ini.php
The allmon.ini.php file needs to have your node name entered into it to work correctly. As user root, change directory and edit the file using your favourite editor.
cd /var/www/html/allmon2
Using your text editor, search for the line starting [XXXXX] and change the XXXXX to your node number. Save the change and exit the file.
At this point you are almost complete, all that is left to do is add your node number and node secret into the appropriate configuration files in the /etc/asterisk directory.
Since I am a Linux command line junkie I use vi to edit all the configuration files on the command line as user root, but you can use any editor of your choice.
cd /etc/asterisk
Start with the extensions.conf file. Search for the line starting with NODE = and delete the XXXXX entry and insert your node number. Save the file and exit it.
Next you need to edit the iax.conf file. This time search for the line starting with
register= and change the XXXXX for your node number and the YYYYYYYYYYYY for your node secret. Be careful not to accidentally delete any other characters in the lines otherwise it will corrupt the configuration file.
In the same file search for the two lines that start with secret = and change the YYYYYYYYYYYY for your node secret. Once you have changed both of the secret entries, save and exit the file.
The final file to edit is the rpt.conf file. Once again open the file using your favourite editor and search for the line starting with XXXXX = radio@127.0.0.1:4569/XXXXX, change the XXXXX entries for your node number making sure not to delete any other characters next to the XXXXX entries.
Further down in the same file there is a line that starts with [XXXXX], once again change the XXXXX for your node number making sure to keep the square brackets at each end of the node number as you edit it.
Finally move down to the very bottom of the file and find the two lines that start with /home/repeater/gpio, once again change the XXXXX entries for your node number.
The final thing to change in the rpt.conf file is to replace my callsign with your own callsign so that the node identifies itself correctly. Scroll through the file until you find the two lines shown below, delete M0AWS and add your own callsign instead making sure you keep all the spaces between words as shown below.
idrecording = |i DE M0AWS
idtalkover = |i DE M0AWS
Once this is done, save and exit the file. At this point your node should be fully configured and will only require a reboot to get it working.
As user root, reboot your raspi using the reboot command.
reboot
Once your raspi comes back online, login using SSH as user repeater and then become root user using the sudo command detailed above.
You now need to create the admin user password for the Allstar Monitor web page on the device. This is done using the following commands as user root:
cd /var/www/html/allmon2
htpasswd -c .htpasswd admin
You will be asked to enter a password twice for the admin user. Make sure you make a note of this user/password as you will need it to login to the web page.
Finally check that the controlpanel.ini.php file is in the /var/www/html/allmon2 directory:
ls -la /var/www/html/allmon2/controlpanel.ini.php
If the file isnβt shown in the directory, enter the following commands to create the file in the correct place as user root and then exit the SSH session:
cd /var/www/html/allmon2
cp ./controlpanel.ini.txt ./controlpanel.ini.php
cd
exit
Once this is done your configuration is complete, logout from the terminal session by entering exit once more and your SSH session will terminate.
Using your favourite web browser enter the IP Address of your raspi into the URL bar as shown below:
http://<Your-Raspi-IP>/allmon2
Note: remove the <> from the URL once you have entered the required information.
Once this is done you should be presented with your node control panel as shown below.
Login using Admin and the password you set above and you are now ready to start using your node.
Itβs a good idea to connect to node 55553 which is a parrot test node to check your audio levels. You can do this by entering the node into the field at the top left and pressing the connect button.
Once connected, tune your radio to 438.800Mhz FM and transmit a test message using your callsign and test123, or something similar. The parrot will then play your recording back to you so that you can hear how you sound. It will also comment on your audio level as to whether it is OK or not.
You are now connected to AllStarLink network and have the world at your finger tips. Below is a small list of nodes in the UK, Australia and America to get you started chatting with other HAMs via your node.
57881 Matrix HAM Radio Space AllStarLink Node (Hosted by Dk1MI)
55553 ASL Parrot for testing
41522 M0HOY HUBNet Manchester, UK
60349 VK6CIA 439.275 Perth, Western Australia
51077 VK6SEG South West Hub B Albany WA
2167 M0JKT FreeSTAR UK HUB 2 freestar.network
53573 NWAG NW AllStar Group Lancashire, UK
27339 East Coast Hub Wilmington NC USA
Thanks to Michael, DK1MI for building and hosting the Matrix HAM Radio Space AllStarLink Node (57881) and getting us all started in the world of AllStarLink!
We hope to be having regular Matrix Netβs on the node soon for all Matrix members and visitors. Weβll organise days/times via the Digital Voice room.
More soon β¦
Following on from my article about my QO-100 Satellite Ground Station Complete Build, this article goes into some detail on the Node-RED section of the build and how I put together my QO-100 Satellite Ground Station Dashboard web app.
The Node-RED project has grown organically as I used the QO-100 satellite over time. Initially this started out as a simple project to synchronise the transmit and receive VFOβs so that the SDR receiver always tracked the IC-705 transmitter.
Over time I added more and more functionality until the QO-100 Ground Station Dashboard became the beast it is today.
Looking at the dashboard web app it looks relatively simple in that it reflects a lot of the functionality that the two radio devices already have in their own rights however, bringing this together is actually more complicated than it first appears.
Starting at the beginning I use FLRig to connect to the IC-705. The connection can be via USB or LAN/Wifi, it makes no difference. Node-RED gains CAT control of the IC-705 via XMLRPC on port 12345 to FLRig.
To control the SDR receiver I use GQRX SDR software and connect to it using RIGCTL on GQRX port 7356 from Node-RED. These two methods of connectivity work well and enables full control of the two radios.
The complete flow above looks rather daunting initially however, breaking it down into its constituent parts makes it much easier to understand.
There are two sections to the flow, the GQRX control which is the more complex of the two flows and the comparatively simple IC-705 section of the flow. These two flows could be broken down further into smaller flows and spread across multiple projects using inter-flow links however, I found it much easier from a debug point of view to have the entire flow in one Node-RED project.
Breaking down the flow further the GQRX startup section (shown below) establishes communication with the GQRX SDR software via TCP/IP and gets the initial mode and filter settings from the SDR software. This information is then used to populate the dashboard web app.
The startup triggers fire just once at initial startup of Node-RED so itβs important that the SDR device is plugged into the PC at boot time.
All the startup triggers feed information into the RIGCTL section of the GQRX flow. This section of the flow (shown below) passes all the commands onto the GQRX SDR software to control the SDR receiver.
The TCP RIGCTL -> GQRX node is a standard TCP Request node that is configured to talk to the GQRX software on the defined IP Address and Port as configured in the GQRX setup. The output from this node then goes into the Filter RIGCTL Response node that processes the corresponding reply from GQRX for each message sent to it. Errors are trapped in the green Debug node and can be used for debugging.
The receive S Meter is also driven from the the output of the Filter RIGCTL Response node and passed onto the S Meter function for formatting before being passed through to the actual gauge on the dashboard.
Continuing down the left hand side of the flow we move into the section where all the GQRX controls are defined.
In this section we have the VFO step buttons that move the VFO up/down in steps of 10Hz to 10Khz. Each button press generates a value that is passed onto the Set DeltaFreq change node and then on to the Calc new VFO Freq function. From here the new VFO frequency is stored and passed onto the communications channel to send the new VFO frequency to the GQRX software.
The Mode and Filter nodes are simple drop down menus with predefined values that are used to change the mode and receive filter width of the SDR receiver.
Below are the HAM band selector buttons, each of these will use a similar process as detailed above to change the VFO frequency to a preset value on each of the HAM HF Bands.
The QO-100 button puts the transmit and receive VFOβs into synchro-mode so that the receive VFO follows the transmit VFO. It also sets the correct frequency in the 739Mhz band for the downlink from the LNB in GQRX SDR software and sets the IC-705 to the correct frequency in the 2m VHF HAM band to drive the 2.4Ghz up-converter.
The Split button allows the receive VFO to be moved away from the transmit VFO for split operation when in QO-100 mode. This allows for the receive VFO to be moved away so that you can RIT into slightly off frequency stations or to work split when working DXpedition stations.
The bottom two Memory buttons allow you to store the current receive frequency into a memory for later recall.
At the top right of this section of the flow there is a Display Band Plan Info function, this displays the band plan information for the QO-100 satellite in a small display field on the Dashboard as you tune across the transponder. Currently it only displays information for the satellite, at some point in the future I will add the necessary code to display band plan information for the HF bands too.
The final section of the GQRX flow (shown below) sets the initial button colours and starts the Powermate USB VFO knob flow. Iβve already written a detailed article on how this works here but, for completeness it is triggered a few seconds after startup (to allow the USB device to be found) and then starts the BASH script that is used to communicate with the USB device. The output of this is processed and passed back into the VFO control part of the flow so that the receive VFO can be manually altered when in split mode or in non-QO-100 mode.
The bottom flows in the image above set some flow variables that are used throughout the flow and then calculates and sets the RIT value on the dashboard display.
The final section of the flow is the IC-705 control flow. This is a relatively simple flow that is used to both send and receive data to/from the IC-705, process it and pass it on to the other parts of the flow as required.
The IC-705 flow is started via the timestamp trigger at the top left. This node is nothing more than a trigger that fires every 0.5 seconds so that the dashboard display is updated in near realtime. The flow is pretty self explanatory, in that it collects the current frequency, transmit power, SWR reading, PTT on/off status and S Meter reading each time it is triggered. This information is then processed and used to keep the dashboard display up to date and to provide VFO tracking information to the GQRX receive flow.
On the left are the buttons to change band on the IC-705 along with a button to tune to the VOLEMT on the 60m band. Once again there two memory buttons to save and recall the IC-705 VFO frequency.
The Startup PTT Colour trigger node sets the PTT button to green on startup. The PTT button changes to red during transmit and is controlled via the Toggle PTT function.
At the very bottom of the flow is the set transverter IF Freq function, this sets the IC-705 to a preselected frequency in the 2m HAM band when the dashboard is switched into QO-100 mode by pressing the QO-100 button.
On the right of the flow there is a standard file write node that writes the 2.4Ghz QO-100 uplink frequency each time it changes into a file that is used by my own logging software to add the uplink frequency into my log entries automatically. (Yes I wrote my own logging software!)
The RX Audio Mute Control filter node is used to reduce the receive volume during transmit when in QO-100 full duplex mode otherwise, the operator can get tongue tied hearing their own voice 250ms after theyβve spoken coming back from the satellite. This uses the pulse audio system found on the Linux platform. The audio is reduced to a level whereby it makes it much easier to talk but, you can still hear enough of your audio to ensure that you have a good, clean signal on the satellite.
As I said at the beginning of this article, this flow has grown organically over the last 12 months and has been a fun project to put together. Iβve had many people ask me how I have created the dashboard and whether they could do the same for their ground station. The simple answer is yes, you can use this flow with any kind of radio as long as it has the ability to be controlled via CAT/USB or TCP/IP using XMLRPC or RIGCTL.
To this end I include below an export of the complete flow that can be imported into your own Node-RED flow editor. You may need to make changes to it for it to work with your radio/SDR but, it shouldnβt take too much to complete. If like me you are using an IC-705 and any kind of SDR controlled by GQRX SDR software then itβs ready to go without any changes at all.
More soon β¦
Itβs been a busy few months, and the folx who I interact with on Mastodon already know that Iβve got lots of irons in lots of fires. I figured itβs probably time for me to do a quick βstate of the shackβ post to give a quick highlight on where all my various projects and activities stand, in case anyone has been wondering about the status of any particular item.
So, in no particular order, here we go:
Check out the Repeater Page and posts tagged repeater for more details.
The quick summary here, is that my Motorola GR1225 died, so I currently have a machine on the air using a pair of Kenwood TK840 radios, the duplexer from the Motorola, and a new USB interface from Repeater-Builder to connect it to the computer that runs the Allstar software. After some very positive feedback on the post I wrote as I was trying to get the machine running again, I decided to start a fund-raiser to support some upgrades to the repeater, and to also help repair and get some additional repeaters on the air, which brings us to the next projectβ¦
Check out the r4e project pages for more details.
r4e is an acronym for Repeaters 4 Everybody.
As a way to support upgrades on my repeater and some additional repeaters that are operated with a purposeful mission of openness and acceptance, and to help bulk up the RF side of the Pride Radio Network, Iβve started the r4e project which some of you may have stumbled across already in the header of my site. If youβre willing to help support some repeaters financially, or with donations of equipment, or to just connect an existing repeater to the pride network, those gifts and actions can go a long way towards our roadmap!
Check out the Subversive Radio Shop for details, and to buy cool stuff!
This project is an offshoot of the r4e project, and is a way to raise additional funds for those projects by selling radio merch that (I hope) is unlike most of the stuff already out there. Buy some cool stuff to support the project!
Check out the Radio Rocket Page and the Radio-Rocket Tagged Posts for more details.
Version 3 of the rocket (Ponzu) had some body tube damage during its first flight, which is now repaired, and itβs ready to fly again. I also built a new ground station that is an all-in one unit with the single board computer, LoRa receiver, an RTL-SDR dongle for receiving APRS packets, touchscreen for launch control and data display, etc. Motors are ordered for the next launch, which will be on June 13th or 14th, as weather permits.
Check out the Narwhal Amateur Radio Society, and our Nets Page (details coming soon, if they arenβt already there!) for more info.
A while back I joined the Narwhal Amateur Radio Society (NR7WL) - theyβre a relatively new club, but have values that I dig. We had kicked around the idea of a club βnetworkβ to be able to connect via digital modes, and potentially a club net. That idea sat for a little bit, but bubbled back up recently, so I took the initiative to set up an Allstar node for the club (61672) which will serve as our hub, and the location for our first club net, for which Iβll be serving as net control!
Net details as follows:
Narwhal Amateur Radio Society Club Net
Time: 1st Tuesday of Every Month at 7pm Pacific, 10pm Eastern
Location: Hosted on the NR7WL Allstar Node (61672) and the Pride Radio Network. The Pride Radio network has bridges that will allow you to connect via DMR, IRLP, System Fusion, M-17, NXDN, P-25, D-Star, Echolink, Hamps Over IP, Hamshack Hotline, plus others!
Check out posts tagged shack for more details.
I havenβt made any major updates to the shack recently, but I have moved a few things around. In the utility space behind the operating position I added a DIY rack made from lumber to move the various computer and network bits and bobs into. My next project in the shack is related to re-doing some of the audio routing - I have designs drawn up in KiCad for an interface device that will sit beside my mixer and convert all the audio to the OHIS standard. The basic reasoning behind this for me, is that by converting the audio to OHIS, I can leverage an existing standard to run a single shielded cat6 cable to each radio and device, instead of the 3 or 4 audio and PTT cables that run to each radio now. I currently have some bursting-at the seams cable management, and doing this should reduce the mess in there quite a bit. The plan is also to eventually replace my aging mixer, and build a new rack mount arrangement to the left of my operating position to house the mixer, interface, and some other related shack equipment.
Check out the ARIP website for more info.
Diversity is something important to our hobby. If you disagree, or think this statement is somehow political, youβre part of the problem in the hobby, and I wonβt engage with you on the subject. If you do feel the same way I do however, the ARIP is one of my projects that is essentially a tool that clubs, individuals, or other organizations can use as a way to show their commitment to inclusion in the hobby. The most recent update on this effort is some changes to the website and methodology to make it more of a self-serve tool. There is more information about that available on the ARIP Website
Check out the Radio League of America website for more info.
Currently, the Radio League of America (RLA) is little more than an idea. That idea being that the amateur radio community is too large and diverse for everyone to have their voices heard by a single national organization. There are many amateurs who have voiced a desire for something different to be available, and the RLA is just one of many potential avenues as that movement takes life. I have committed to at some point in the near future getting together an initial presentation of what that might look like, and setting up a recurring (probably quarterly) series of meetings for people who are interested to see if it is something worth fleshing out in more detail and organizing around.
Ever since my QO-100 ground station has been operational Iβve been using my NodeRed QO-100 Dashboard to control my IC-705 and GQRX SDR software to drive my NooElec SmartSDR receiver. This gives me a full duplex ground station with both transmit and receive VFOβs synchronised.
This solution has worked incredibly well from the outset and over time Iβve added extra functionality that Iβve found to be useful to enhance the overall setup.
The latest addition to the ground station solution is a Sennheiser Headset that I picked up for just Β£56 on Amazon (Much cheaper than the Heil equivalents at the HAM stores!) and have found it to be excellent. The audio quality from both the mic and the headphones is extremely good whilst being light and comfortable to wear for extended periods.
To incorporate this into the ground station the headset is connected to my Kubuntu PC and the audio chain to the IC-705 is sent wirelessly using the latest version of WFView. This works extremely well. The receive audio comes directly from the GQRX SDR software to the headphones so that I have a full duplex headset combination.
Audio routing is done via pulse audio on the Kubuntu PC and is very easy to setup.
Since I no longer have a mic connected to the IC-705 directly I found that I needed a way to operate the PTT wirelessly and this is where the latest addition to my NodeRed QO-100 Dashboard comes in.
Adding a little functionality to the NodeRed flow I was able to create a button that toggles the IC-705 PTT state on and off giving me the ability to easily switch between receive and transmit using a simple XMLRPC node without the need for a physical PTT button.
The PTT state and PTT button colour change is handled by the Toggle PTT function node shown in the above flow. The code to do this is relatively simple as shown below.
The entire QO-100 Dashboard flow has grown somewhat from itβs initial conception but, it provides all the functionality that I require to operate a full duplex station on the QO-100 satellite.
This simple but, effective PTT solution works great and leaves me hands free whilst talking on the satellite or the HF bands when using the IC-705. This also means that when using my IC-705 it only requires the coax to be connected, everything else is done via Wifi keeping things nice and tidy in the radio shack.
The image above shows the QO-100 ground station in receive cycle with the RX/TX VFOβs in split mode as the DX station was slightly off frequency to me. The PTT button goes red when in TX mode just like the split button shown above for visual reference.
As you can probably tell, Iβm a huge fan of NodeRed and have put together quite a few projects using it, including my HF Bands Live Monitoring web page.
More soon β¦
Since Iβve been using my Icom IC-705 on the QO-100 satellite Iβve been getting no end of unsolicited great audio reports with one Op even saying I have the best audio heβs ever heard on the satellite.
Most people are surprised when I tell them that I am using the stock fist mic that comes with the radio. Itβs nothing special, in fact itβs rather cheap and plastic, not particularly good quality however, it does seem to have a good sounding mic insert.
The other great thing about the IC-705 is that it has a two channel parametric equaliser built into the radio. Many people donβt realise this and miss out on the massive improvement they can make to their transmitted audio with just a few simple adjustments.
The stock fist mic has a very flat response across the audio frequency range out of the box and doesnβt sound particularly inspiring. Many see this as a negative and often just replace the mic with either a headset (probably from Heil), a boom mic (again probably from Heil) or another, better quality fist mic. All of these options cost varying amounts of money when in reality none of them are necessary.
Starting from a flat audio response is actually a good thing as it makes the equaliser adjustments more pronounced, making it easier to adjust the settings to suit your voice.
We all have different voices but, there is one thing that is pretty much the same for everyone and thatβs the frequency range in which the articulation of the words and sounds we make can be found. Itβs this part of the voice that is often lacking when we struggle to understand what the DX station is saying.
Itβs become common place on the HAM bands these days for stations to boost the bass frequencies and reduce the mid and high frequencies with the net result of a horrible bass ringing sound and muddy mid range often making it very difficult to understand what is being said.
Having spent some considerable time watching the great videos on audio from the late Bob Heil, K9EID itβs clear that the most important frequencies to enhance are those around 2.5khz as this is where all the articulation is in the human voice.
To this end I set about setting up the audio on my IC-705 QRP radio so that my voice sounded such that it is easy to comprehend even in the most difficult of situations on air. This doesnβt mean that it has to be very harsh and overly bright, quite the opposite in that to be heard clearly in all conditions on air oneβs audio needs to be balanced across the frequency range with an enhancement in the 2.5Khz frequency range.
To reduce the unwanted, muddy bass the first thing to do is change the transmit bandwidth for the βWideβ setting to 200-2900Hz. This will cut off the bottom 100Hz from the voice reducing the overall bass output from the standard fist mic that comes with the radio. This will ensure a 2700Hz wide SSB signal, the recommended max for QO-100 operations and the preferred bandwidth on the HF bands.
On top of this I made a further reduction of 2dB on the TX Bass setting to help balance out the overall audio response of the mic insert.
Next I set about enhancing the higher frequency response of the mic insert and found that it required an increase of 4dB to bring out the articulation of my voice. This enhanced my audio considerably compared to the standard output from the fist mic and improved the intelligibility of my voice considerably, especially in difficult band conditions.
To complete the setup I set the compression to 3 and mic gain to 35 so that the overall drive level is increased slightly giving a greater average output from the radio.
Once Iβd got the audio setup correctly I enabled the configuration by setting the Transmit Bandwidth (TBW) to the βWideβ config in the IC-705 Function menu so that the correct settings were made active.
Ever since making these relatively easy changes I have had no end of unsolicited great audio reports from stations asking me what mic I am using and how Iβve managed to get such good audio from the IC-705. Many are surprised that I am using the OEM fist mic that comes with the radio and Iβm sure there are those who donβt believe me!
Of course all voices are slightly different and these settings may not be perfect for your voice but, all those that have tried these settings have told me that their audio sounds better than ever and that DX stations often comment on how good their audio is.
I also went through the same exercise with my Yaesu FTDX10 with itβs standard fist mic and again achieved excellent results with itβs 3 channel parametric equaliser. Iβll go through the somewhat more complicated setup for the FTDX10 in another article soon.
Using just 20w I had a quick chat with Callum of DXCommander fame on the 80m band using my Inverted-L antenna.
Very happy with the signal report and being able to hear myself after the event was very interesting. Iβve not used my Yaesu FTDX10 much on SSB and so I was very happy to hear that the audio sounded nice and clear.
More soon β¦
![]() |
Cable with large distance between the conductors (Schnerzinger) |
The first example had a large distance between the conductors and was placed on its own stand to get distance from the floor. It was designed to minimize capacitance between conductors. Another design criterion was to minimize the impact of vibrations created by the speakers themselves.
![]() |
Twisted multi-conductor cable |
There is no shortage of science-based claims for how cables affect sound. There are so many of them that this blog post, where my ambition is to say something about all of them, gets to be a little longer than I would have liked it to be. If you are impatient, you may go to the end and just see the conclusion. For the curious, I will consider the claims one by one.
After resistance, it isΒ inductanceΒ andΒ capacitanceΒ that matter.Β But is a small inductance or a small capacitance preferable? Ideally, both should be zero. A cable can be made with a small value for one by getting a large value for the other. But it is not possible to produce a cable with both low inductance and low capacitance. And whether you have much of one or the other, they will both influence the treble. A hand-waving way of saying it, is that inductance blocks high frequencies while the capacitance short-circuits them. And even in a regular short speaker cable, this means that the bass will arrive a little later than the treble (dispersion).
But in a normal short speaker cable, the difference in time delay does not correspond to much more than if the tweeter element in the speaker was moved a maximum of one mm back, so the consequence is minimal. In addition, the attenuation of the treble is so small that it can be neglected in most cases. If you still want to minimize the loss for high frequencies, it pays to design for a low inductance rather than low capacitance. In other words, it is an advantage that the conductors are close together as they are in most cables, and not far apart as in the cable in the top picture. Iβll come back to why at the end of the article.
Some cable designers emphasize minimizing the reflection of the signal from the ends of the cable. That means adapting the load to the cable's characteristic impedance. This is well known from cable TV where the impedance of the transmission line is 50 or 75 ohms. There are several objections to impedance matching for speaker cables. First, it is not possible to make cables with impedance as low as 4 or 8 ohms as is typical for speakers. But even more important is that a TV cable is long in relation to the wavelength of the signal. That is what makes the transmission line model applicable. This is not the case for audio cables where the highest treble has a wavelength of at least 10 km and the bass even longer. Therefore, speaker cables cannot be understood as transmission lines and all talk of impedance matching is meaningless.
![]() |
Pupin coil on insulator on top of a telephone pole (American Electrician 1903) |
Others emphasize that their cables avoid resonance. This may sound convincing, as a system that has inductance and capacitance always has a resonant frequency. Even here there is a mixture of two models. The low frequency model of a cable, the one that applies to audio, consists of a discrete inductance and a discrete capacitance. Using typical values, one will find a resonance frequency of a MHz or two, which is far higher than what is audible. This fact alone should indicate that the effect is insignificant. Even more important is that at these frequencies, there is another cable model that applies, namely the transmission line model mentioned above. It has distributed inductances and capacitances and that gives no resonance. Therefore, the whole resonance problem is yet another imagined problem due to incorrect application of models.
Closely related to resistance is the skin effect. This means that the higher the frequency, the more current is displaced to the outer part of the conductor. Even at 50 Hz, as in the electricity supply, the current does not penetrate much more than one cm into a copper cable. At 20 kHz, the skin depth is about 0.5 mm. This means that for thick cables (2.5 mm2 and more or AWG 10 and less) the effective resistance increases slightly with frequency. The proximity effect, which is how currents in the two conductors influence each other, also contributes in the same direction.
![]() |
Flat cable which is 12.5 cm wide (Magnan Audio Cables) |
A third way to deal with the skin effect is to make a hollow coaxial cable, as used for cable TV. The expensive Pear Anjou cable that annoyed James Randi, was built this way (see part 1). However, compared to the increase in impedance due to the inductance of the cable, the contribution from the skin and proximity effects is not very large. Still, it does not hurt if a cable minimizes these effects, but it may be questionable whether it is worth the money. The skin and proximity effects are real effects that can be measured. They illustrate the fact that even if something can be measured and analyzed, it does not necessarily mean that it is important. Designs like that of the twisted multi-conductor cable shown above are despite this reservation to be recommended.
How can it be that some exotic cables are made with a large distance between the conductors to get low capacitance, when I have just said that it pays to have small inductance? Often it is because they have focused on a memory effect found in capacitors and which is due to dielectric absorption. Even when a capacitor is what appears to be fully discharged, it may still remember some of the charge. I have illustrated this in a YouTube video. This is a real effect, especially in capacitors with high capacitance, much larger than in a cable. It has to do with the quality of the material between the conductors. Materials such as Teflon minimize it, but even better is air such as in the cable that floats on stands above the floor in the picture above. A cable supplier can make it a point that their Teflon insulation cures such effects and explain how cables can be 'non-linear' and smear the sound and give a bad stereo image. Is it real? Not really, as the memory effect is linear since it can be modeled with a network of capacitors and resistors, none of which can distort a signal non-linearly. This has been illustrated by analog guru Bob Pease.
Then there are those who emphasize the importance of properΒ materials in the conductors. As mentioned earlier, good conductivity is important. With its good conductivity, acceptable price, and good mechanical properties, it is no wonder that copper is so popular. It is barely beaten by silver which has 5% lower resistance than copper. Gold does not come out so well as it has almost 50% more resistance, but it is well suitable for electrical contacts as it does not oxidize. You can get expensive speaker cables made from silver. But the question is whether it is not better to increase the copper cross sectional area by 5% rather than use the much more expensive and mechanically more fragile silver.
There are also many who emphasize the purity of the material. Here it is appropriate to say something about how a signal is conducted along a cable. A common analogy is a water pipe. In a cable, it is the electrons that are said to flow like water in the copper. But this analogy starts to falter when you look at the transport velocity of electrons in a material like copper. It increases with frequency but never becomes more than a few tens of m/s. Let's say that a power plant located 500 km away starts up and the current is transported by electrons traveling at 10 m/s. Then it will take almost 14 hours before I notice that the power has been turned on! Here, the analogy with the water pipe has obviously broken down.
![]() |
Nobody likes corroded cables |
The same applies to those who argue that there will be distortion if the cable consists of several strands. For this reason, there exist rigid and unwieldy speaker cables made from solid copper. The idea is that there will be a rectification effect in the transition between the various strands, i.e. between copper and oxidized copper, much like in a crystal in an old-fashioned crystal radio. But this is also energy which is lost as heat. Moreover, such a distortion should be possible to measure, but I have never seenΒ such measurements. The argument that the cable needs a break-in period of several hours or more, as well as the idea that it is important which direction a cable is connected also fall on its own unreasonableness. Remember, however, that it is still important that both speakers are polarized in the same way, so it does matter which of the two cable ends is connected to which of the speaker terminals. But this is a pure acoustic effect to ensure that both speaker diaphragms move in and out in phase and it has nothing to do with the properties of a cable.
When it comes to tube amplifiers, some may have noticed that knocking on a tube sometimes can be heard in the speaker. This is a mechanical or triboelectric effect that also can happen in poor quality microphone cables. The danger is that the sound pressure from the speakers themselves can cause the cables to vibrate and create distortion. Some people get hung up on this in the design of speaker cables and argue for choosing geometry and insulation materials based on it. Since it is possible to design microphone cables which are robust against such effects with their very low signal levels, it goes without saying that this cannot be an important effect in a speaker cable where signals are so much stronger.
There is also an effect which in a way is the opposite of that in the previous paragraph. Due to the large currents in a speaker cable, there may be forces between the conductors that can make them attract and repel each other. If this leads to movement, it can give rise to a new signal which can disturb and distort the actual signal. This is used as an argument for putting the conductors far apart. But the cure must rather be to make a mechanically stable construction. Besides, this is also an effect which should be measurable. I have even tried to measure it myself without success. Therefore IΒ doubt if there is anything here to worry about.
My last point is an effect which is easy to forget. There is a possibility that a speaker cable will act as an antenna and pick up unwanted signals. This applies especially to signals that are outside the audible range. They are sent 'backwards' into the amplifier. What happens next depends on how well the amplifier is designed. Such signals can be from nearby radio transmitters. Medium and long wave frequencies are often worst. But it can also be from a noise source that not many people think about and which can be in their own living room, namely a plasma TV. The open cable solutions with a large distance between the conductors are clearly the worst. This has been confirmed by both measurements and listening tests. The best cables to minimize such interference are shielded (coaxial) cables and twisted multi-conductor cable, but cables with the conductors close together are also good.In fact, there are several amplifiersΒ of minimalistic design in hi-fi that are more easily influenced by external factors than other more robust constructions. These can be amplifiers that professionals steer clear of, but which are still large in the high-end hi-fi market. This may be due to sensitivity to external radiation, or that the amplifier is only marginally stable and thus extra sensitive to capacitive loads. Then a standard low-inductance cable, implicitly one with a high capacitance, could create problems. Finally, it should be mentioned that tube amplifiers, which often have a large output impedance, are more easily affected by the complex interaction between a cable, the crossover filter in the loudspeaker and the loudspeaker elements than other amplifiers.What should we conclude, is there magic in speaker cables? We have looked at the effect of a large and a small distance between the conductors, on reflection and resonance, on skin and memory effects, on the purity of the conductor, whether break-in and polarizing the cable is important, on vibration effects and finally on potential interference from external signals. One factor that has deliberately not been discussed is the aesthetic value. It should not be underestimated that hi-fi sells both on acoustic performance and on pleasing design.
But from a purely performance point of view, I end up with the finding that low-resistance, low-inductance cables are well supported. The most common and cheapest type is with conductors next to each other, but they may be twisted together as well. This means that the cables can be quite affordable as long as they have a cross section of at least 2.5 mm2 (AWG 10). In professional audio, 2 x 4 or 2 x 6 mm2 are often used (AWG 6-3), unless they use active speakers where there is hardly any cable at all. Twisted multi-conductor cablesΒ are often not very expensive and may possibly have an edge at high frequencies. In any case, the longer the cable, the thicker it should be. Due to the possibility of radio frequency interference, exotic cables with a large distance between the conductors are something I want to warn against. The same goes for high-inductance cables with discrete inductors embedded in them, all other exotically designed cables have no negative effects, other than possibly on your wallet. And since no cable is ideal, the recommendation is to make it as short as possible, but still the same length for each speaker.
Sources:
Β (First published in Norwegian in: Magiske hΓΈyttalerkabler - del 2)
The post "Magical speaker cables - part 2" first appeared on the LA3ZA Radio & Electronics Blog.
![]() |
Pear Anjou speaker cable |
A climax in the cable dispute may have been reached in 2007 when the skeptic James Randi offered a reward of 1 million dollars if anyone could prove that speaker cables costing $2750 for a pair of 1 meter long cables provided any improvements. He was provoked by the claims of theΒ Pear Anjou cables shown in the image.
There was a time when connecting speakers to an amplifier was simple. Speakers came with a thin cable and that was it. This cable was very unassuming and easy to hide as shown in the image below. Then someone came along claiming that this wasnβt good enough. Neither were the tiny DIN-rated speaker plugs, even though they followed German industry standards (DIN once meant Deutsche Industrie-Normen). Since then, there has been no end to the flood of increasingly elaborate and expensive speaker cables.
![]() |
Thin cable from the 1970's with DIN plugs |
Music is something that means a lot to many and it has an almost direct path into the emotions. There is therefore no doubt that what we hear and experience is what counts. But we are talking here not just about ordinary ears, but golden ears - those who through training and experience have a special ability to perceive what others cannot hear. Although it is easy to cheat in this field, there is no doubt that such golden ears exist. But they have their limitations. They are good for analyzing and spotting deficits in sound reproduction, but they provide very few clues as to what it takes to design equipment that sounds correctly.
The story about the Norwegian company Electrocompaniet illustrates the value of golden ears. In the 1970's they came up with one of the first power amplifiers that fixed what was called βtransistor soundβ. Many could hear the problem with early transistor amplifiers, but few knew what caused it before the Finnish professor Matti Ottala explained it as transient intermodulation distortion. Today it is usually called slew-induced distortion. This story demonstrates how an audible problem was not fixed until someone managed to quantify it with measurements. Read the fascinating story of Svein-Erik BΓΈrja with the golden ears, and Terje SandstrΓΈm with the engineering talent in British Hi-Fi News from 2011. It shows how important it is to quantify with numbers what matters to the audible impression.
![]() |
A speaker cable which is too thin along with one that is better |
It is also easy to explain from measurements. It is first and foremost resistance in ohms per meter that explains it. For the thinnest cable it is 0.045 ohm/m or 0.45 ohm for a 5 m cable (back and forth). The voltage of the amplifier will be divided between the resistance of the cable and the resistance or impedance of the speaker of nominally 4 ohms. This gives a loss of 1 dB which is audible. Even more audible will be the coloration of the sound that comes from the speaker's impedance varying with frequency. It can cause both a resonant top in the bass and a raising of the treble. Another argument for low resistance is that this helps the amplifier dampen speaker motion once a signal has stopped. This is called the damping factor.
Here many with a scientific or technical background stop and reject all other arguments. My point in the rest of this article and in part two, is to discuss the main technical arguments that have been raised. The various manufacturers don't really promote their solutions solely on the basis of subjective listening impressions. Virtually everyone has arguments based on some physical principle. They like to claim that they are particularly good at this single aspect where the company believes they have a special insight. This is formulated in scientific terms. But then we are really at the heart of what this blog is all about. If it is formulated in scientific terms, it should be possible to analyze with respect to whether this aspect is important or not.
![]() |
Twisted multi-conductor cable |
I have already shown the most common types of cables in the figures above. Here are two slightly more exotic types. The first is a cable made by twisting together many thinner conductors, in this case a do-it-yourself cable. It is designed to have the least possible inductance as well as to minimize the consequences of the skin effect. The latter is the tendency for alternating current only to flow in the outer part of a conductor.
![]() |
Cable with large distance between the conductors (Schnerzinger) |
But what are the factors which are of real importance? In part 2, we consider those aspects, beyond resistance, which are emphasized in cable design.
References:
(First published in Norwegian in: Magiske hΓΈyttalerkabler)
The post "MagicalΒ speaker cables - part 1" first appeared on the LA3ZA Radio & Electronics Blog.
On a whim, I took another look at the LM386-4 one cold Sunday afternoon this November. My focus was to drive a loud speaker and not headphones. I wonβt personally use the LM386 for a headphone amp as we enjoy so many better options. For example, an op-amp driving a pair of TO-92 followers, or perhaps placing 2 NE5532 op amps in parallel as the headphone PA stage.
Since the mid 1970βs the LM386 has enjoyed popularity amongst hobbyists for low to flea power AF power amplification. The NE612 mixer IC plus the LM386 have literally formed the basic building blocks of innumerable radio receivers amongst Hams and hobbyists for decades.
Although, imperfect like all other linear ICs, the LM386 design team delivered a simple, flexible, low power AF amp with reasonably low distortion.
This part is noisy though. The input noise density = ~ 50 nV/β(Hz)Β β aboutΒ 10X that of an NE5532 op amp. So. if you use this part in high gain mode [with a gain of 100 to 200] and drive it with a low-level audio signal, youβll really hear the noise (hiss) in your speaker.
Others online have provided detailed analysis about each stage of the LM386, so I wonβt bother. However, I will comment about why it might be noisy. Normally, in modern AF power amps. the differential input pair emitters get 50 - 100 ohms of degeneration to boost linearity at the expense of noise. Other than that, in the IC only current sources connect to the emitters (and usually active loads to the collectors -- i.e. no resistors), However, the LM386 input pair get multiple large value resistors connected to their emitters. This translates into lots of Johnson noise from thermal agitation within conductors, plus related high-level input current noise that all gets amplified by the NPN voltage amp and delivered to the output stage.
1. Gain = 20 Mode
2. Bass Boost and More
3. Gain = 50 Mode
4. Gain = 200 Mode -- plus lifting and AC bypassing Pin 2
I'm avoiding math in 2022, as data shows that my blog readers don't care for it.
Above βΒ I kept advancing the gain knob until clipping appeared on the top half.
Above βΒ Switch to FFT mode, although you can still see the yellow coloured sine wave. At this point, the tones are almost level and 45 - 47 dB down. You may easily hear distortion at this level.
If you listen to this amplifier with music through a speaker, it sounds muffled and somewhat lacks the important mid frequencies for both voice and music. The hiss is definitely attenuated though. Further, the voltage gain goes from 20 without feedback down to around 8 with the 10K + .033 Β΅F network added. There are other potential side effects to with such heavy feedback which I'll show soon.
Above β The bass boost circuit may produce weird distortion when driven hard in some circuits. The bottom half of the sine wave clips initially.
Let's adjust the feedback network and perhaps find something that works better.
In a classic PA with a differential transconductance input pair, 1 BJT base serves as the input stage while the other base receives the negative voltage feedback. The transconductance pair subtracts that negative feedback from the input and passes the difference voltage onto the voltage amplifier stage that follows. This doesn't happen in the LM386 -- negative feedback goes to the non-inverting BJT emitter which is often also the input side of the input emitter coupled pair.
Negative feedback also affects an amps gain, bandwidth, frequency response, plus its input and output impedance (although the output impedance is just fractions of an ohm). When we add AC feedback between pins 1 and 5, our network is in parallel with the 15K resistor and may be affected by other amplifier parameters including the gain and input impedance.
From the data sheet, In low gain mode, we should strive to keep the amplifier's closed loop gain 10 or greater which happens with the 10K resistor in our R C network. It's quite easy to turn your LM386 into an oscillator with too much feedback. I've noticed that feedback networks that look OK in SPICE simulations may actually oscillate in real life bench work -- especially with higher gain and/or drive.
I took the bass boost circuit example, kept the 10K resistor, and tried different capacitor values. If you lower the 10K resistor, you'll have to watch for oscillations at input drive levels high enough to cause distortion. This also may also reduce the LM386 voltage gain considerably.
Above β Our base schematic to evaluate different values of C1 and view the resultant FFT and voltage gain.
Above β FFT at 5 Vpp where C1 = 0.01Β΅F. Outstanding results! This turned out to be the best feedback capacitor of the few I tried. The LM386 voltage gain dropped to just under 11 with that particular capacitor value for C1.
Above β Driving it as little harder to give 623 mW output power. Still fairly clean compared to other tracings.
Above β Pushed a little harder to 761 mW. Things are getting ugly. C1 still = 10 nF. Let's decrease the C1 value by a decade to 1 nF:
Above β FFT at 5 Vpp where C1 = 0.001Β΅F. While not as impressive as when C1 = 10 nF, it's still quite good and the LM386 voltage gain is around 19.
Above β FFT at 7.14 Vpp or 759 mW output power, The second harmonic is somewhat better than the case where C1 = 0.01uF.
Above β FFT at 5 Vpp where C1 = 470 pF.Β Another favourable reduction of harmonic distortion when the LM386 amp is running at reasonably high, unclipped power levels. I measured no loss in voltage gain with a 470 pF cap + 10K resistor.
I also tried a 220 pF cap - it worked somewhat, but the harmonic suppression started to fall off at this point. The overall best unclipped harmonic suppression occurred where C1 = 0.01 Β΅F in my experiments, albeit with 45% voltage gain loss.
To decide on a C1 value, it's important to listen to it to with actual audio to ensure that any frequency peaks, or more importantly, the low pass effects caused by the network doesn't wreck the audio you listen to.
Above β FFT at 5 Vpp with a 1K2 and 10 Β΅F cap between Pins 1 and 8. Although we see the 2nd harmonic at about 52 dB down, it's still OK for the LM386. Signal noise will appear louder compared to the "Bass Boost" feedback variants with the same output power.
Last section. Here's the famous schematic used by millions to make simple DYI audio projects:
Connecting Pins 1 and 8 AC bypass a 1.35K emitter resistor in the input pair -- and unleash the hounds. We get a full menu of gain, noise, and potentially harsh sounding distortion.
Above βIn this separate experiment to showcase the worst-case scenario, I've manipulated & then pushed this particular amp into raucous distortion. Note the strong 3rd and 5th harmonics relative to the 2 even harmonics. This is worst case fuzz box stuff. While this might sound bad with your ears, it's great fun to see it on a DSO.
Above β Back to the main experiments using the schematic shown above... The FFT at 5 Vpp or 372 mW output power. The 2nd harmonic lies at -46 dBc.The various tones do not go down much at lower input signal levels.
Above β A slight increase from 7.01 to 7.09 Vpp. The 3rd harmonic is about 41 dB down.When in full gain mode, the LM386 tends to offers more odd harmonics
Above β Top and bottom sine wave clipping translates into wretched distortion.
My question -- will feedback similar to what we used in the Bass Boost variants lower the harmonic distortion?
Above β At our standard of 5 Vpp, the affects of feedback leap out at us. [ 10K plus 0.01 Β΅F cap ] Feedback is our friend?Β However, the gain dropped to around 100 and you will hear lots of high frequency roll off.
Above β Vpp = 5 with a feedback network consisting of a 10K + 0.001 Β΅F cap. The response is lack luster compared to 10K + 0.01 Β΅F capacitor. The drop in voltage gain was only 2%. The 2nd harmonic is maybe 1-2 dB better than without the feedback network?
Above β There's an old trick left to try. Normally, most builders will ground Pin 2 like I did throughout this blog post. What if we AC couple Pin 2 to ground through capacitor C2?. This may help to better DC balance the input pair bases ( may reduce DC offset ) and perhaps even bypass some portion of the distortion to ground.
Does this work?
The bottom of the red line is the exact peak of the each tone with Pin 2 shunted to ground. Above the red line is the measured improvement for that particular tone caused by C2. I installed a switch across the C2 capacitor to make comparisons. . C2 = 0.01 Β΅F in this particular experiment.
In my experiments with a gain of >=150 and no feedback network, when the LM386 is pushed into harmonic distortion, C2 lowered the harmonic tones by 4 to 8 dB. I tried C2 values of 0.01 to 0.27 Β΅F
and changing the value of C2 within that range seemed to make no significant difference. Replacing C2 with a resistor of any value did not work to lower distortion.
C2 seemed to have less of an effect when the LM386 gain was lower than 150. At Gain = 20 with no feedback, I observed a maximum 2-3 dB maximal improvement in any 1 tone. With feedback, the effect diminished a little further, however, results were inconsistent. C2 does not appear to lower the harmonic distortion when the audio signal is unclipped -- rather, it seems to reduce distortion due to clipping when it happens.Β
I performed other experiments such as bringing the feedback to Pin 2 with Pin 2 connected to ground via a resistor or resistor + capacitor (like what you do with an op-amp or discrete AF amplifier). I also tried lowering the feedback 10K resistor value at various gain levels. Often enough, the result was that the LM386 would go into a writhing spasm when pushed into distortion. See below.
Above βFancy feedback experiments often resulted in the above tracing. It seemed better to exploreΒ simpler ways to lower distortion.
Wow, this was a lot of work, but proved fun. I encourage you to perform your own experiments with the LM386. While no panacea, and a little long in the tooth, the LM386 reflects a simpler, mostly analog time for many of us home builders.
I suggest you consider using the LM386 with lower gain and build up your audio signal voltage with a low noise preamp using an op-amp like the NE5532.Β
Further, consider adding feedback [ 10K plus some value of C1 ] from Pins 5 to 1 and also AC coupling Pin 2 to ground. I did both of these tricks in my 2 photographed bench CD player listening tests shown in Section 2.Β
γ¨γ¦γγγΒ