The last 12 months have been hectic and my hobby has suffered as a result. However, things are "hopefully" starting to calm down now and I can return to my beloved pastime.
I always book vacation time over Christmas and as it's a nice block of time, that's when I undertake my Christmas project. I did have something in mind for this year, but as usual, events overtook me.
Long story short, I made a new friend Dennis, and Dennis has a hearing problem. Whilst we were chatting he mentioned that for various reasons he didn't want to use a commercial hearing aid, so he'd tried to assemble one for himself using an off-the-shelf headphone amplifier but it wasn't really working. After inspection and a little test, the reason became obvious. It was designed for line-level audio input and not direct connection to a microphone.
Now I could just have added a pre-amplifier, but even then the performance wasn't going to be great. Since size isn't too much of a problem, it occurred to me that some of the available space could be used for adding some audio filters. This opened a complete can of worms.
Depending on the listeners situation, the filter requirements are going to be different. Maybe you're just wanting to enhance the speech audio from someone you are talking too face to face. Or maybe you want to filter out some background noise. Maybe you are at the opera and want to enhance the vocal range but subdue the music from the orchestra and it soon became apparent that this is the sort of project you can make as simple or complex as you want.
So, some type of adjustable filter would be required. A band-pass filter would probably be ideal. This would allow frequencies within a certain range to pass through, but attenuate frequencies that are either too high, or too low. To make it truly configurable, you need to be able to adjust the lower minimum frequency, and the upper maximum frequency thus leaving the pass frequency range or pass band in the middle.
Something with a preamplifier, band-pass filter, and headphone audio amplifier in a box with a rechargeable power source seems like it would do the job nicely. But how to design and test a suitable filter design.
I don't have any equipment that I can easily utilise to measure the performance or profile audio filters. My spectrum analyser has a sweep generator built it, so this is ideal... but it only works down to around 30 KHz which is wayyy too high for audio work. I need something that will go from near DC (0 Hz) to around 20 KHz.
I need an audio spectrum analyser; or Audio Analyser as they are called. They exist, and you can buy them, but when I saw the price I thought bugger that, I'll make my own.
The process to profile or measure the response characteristics of a filter is pretty simple. Inject a signal of a known frequency and amplitude into the filter input, and measure the amplitude at the output. Repeat that for a selection of frequencies from the filters minimum to maximum frequency range, record the results, plot them on a piece of graph paper and hey-presto, you get the filter profile, and yes, I "could" do it that way. But it's very labour intensive and if you are want to tweak your filter design and re-profile it, it starts getting very time consuming. What's needed is a gizmo to do the hard work for you.
More soon...
I always book vacation time over Christmas and as it's a nice block of time, that's when I undertake my Christmas project. I did have something in mind for this year, but as usual, events overtook me.
Long story short, I made a new friend Dennis, and Dennis has a hearing problem. Whilst we were chatting he mentioned that for various reasons he didn't want to use a commercial hearing aid, so he'd tried to assemble one for himself using an off-the-shelf headphone amplifier but it wasn't really working. After inspection and a little test, the reason became obvious. It was designed for line-level audio input and not direct connection to a microphone.
Now I could just have added a pre-amplifier, but even then the performance wasn't going to be great. Since size isn't too much of a problem, it occurred to me that some of the available space could be used for adding some audio filters. This opened a complete can of worms.
Depending on the listeners situation, the filter requirements are going to be different. Maybe you're just wanting to enhance the speech audio from someone you are talking too face to face. Or maybe you want to filter out some background noise. Maybe you are at the opera and want to enhance the vocal range but subdue the music from the orchestra and it soon became apparent that this is the sort of project you can make as simple or complex as you want.
So, some type of adjustable filter would be required. A band-pass filter would probably be ideal. This would allow frequencies within a certain range to pass through, but attenuate frequencies that are either too high, or too low. To make it truly configurable, you need to be able to adjust the lower minimum frequency, and the upper maximum frequency thus leaving the pass frequency range or pass band in the middle.
Something with a preamplifier, band-pass filter, and headphone audio amplifier in a box with a rechargeable power source seems like it would do the job nicely. But how to design and test a suitable filter design.
I don't have any equipment that I can easily utilise to measure the performance or profile audio filters. My spectrum analyser has a sweep generator built it, so this is ideal... but it only works down to around 30 KHz which is wayyy too high for audio work. I need something that will go from near DC (0 Hz) to around 20 KHz.
I need an audio spectrum analyser; or Audio Analyser as they are called. They exist, and you can buy them, but when I saw the price I thought bugger that, I'll make my own.
The process to profile or measure the response characteristics of a filter is pretty simple. Inject a signal of a known frequency and amplitude into the filter input, and measure the amplitude at the output. Repeat that for a selection of frequencies from the filters minimum to maximum frequency range, record the results, plot them on a piece of graph paper and hey-presto, you get the filter profile, and yes, I "could" do it that way. But it's very labour intensive and if you are want to tweak your filter design and re-profile it, it starts getting very time consuming. What's needed is a gizmo to do the hard work for you.
More soon...
RSS Feed