16-05-2020, 07:12 PM
See the attachment. While it is a form of T switching, it's more commonly known as "double switching" or "dual switching".
For stereo, the difference in price (and space taken) of 4 pole switches vs 2 pole is minimal, so it's a good technique. The improvement is of the order of 20dB at 10kHz, depending on the various resistances involved. Better than even a rotary switch with ever other position grounded.
I claim no originality, but I did this about 20 years before I saw anyone else mention this particular form (this extract is from "Self on Audio", published in 2010 IIRC). This was before the internet, of course, and in those days access to schematics and even good text books and magazines was very limited to a schoolboy at a rough comp with no income who lived in the middle of nowhere. Back then, I was probably a lot more creative as a result - now you just ask the internet, but of course you don't learn so much from simply copying circuits...
Anyway, I built a simple 2 input video switch and 3 output distribution amplifier, and used a 4053 to do the switching. I didn't bother testing how well a single section of it performed, so can't say how much of an improvement it brought, but as I had 3 SPDT switches in the package, and no application for the other 2, I thought I might as well double-switch. Despite being built on 0.1" matrix board, it worked well, with excellent bandwidth given the simplicity of the design.
I still have it in the attic - it was for my NICAM tuner, which had the useful option of VCR input via a Scart socket. As the pictures on my website are so poor, I have been planning to dig it out and take some new pictures with a camera that was built this millennium! At that point, I'll run some measurements - it'll be interesting to see how I did all those year ago. We have a test signal generator at work which can generate NICAM, so I'll also investigate the audio performance, which won't be stellar, as I was only 17 when I designed it. Needless to say, the NICAM bits were not my design!
For stereo, the difference in price (and space taken) of 4 pole switches vs 2 pole is minimal, so it's a good technique. The improvement is of the order of 20dB at 10kHz, depending on the various resistances involved. Better than even a rotary switch with ever other position grounded.
I claim no originality, but I did this about 20 years before I saw anyone else mention this particular form (this extract is from "Self on Audio", published in 2010 IIRC). This was before the internet, of course, and in those days access to schematics and even good text books and magazines was very limited to a schoolboy at a rough comp with no income who lived in the middle of nowhere. Back then, I was probably a lot more creative as a result - now you just ask the internet, but of course you don't learn so much from simply copying circuits...
Anyway, I built a simple 2 input video switch and 3 output distribution amplifier, and used a 4053 to do the switching. I didn't bother testing how well a single section of it performed, so can't say how much of an improvement it brought, but as I had 3 SPDT switches in the package, and no application for the other 2, I thought I might as well double-switch. Despite being built on 0.1" matrix board, it worked well, with excellent bandwidth given the simplicity of the design.
I still have it in the attic - it was for my NICAM tuner, which had the useful option of VCR input via a Scart socket. As the pictures on my website are so poor, I have been planning to dig it out and take some new pictures with a camera that was built this millennium! At that point, I'll run some measurements - it'll be interesting to see how I did all those year ago. We have a test signal generator at work which can generate NICAM, so I'll also investigate the audio performance, which won't be stellar, as I was only 17 when I designed it. Needless to say, the NICAM bits were not my design!
