I needed some music for working in the basement office for my tiny-budget internet startup and Instead of the usual beige/black box computer speakers I got one of those funny T-amps from ebay powering a pair of KEF Cresta 10s I found at a garage sale.
So far, so cheap. My bright idea was to make a cable to use the 12v "+4" from my computer's "20+4" connector to power the amp since it wasn't being used for anything. I like this idea. There are far too many mains adaptors under my desk already (monitor, printer, scanner, laptop, mouse!) and the computer PSU is probably more efficient than the average wall wart.
Now I've made the cable I've discovered audiable interference from the power rail isn't always just the prefered subject of people who buy $5,000 mains cables. Moving the mouse with the amp on sounds exactly like the thumps and tweeks the speccy +2a made connected to a TV.
(going heavy on UNIX pipes makes a sound a lot like bad plumbing, and compiz drives it nuts!)
I'm pretty sure it's distortion from the power line and not the sigal because with the amp powered by eight AAs the background is dead silent. My geek intuition tells me the polution on the power rails could be filtered with some combination of capacitors and chokes, but that's about where my knowledge ends. If pressed, I'd say the former in parallel and the later in series, but really I'm just guessing. I'm a software engineer, and know really very little about electronics. Searching brought up some interesting pages, but mostly things that assume too much knowledge.
Just to prove I've Googled high and low, here are some of the pages I've found:
http://en.wikipedia.org/wiki/Low-pass_filter
http://www.latticesemi.com/lit/docs/technotes/tn1068.pdf
http://www.williamson-labs.com/480_dec.htm
Can it be done with a simple network?
So far, so cheap. My bright idea was to make a cable to use the 12v "+4" from my computer's "20+4" connector to power the amp since it wasn't being used for anything. I like this idea. There are far too many mains adaptors under my desk already (monitor, printer, scanner, laptop, mouse!) and the computer PSU is probably more efficient than the average wall wart.
Now I've made the cable I've discovered audiable interference from the power rail isn't always just the prefered subject of people who buy $5,000 mains cables. Moving the mouse with the amp on sounds exactly like the thumps and tweeks the speccy +2a made connected to a TV.
(going heavy on UNIX pipes makes a sound a lot like bad plumbing, and compiz drives it nuts!)
I'm pretty sure it's distortion from the power line and not the sigal because with the amp powered by eight AAs the background is dead silent. My geek intuition tells me the polution on the power rails could be filtered with some combination of capacitors and chokes, but that's about where my knowledge ends. If pressed, I'd say the former in parallel and the later in series, but really I'm just guessing. I'm a software engineer, and know really very little about electronics. Searching brought up some interesting pages, but mostly things that assume too much knowledge.
Just to prove I've Googled high and low, here are some of the pages I've found:
http://en.wikipedia.org/wiki/Low-pass_filter
http://www.latticesemi.com/lit/docs/technotes/tn1068.pdf
http://www.williamson-labs.com/480_dec.htm
Can it be done with a simple network?
Just put a 1000uF cap across the power lines positive to negative (so in parallel) and see if that helps.
OK. That's what I thought (although it feels wrong soldering things across the +ve to -ve wires!)
I'll probably be buying the components from Maplin.
Item AT20W (see here) is a 1000uF capacitor.
Because delivery is likely to cost more than the actual components, I'd prefer to order everything I might need at once, and have stuff left over than order twice.
Would it be worth getting any other things too? Most things I find on Google use three capacitors (usually 10uF, 1uF and 0.1uF) and some resistors or inductors.
Will the cap act as a low-pass filter? I'm supposing that DC can be seen as the lowest possible frequency, so a low-pass filter at a very low frequency will remove the noise? That seems to make *some* sense to my confused mind!
I'll probably be buying the components from Maplin.
Item AT20W (see here) is a 1000uF capacitor.
Because delivery is likely to cost more than the actual components, I'd prefer to order everything I might need at once, and have stuff left over than order twice.
Would it be worth getting any other things too? Most things I find on Google use three capacitors (usually 10uF, 1uF and 0.1uF) and some resistors or inductors.
Will the cap act as a low-pass filter? I'm supposing that DC can be seen as the lowest possible frequency, so a low-pass filter at a very low frequency will remove the noise? That seems to make *some* sense to my confused mind!
So what happens if you disconnect the input to the T-amp? Just powering it from the computer? Maybe driving it from another signal source. It goes silent doesn't it? If it does capacitors won't help. You need to do something about that ground loop in that case. Maybe some kind of common-mode choke would help.
I have a computer monitor with speakers. The speakers are powered from the monitor. I get all kinds of weird noises out of the speakers... maybe it's an alien civilization trying to contact me? Yay! I'll be famous! 😀
I have a computer monitor with speakers. The speakers are powered from the monitor. I get all kinds of weird noises out of the speakers... maybe it's an alien civilization trying to contact me? Yay! I'll be famous! 😀
I just did an experiment. The only other signal source I have handy is my laptop (I have a standalone CDP, but that is nested in a jungle of wires)
So, with the T-amp powered by the desktop PSU, and the signal from the laptop, I get exactly the same effect (moving mouse on desktop is audiable) but ONLY if the laptop is plugged in!
If the laptop is running from the battery, the background is dead silent!
Now I'm *really* confused!
Maybe I should just get a mains adaptor... *sigh* yet another more box under the desk to warm my feet on in winter I suppose. Also, I'm skint so this isn't my favourite option.
Unless this ground loop problem can be fairly easily dealt with, I'll probably just get an adaptor though. A separate adaptor would fix it, right?
BTW, Now I know how all the newbs I talk to on programming forums I hang out on feel. Please try to bear with me 🙂
So, with the T-amp powered by the desktop PSU, and the signal from the laptop, I get exactly the same effect (moving mouse on desktop is audiable) but ONLY if the laptop is plugged in!
If the laptop is running from the battery, the background is dead silent!
Now I'm *really* confused!
Maybe I should just get a mains adaptor... *sigh* yet another more box under the desk to warm my feet on in winter I suppose. Also, I'm skint so this isn't my favourite option.
Unless this ground loop problem can be fairly easily dealt with, I'll probably just get an adaptor though. A separate adaptor would fix it, right?
BTW, Now I know how all the newbs I talk to on programming forums I hang out on feel. Please try to bear with me 🙂
Hi Jim,
add Rapid Electronics and Dalbani (for difficult semicons) to your retailer's list.
www.rapidonline.com
www.dalbani.co.uk
add Rapid Electronics and Dalbani (for difficult semicons) to your retailer's list.
www.rapidonline.com
www.dalbani.co.uk
Ah, it's clear I'm in way over my head.
I still really like the idea of powering it from the computer (another neat advantage would be that the amp will turn off when I switch the computer off!).
Hardly the words of someone normally considered a hacker by his peers, but *sigh*, I'll stick to programming - this stuff is just too low level for me!
I have a rough idea that the ground on the signal and the power being the same is causing a ground loop, but that's about as far as I get.
For example, I have no idea which of these I would use for a common mode choke, or even if that would really solve my problems:
http://www.rapidonline.com/searchresults.aspx?style=0&productNo=&kw=choke&Manucode=&supplierID=All
Probably, a nice mains adaptor it is then. The 3A one on this page looks OK:
http://www.maplin.co.uk/Search.aspx?criteria=XM20W&doy=7m8&source=15
However, it is 13.8v, but having searched the forums here, a few other people have reported using it without any problems. Probably, I'll add a heatsink to the amp chip to be sure.
I'll still be reboxing the t-amp, mind - that at least can be done with just a bit of dumb soldering 🙂
I still really like the idea of powering it from the computer (another neat advantage would be that the amp will turn off when I switch the computer off!).
Hardly the words of someone normally considered a hacker by his peers, but *sigh*, I'll stick to programming - this stuff is just too low level for me!
I have a rough idea that the ground on the signal and the power being the same is causing a ground loop, but that's about as far as I get.
For example, I have no idea which of these I would use for a common mode choke, or even if that would really solve my problems:
http://www.rapidonline.com/searchresults.aspx?style=0&productNo=&kw=choke&Manucode=&supplierID=All
Probably, a nice mains adaptor it is then. The 3A one on this page looks OK:
http://www.maplin.co.uk/Search.aspx?criteria=XM20W&doy=7m8&source=15
However, it is 13.8v, but having searched the forums here, a few other people have reported using it without any problems. Probably, I'll add a heatsink to the amp chip to be sure.
I'll still be reboxing the t-amp, mind - that at least can be done with just a bit of dumb soldering 🙂
- Status
- Not open for further replies.