just for thought... someone discussed that a good amp gives a sense of depth, like a 3-d effect. In my opinion, 3-d effect is caused by the stereo sound which we hear, that means two separate chanels. Left+ right tells us the exact distance and location of the noise source. Now, a question is that does crosstalk reduces this stereo effect? I meant a bit of signal leaked into the other channel, at a large degree it almost cancels out the stereo sound.
If sharing the same power supply for two chanels, how much is the signal leak? how about using stereo potentialmeters? I know that if i touch my potentialmeters, I get buzzing from my amp. This means that I am feeding my amp with signals, so that means possibilities of crosstalk can also happen there.
any comments please?
If sharing the same power supply for two chanels, how much is the signal leak? how about using stereo potentialmeters? I know that if i touch my potentialmeters, I get buzzing from my amp. This means that I am feeding my amp with signals, so that means possibilities of crosstalk can also happen there.
any comments please?
Well I tested my amp with RMAA and it got a very poor rating for crosstalk (something around 43DB) I did some crude tests, and yes I could hear channel leakage. I did this by running 1 channel into a dummy load, the other channel had no input but had a speaker attached. I could here the music at a very low level from the speaker attached to the channel with no input. However this was at such a low level I had to have my ear near the speaker to hear it.
So my guess (and it is very much a guess) would be that while crosstalk could cause subtle degradation of fine details, it probably wouldn't have much of an effect on the stereo imaging unless it was truely awfull!
Tony.
So my guess (and it is very much a guess) would be that while crosstalk could cause subtle degradation of fine details, it probably wouldn't have much of an effect on the stereo imaging unless it was truely awfull!
Tony.
wintermute said:I did this by running 1 channel into a dummy load, the other channel had no input but had a speaker attached.
Hi,
was this input terminated with source output impedance during test?
Regards
Most likely powersupply. I rewired it using heavy gauge wire (8gauge) and the crosstalk figures improved by about 6db I think it went from very poor to poor 🙂
I did something else which also seemed to improve it by about 6db but can't remember what it was...... Just searched and it was shortening the length of wire which went to my load resistors, that improved the cross talk by 6db as well. It was originally down around 30db, but with the two changes got it to 43db.
I have since tested with my marantz DV18 as the source for the RMAA test tones and got 57db at approximately 25W and 64db at approx 5W. I didn't measure the voltage when doing the previous tests, but I think that it would have been similar. One possibility for the better results with the DV18 as source, is that I think the amp was oscilating when connected to the sound card, which could have been stressing the Power supply.
all here if you want to look......
http://www.diyaudio.com/forums/show...&perpage=15&highlight=wintermute&pagenumber=1
Tony.
I did something else which also seemed to improve it by about 6db but can't remember what it was...... Just searched and it was shortening the length of wire which went to my load resistors, that improved the cross talk by 6db as well. It was originally down around 30db, but with the two changes got it to 43db.
I have since tested with my marantz DV18 as the source for the RMAA test tones and got 57db at approximately 25W and 64db at approx 5W. I didn't measure the voltage when doing the previous tests, but I think that it would have been similar. One possibility for the better results with the DV18 as source, is that I think the amp was oscilating when connected to the sound card, which could have been stressing the Power supply.
all here if you want to look......
http://www.diyaudio.com/forums/show...&perpage=15&highlight=wintermute&pagenumber=1
Tony.
- Status
- Not open for further replies.