Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › M³ Project Announcement
New Posts  All Forums:Forum Nav:

M³ Project Announcement - Page 19

post #271 of 565
12/12/04 at 7:41pm
Quote:
Originally Posted by JohnFerrier
You'll need to verify that between positions 2 and 3 that power doesn't turn off... Make before brake.
JF
JF, thanks for the suggestion. There're several "shorting" rotary switches available at Mouser, e.g. 10WA367 and 10WR046. For my PPA, I'm even planning on putting the Linkwitz crossfeed and bass-boost all on another rotary switch (4P6T 6 position = 3 crossfeed levels X 2 bass-boost settings) . This way I will have one for the power and another for all the different effects. I think this idea could also be borrowed for M3 project, although I think it would be left to individual's preference.

I'd like to know if having so many wires cramped together would have some issue with cross talk. That's my only concern.
Reply
post #272 of 565
12/12/04 at 7:57pm
Quote:
Originally Posted by koladd
I'd like to know if having so many wires cramped together would have some issue with cross talk. That's my only concern.
It would be good to keep things seperate, especially power and signal. This will be less of a problem if you've got a good power source (with no ripple). Batteries are an excellent power source for audio (except for the need to recharge or replace...)

You seem like a very thoughtful person. It's good to think all this out before you start building...

Maybe you've seen the following link before. You can enter part numbers of all kinds of electronic parts that you are looking for (not just chips)...
http://www.findchips.com/
Maybe you found your switch this way as Mouser is the only one that carries it. Anyway I like that link and using Digikey's website...


JF
Reply
post #273 of 565
12/12/04 at 10:30pm
Thread Starter 
I suggest using a DPDT power switch, which connects the amp circuit either to the power source or a 1 Ohm resistor which rapidly discharges the onboard power caps, reducing or eliminating turn off noise.
Reply
post #274 of 565
12/12/04 at 11:51pm
Hi morsel,

The other day, I went back through some of my original posts and noticed that you posted several answers for me. Thanks. How was I to know who you were? That was before I knew morsel was morsel (so to speak). You had a different avatar then--it was a potatoe, or a gear, a mushroom, or a flower, or something...I can't remember...


JF
Reply
post #275 of 565
12/12/04 at 11:52pm
Quote:
Originally Posted by morsel
I suggest using a DPDT power switch, which connects the amp circuit either to the power source or a 1 Ohm resistor which rapidly discharges the onboard power caps, reducing or eliminating turn off noise.
Sounds great, Morsel. This will totally eliminate the turn off noise. I don't have an idea about the power rating of the resistor. I don't want to use a chunky resistor in this case. But I doubt if 1/4 W resistor will be able to handle the peak current at the instance of turning off. Aside from that, this solution would work only for power off pop elimination, but not for power on, if I'm thinking right.

JF, thanks for the link. I'm not aware of it before. I think I'll use it from now on whenever I need any parts. I happen to have a power supply with 0.25 mV RMS Ripple specified on the datasheet. I guess in this case it would not be a big issue with the crosstalk. You are right, shielding is almost impossible with these rotary swithces.
Reply
post #276 of 565
12/13/04 at 12:00am
Quote:
Originally Posted by koladd
But I doubt if 1/4 W resistor will be able to handle the peak current at the instance of turning off. Aside from that, this solution would work only for power off pop elimination, but not for power on, if I'm thinking right.
I = sqrt(P/R) = sqrt(0.25 W / 1 ohms) = 0.5A. This should work. It will only be a brief spike too...


JF
Reply
post #277 of 565
12/13/04 at 12:06am
Quote:
Originally Posted by JohnFerrier
I = sqrt(P/R) = sqrt(0.25 W / 1 ohms) = 0.5A. This should work. It will only be a brief spike too...
JF
You're probably right. However, P_peak=U^2/R=24^2 V/1 ohm=576W. That's the peak power at the moment of switching. My concern is if this would be bad for the resistor.
Reply
post #278 of 565
12/13/04 at 12:35am
Quote:
Originally Posted by koladd
You're probably right. However, P_peak=U^2/R=24^2 V/1 ohm=576W. That's the peak power at the moment of switching. My concern is if this would be bad for the resistor.
Actually, I was thinking that this resistor was used to short the headphone output when powered off (to prevent a turn-off pop). In this case, a spike would likely not exceed 0.5A by much, if at all...

I'm not sure. I guess I wouldn't do it this way...

I like seperate power and signal switches. But, my amp is not a portable rig. It's easy for me (by design) to get into the habit of a power on/power off sequence.


JF
Reply
post #279 of 565
12/13/04 at 12:56am
I guess I was wrong. The resistor rating should be RMS, so a transient probabily won't matter at all. But since during the switch off, the capacitors and 1 Ohm resistor from a R-C loop with initial voltage 24V due to the charge in capacitor, the peak power calculation would still be correct, theoretically.

I was actually impressed by your design, in which the switch are laid out in such a way that it would be easy to form a habit. Nice discussion with you though. I have something in mind (some other idea) but I'll shut up for the moment until I test it out.
Reply
post #280 of 565
12/13/04 at 1:30pm
Thread Starter 
Alternatively, one could follow the time-honored practice of turning the volume all the way down before powering off.

JF, your memory is better than mine. I was ready to deny having a different avatar but then I recalled that it was indeed a hot chile pepper.
Reply
post #281 of 565
12/13/04 at 2:18pm
A 4P3T switch between amp and hp jack will introduce it's own turn-on/turn-off clicks and noises. And it will be in the circuit when the amp's in use, so it has to be a high quality switch. If you're switching the AC along with the output, you'll have an AC field jus ta fraction of an inch away from the audio output.

- Eric
Reply
post #282 of 565
12/13/04 at 2:33pm
Quote:
Originally Posted by morsel
JF, your memory is better than mine. I was ready to deny having a different avatar but then I recalled that it was indeed a hot chile pepper.
I was trying to surprise you and find it at the following site: http://web.archive.org/web/*/http://www.head-fi.org (Actually, I was going to try and trick you by posting it with the message "Is someone's firewall down?"...to test your steel...) Archive.org is great for finding info (and graphics sometimes) from past websites... Unfortunately, in head-fi's case, they link to the current avatars...

Take care.


JF
Reply
post #283 of 565
12/13/04 at 2:40pm
Quote:
Originally Posted by eweitzman
A 4P3T switch between amp and hp jack will introduce it's own turn-on/turn-off clicks and noises. And it will be in the circuit when the amp's in use, so it has to be a high quality switch. If you're switching the AC along with the output, you'll have an AC field jus ta fraction of an inch away from the audio output.

- Eric
Eric, the hp jack is disconnected when the the power is switched on. The idea of using rotary switch is that you heaphone is not in a closed circuit when the power is switched on. The switch is for DC power source and it's a shorting type . If you short the position 2 and 3 (power is switched on from position 1 to 2) associated with the power poles on the rotary switch, the power will remain connected when the headphone is switched on (from position 2 to 3). I don't see where this would introduce clickes due to switching. I agree that a high quality swtich needs to be used. I'm going right now for Lorlin switches. The power rating should be good. I hope it works well.
Reply
post #284 of 565
12/13/04 at 3:11pm
koladd,

I understood the purpose of the sequential switching. It's clever: I think I might use it in my next amp.

When you use switch position 3, there will be clicks on turn-on and turn-off because of (1) contact bounce (2) load changes on the amp and (3) double load changes on the amp: the M³ has an active circuit on both hot and "ground" so you have two amps per channel being loaded and unloaded, and maybe not at precisely the same moment either because no switch is perfect. Two bounces/load changes per on, two per off. Per channel.

I made a passive linestage. I thought I'd be clever, and wired the input selector to switch both hots and grounds to reduce crosstalk. (It's a milspec surplus Greyhills 4P6T switch.) Switching the grounds had the nasty side effect of permitting slight DC offsets in the disconnected components: some of them made little noises when switched on/off. I went back to a traditional approach.

These noises would probably be no worse than what you'll get inserting and removing the headphone jack, though. But since your goal is to eliminate the noises ... nevermind ...

- Eric
Reply
post #285 of 565
12/13/04 at 7:41pm
Quote:
Originally Posted by eweitzman
koladd,
there will be clicks on turn-on and turn-off because of (1) contact bounce (2) load changes on the amp and (3) double load changes on the amp: the M³ has an active circuit on both hot and "ground" so you have two amps per channel being loaded and unloaded, and maybe not at precisely the same moment either because no switch is perfect. Two bounces/load changes per on, two per off. Per channel
This makes sense to me. I haven't thought about the load change as a factor. My current switch for headphone (since I haven't implemented the rotary switch idea yet) is DPST rocker switch for the headphone outputs. Frankly, I can't hear the pop when I switch them on, given that the power is already switched on. I believe they do exist, but the level must be much lower than the pop due to the power switch on. Would you think that connect some resistors (with lower resistance than the headphone) between poles for headphone output and position 2 will give a less dramatic change in load when the rotary switch goes from position 2 to 3, since the step change in load is smaller from a buffer's perspective?
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › M³ Project Announcement