Originally Posted by gsferrari Back on topic - I have added certain people to my ignore list and anyone who wants to proceed with a sensible discussion will be advised to do so.
I have not yet looked for the 3PDT switch but will be back in a few minutes with the part numbers and prices. I dont think a group buy is necessary - lets just set up a guide for anyone who wants to do the same project with ordering details etc.
That will be useful.
I apologize that this thread got messed up despite my best intentions and despite my requests in the beginning to keep things clean. But sometimes, just as in real life, you are never completely in control.
sovkiller already posted 2 numbers, that post probably got lost in the shuffle, so I'll repost them here:
mouser part number 690-A323S1YZQ
digikey part number ckn1037-nd
Another 3pdt switch I found (for the cheapskates among us) is 108-1M31T1B1M1QE at mouser
Mouser # mouser part number 690-A323S1YZQ is what KG also recommended. Just lost in this thread. Someone will clean it up hopefully.
this costs $10.00 for a single piece.
The Digikey alternative : ckn1037-nd costs $14.00
I am going to get the Digikey version.
Anyway - I think this part is good to go.
Any suggestions for cables?
Edwood - you make cables right? - can you make a shielded pair terminated at 1/4" on one end and tinned at the other end? Will the homegrown 3 strand wire work for this purpose?
Originally Posted by gsferrari Any suggestions for cables?
Edwood - you make cables - can you make a shielded pair terminated at 1/4" on one end and tinned at the other end? Will the homegrown 3 strand wire work for this purpose?
Cheers!
If you really must go silver, Homegrown Audio is a waste of money. Unless you are going with unshielded twisted pair configuration (each pair of wire must be evenly twisted with signal and ground for each pair, or it won't work) it is not worth going with pre twisted configuration. With twisted pair, it's not really necessary for shielding, especially in this application.
I've had great experience with just plain Canare Starquad. It is shielded with a tinned braid. Anyone recommend whether or not to solder the shield to the ground on the 1/4" Plugs (the two that go into the two amps) and (or not) to the ground on the other end of the wire terminated at the switch? Canare L-4E6S Star Quad 4-Conductor Microphone Cable.
I'll have to check if any of the Star Quad Wires are twisted or not. There are four stranded copper wires in there, of course.
OK - if we use insulated jacks and a metal case, can we ground the shield to the metal case and leave the circuit grounds isolated from this? That will be a equipotential surface on the case, shielding and leave the signal wires alone.
Thanks for the links Sov - checking them out now...
Originally Posted by gsferrari OK - if we use insulated jacks and a metal case, can we ground the shield to the metal case and leave the circuit grounds isolated from this? That will be a equipotential surface on the case, shielding and leave the signal wires alone.
Thanks for the links Sov - checking them out now...
If you really must connect the case to a ground (not necessary in my opinion), then ground you're connecting to should also be switched by the switch to correspond to the active amp. I suggest you use shielded cables all the way up to the 3 pole switch and forget about grounding the enclosure.
Btw, balanced equipment exits for some good reasons, one of which is this: with balanced equipment, you can connect all (case) grounds together no problemo... Also see, why not wye: http://headwize.com/tech/wye_tech.htm
These look good. but for the moment I am sticking to Dr. Gilmore's suggestion and going with the mouser part.
I am still not clear about cables. Do I or dont I have to ground the shielding? Can I use un-shielded wire? or shielded cable without grounding the shielding?
Proved that despite its huge size the CD3000 can be shoved down one's throat.
Joined
Jul 31, 2002
Posts
12,902
Likes
28
Quote:
Originally Posted by gsferrari I dont think this will work in this project...looks like a foot-switch and I think it has to be held down - no toggle.
Some foot switches do not require to be pressed all the time, and are just that, switches, and some are not even so hard to push, and you can do itwith the hand easily....one thing is sure they are rough enough to support abuse.
Originally Posted by gsferrari Does this mean I dont have to connect the shield to the ground?
If you're talking about the shield in the TRS connector, you do connect it to the corresponding position in 3rd pole of the switch, but not to the metal box. Doing that would defeat the purpose of the 3rd pole.
Don't think in terms of "ground" when building this. You don't have common ground here. For that you'd need balanced cables. What you have are two signal sources, each with 3 signals (one happens to be called ground, but it's a signal wire). All you want is to connect all 3 simultaneously to the headphone, and at the same time float the 3 wires from the other amp.
The simplest design is to have the box floating, so it doesn't have to be metal. If you want to really "ground" the box, i.e. make it part of the signal wires, then connect it to the headphone ground (i.e. jack sleeve). That way, when you flip the switch the box is grounded to whatever ground is "active". I'm not sure that this would be better than floating the box. It may be worse. People experienced with RF should probably comment on this.
Btw, if you use a 3 postion switch (i.e. source A, source B and off), you can probably do a sort of double blind test if someone else controls the switch. You won't know after an "off" if the sources have been switched or not.
So ON-OFF-ON right? Yeah - that is a good point. I have decided to use un-shielded cable of short length and follow the second circuit in my original post. I will then experiment and learn as I progress...need to get started somewhere.
Originally Posted by gsferrari So ON-OFF-ON right?
Yes. That's what I mean.
I theory you can even to an ABX test with this. For each trial the experimenter would have to play first A, then B and finally randomly A or B. The problem is that the random selection is done by a human.
I'm sure Hirsch knows far more about double blind tests than I do.
The most difficult part I see here is not building the switch, but matching the levels of the contesting amps. There is a popular conception that a minute difference in levels can influence the perception of what sounds "better".
I see at least three ways, in increasing order of difficulty: (i) measure the voltage w/o load, (ii) measure the current flow into the phones, and (iii) measure the actual sound output. Maybe you want to start another thread on this topic...
Shouldnt be too hard to measure outputs with a DMM and match the amps...although you can also do this by ear.
I will have a CRO at the HeadFi meet where I plan to take this switchbox.
I am committed - just ordered the Hammond Case and the 3PDT switch from Mouser. Now I need to get the cables...I will use solid silver hookup wire and keep everything as short as possible.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.