I believe this should work. Anyone want to test it? I don't have any MC14016's around. Any 4016 device will work.
(http://experimentalistsanonymous.com/other/TBSPDTLED.gif)
-Colin
:::Scratches head::: Alittle more clearer labeling might be nice but I think I know where your going with this. I'll play with it alittle later but I can see some possible "improvements". Just letting you know that this is not being ignored.
Andrew
You might want to try something more like this....
(http://64.7.66.217/circuits/4016.gif)
In this example, one of the switches is wired as an inverter which makes the transfer time between on and off almost instantaneous. It also reduces the switch requirements even more so you only need an SPST now. Another advantage of doing things this way is that the LED current is not going through the 4016 which will make for a cleaner signal environment.
Take care,
-Peter
PS: You're still making me scroll horizontally. :(
Interesting, Peter, I've never seen a design like that. I was hoping to get along to SPST some time soon. I didn't realize the switching wouldn't be instantaneous.
Sorry about making you scroll. I like big schematics, I can't lie.
Hey, could you change my nick to EAColin? The fact that I make the poster-column on the main page bigger has always bugged me, and I like having my name in my nick. It'd also help a tiny bit with the scrolling.
-Colin
One source of noise in switching is the time in-between the two sources being engaged. It's not very much, but it's sure there. If you invert the signal from a single control line, you'll get 1/2 as much switch bounce too. The bounce you do get will act a little like a pulse-width-modulated (PWM) blend between the two signals. It would be a good idea to bypass the switch with a small cap too. (maybe 0.1uF) You could also add a D type flip-flop to make an alternating on/off with a momentary switch. Lots of options.
On your handle.... I don't mind scrolling in threads where schematics are posted at all. That's not the issue.... it comes with the territory. It's every other thread that you post in. If you shorten your sig graphic to 300 pixels wide like this...
(http://64.7.66.217/circuits/c1.jpg), or this
(http://64.7.66.217/circuits/c2.jpg), or this
(http://64.7.66.217/circuits/c3.jpg), then is won’t cause people who generally use less screen real estate to have to scroll horizontally (like puretube and myself for starters). Also remember that different browsers on different platforms have different deficiencies. When you design for the web, you have accommodate the lowest common denominator.
Take care,
-Peter
BTW: I'm just a mod and not an admin, so I couldn't change your handle anyway. Thanks for asking though... If you really do want to go that route, another option would be to just add a space (I think that would work), but I think it's fine as it is. Also... if you like, you can do anything you want with that snippet. ;)
Quote from: Peter SnowbergYou might want to try something more like this....
(http://64.7.66.217/circuits/4016.gif)
Peter:
Could a 4066 be subbed for the 4016? I've fought the 4053 battle and lost. I struggled with the Wicked Switch over the weekend and failed. This looks like something I could handle.
Hi David,
You could... but from what other's experiences have been, your mileage may vary. Andrew has a LOT more experience in that area than I do.
Honestly, I do the 3PDT bypass thing or use H11F1s (expensive when you use 3!). I've seen enough struggle to avoid the CMOS switches for effects so far. ;).
Danelectro apparently uses the 4053 so it can be done.... somehow. The enigma continues.
Best of luck,
-Peter
some people (who got a lotta juice from their wallwarts...) use good small signal relays and are happy with it...
interesting stuff, is this technique generally considered "true" bypass or is there some difference in tone?
thanks
Good question.
I would call this more of a "semi-true" bypass because the signal is still forced through a couple of caps and the headroom is limited by the CMOS switch. I think it's in a gray area between true and tone-sucking.
In one definition... the signal is lot being loaded by a small input resistor all the time or subject to filtering from that resistor and the input cap (small value) is, so in that sense it is true, but the bias resistors will still load it so it's a tradeoff.
Maybe it's "true" in the sense that fast food is still "food". :? It's not a real restaurant though in my book.
Take care,
-Peter
...nor "homecooking" !
to be serious: it is not T.B., coz it is "semi-conducted", i.e. it goes thru some pn or np semiconductor stuff that needs an operating voltage to be able to fully conduct, and not thru physical conductive material like copper/silver/gold/aluminum... you name`em.
Isn't that sort of how Dunlop does the switching in their late-model wahs? I've got a 'Jimi Hendrix Fuzz-Wah' that uses a SPST for the wah on/off, and it has 3 monster MC14016 ICs on the board. By the way, I took out the smelly old guts and built a vox-type circuit with a red fasel to put in it.
Regards
RDV
Aside from not being true bypass, it seems like the savings aren't even that much, at least for us diyers. At a quick glance... Small Bear has a Carling SPST for $6, which is the same price aron sells his 3PDT's for. I assume there is much more benefit in large quantities or the big companies would just use 3PDT's too.
QuoteAndrew has a LOT more experience in that area than I do.
I heard my name used in a non-cursing manner which is a rare thing for me so I'll chip in what I can.
QuoteCould a 4066 be subbed for the 4016? I've fought the 4053 battle and lost. I struggled with the Wicked Switch over the weekend and failed. This looks like something I could handle.
In this scenario you could sub a 4066 for a 4016. Its not like the 4016 flip-flop switch that has been disscussed before that takes advantage of the internal transistor layout of the switch.
This circuit is more optimized just for bypass switching. I thought about releasing a prewired Wicked Switch that is meant for bypassing only but decided to give people the option to do what they want. The WS are meant as general switches to do the same thing as any DPDT switch whould so you could even replace mini-toggles and the like.
I've been thinking about playing with the 4053 to see what it can do. I also need to do some updates to the WS article. Theres a bunch of stuff I didn't get a chance to add on the first go.
If you have had problems with the WS your welcome to post a message, PM, or email me and I'll help out where I can. I have had many successful build reports.
Quoteinteresting stuff, is this technique generally considered "true" bypass or is there some difference in tone?
It depends on how much of a "tone purist" your are or how you define "true bypass". In the most technical manner it is not. In an audio manner I think its good enough. I haven't had a client complain about the switching in the stuff I've built, just the sound of the effect. :roll: What I belive is that you'll get as close as possible to true bypass before going to mechincal means.
I will also point out that this circuit does not allow for the alternate effect input grounding on bypass method of switching wiring since you are giving up a switch to do the inversion.
When I was playing around with solid state switching I did something like this in the beginning. Peter did some things in his schematic that I was going to point out. Mainly that you should not let the switching control pins float and hope that they hit the state you want. You should use pull up/down resistors to make sure the state you want is set. The debouncing is a good idea too.
I did another variation on this by using pull up resistors on each set control signals and using a SPDT ground the desired control pins. It gave access to all four switches.
With this type of switching as the battery voltage drops the voltage to the control pins will also drop poossibly to the point of the switch not operating properly. Its switching performance can be dependant on the input voltage meaning that unless it's voltage is at near logic levels the switch may not open/close fully especially with the 4016. It may open/close partially creating noise and unwanted resistance. Sometimes the switch will even oscilate between states when the voltage hits the right point. Just a word of warning with this type of switching.
Ah I'm typed out.
Andrew
Quote from: TriffidAside from not being true bypass, it seems like the savings aren't even that much, at least for us diyers. At a quick glance... Small Bear has a Carling SPST for $6, which is the same price aron sells his 3PDT's for. I assume there is much more benefit in large quantities or the big companies would just use 3PDT's too.
Cost wise its about the same in my opinion. The cost of the switch and IC are about the same as a footswitch.
There some places where you can't get mechincal DPDT switches. Using solid state swtiching allows for any kind of switch to be used. You can do long range switching easier with little sound degradation or noise. You are not at the mercy of switch contact issues affecting your audio signal. If your switch fails you can replace it easily without having to hack your effect up or your audio signal suffering.
I'm not saying its perfect and there are downsides that I'm sure everyone can see but its should be given more credit then it has gotten. It does solve some issues.
Andrew
Andrew:
I'm not dissing the "Wicked Switch" design. When I worked on it Friday, I couldn't figure out to save my life how to interface an effect to it. That seems to be due to the cold medication I took Friday night, not your design. The DPDT switch representation you incorporated into your diagram threw me off. I went back and traced the circuit paths with crayons (much to my daughter's disgust) and I think I understand it now. Peter's trick appealed to me because I like to use as few components as possible. I've got a WS on breadboard that I'm planning to hook up to a Flatline tonight.
Is there any possibility that you would consider doing a layout for the SPST / inverter model of WS on perf?
QuoteDanelectro apparently uses the 4053 so it can be done.... somehow. The enigma continues.
I did try a design with the 4053 that did sound OK.
http://members.shaw.ca/roma/switching.html
Zeta:
I tried one of those too. Which one did you get to work?
The one that says "Using the CMOS 4053 " from the same url.
"True bypass" as it's commonly used means that when the effect is bypassed, there is no connection to the effect signal path at all, not even through turned-off solid state switches. So it's not true bypass.
However, careful design of CMOS switches can be quite good. See the articles on bypassing at GEO -
- bypassing with the CD4053
- the technology of bypasses
QuoteDanelectro apparently uses the 4053 so it can be done.... somehow. The enigma continues.
Actually, it doesn't. The bypassing with the CD4053 at GEO is how Danelectro does it.
Buffered bypass or something like the Clinton bypass comes closest to true bypass with an SPDT.
R.G. -
Since you're weighing in on this, perhaps you can help clear something up.
I'm not trying to question you, or put you on the spot, but after reading your article about the CD5053 bypass, I was... mostly stumped. I didn't get what you were doing with the biasing. I investigated other options: the Wicked Switch, etc. I got down to a 4053 implementation (the one Zeta did) from Laurier Gendron. It didn't work (well, not yet). I saw some comments you posted about the 4053 having a limit of 7 volts on something. I'm about ready to give up on electronic switching forever!
A couple of questions, if I may?
What does the 7 volts reference? It's not VDD, is it?
Could this thing be biased with your 386 trick?
Quote from: DavidI'm not dissing the "Wicked Switch" design. When I worked on it Friday, I couldn't figure out to save my life how to interface an effect to it. That seems to be due to the cold medication I took Friday night, not your design. The DPDT switch representation you incorporated into your diagram threw me off. I went back and traced the circuit paths with crayons (much to my daughter's disgust) and I think I understand it now. Peter's trick appealed to me because I like to use as few components as possible. I've got a WS on breadboard that I'm planning to hook up to a Flatline tonight.
Is there any possibility that you would consider doing a layout for the SPST / inverter model of WS on perf?
Ah its ok to knock it, its not really much of a design. I too like the one IC design but it does have some issue that have been pointed out. It another tool in the box. I did the WS that way because of certain design parameters I originally started with when I started playing with them. If you do have any question let me know.
I have designed a WS that has a pair of SPDT switches in one board. Each switch used half of the 4066 and the logic ICs were used to make a pair of control circuits. I thought posting those would be too much for people.
Get well. Being sick sucks.
Andrew