Electronic Bypass
- mictester
- Old Solderhand
Information
I found that I had a lot of single pole foot switches, and a lot of low current bicolour LEDs. I've used 4066s for years in audio mixers (for switching PFL) and have a lot of those too. I thought that it might be a good idea to try the 4066 for silent FX switching. The audio paths are biased to half the supply rail - this is common with op-amp circuits, and I found that it helped the 4066 to operate without an audible "pop" when switching. I used all four of the switches in the 4066 - I tried just using two, but there was a little breakthrough. The 4066 shows about 20 Ohms through it when on, and several Megohms when off. The "switch contact" connections on the 4066 are reversible, so laying out the PCB is easy.
As you can see, it's literally a "back of an envelope" design scribble! I'll draw up a better version electronically, and a suggested PCB / stripboard layout later this week.
How does it sound? Like it's not there (which is probably its best recommendation).
Possible modifications - the op-amps are set for unity gain, but in reality there will be a (minimal) loss. With the addition of a couple of resistors and capacitors per op-amp, you could easily introduce some gain to get exactly the same level out as in (I found it wasn't necessary as the level out was within 0.2 dB of the signal in). You could also "tailor" the frequency response by tweaking the capacitor to resistor ratios around the op-amps.
As you can see, it's literally a "back of an envelope" design scribble! I'll draw up a better version electronically, and a suggested PCB / stripboard layout later this week.
How does it sound? Like it's not there (which is probably its best recommendation).
Possible modifications - the op-amps are set for unity gain, but in reality there will be a (minimal) loss. With the addition of a couple of resistors and capacitors per op-amp, you could easily introduce some gain to get exactly the same level out as in (I found it wasn't necessary as the level out was within 0.2 dB of the signal in). You could also "tailor" the frequency response by tweaking the capacitor to resistor ratios around the op-amps.
"Why is it humming?" "Because it doesn't know the words!"
- mictester
- Old Solderhand
Information
As promised:
A tidied-up version of the electronic switching.
A tidied-up version of the electronic switching.
"Why is it humming?" "Because it doesn't know the words!"
- mictester
- Old Solderhand
Information
Hiaskwho69 wrote:hi i've been looking for good electronic switching and found this can iask if you have some PCB layout on this? thanks a lot fors haring
I've got both PCB and Vero layouts for the electronic switching circuit, but haven't got them here in electronic form. I'll scan them in the next day or so.
"Why is it humming?" "Because it doesn't know the words!"
- earthtonesaudio
- Transistor Tuner
The series switch should come before the shunt switch, in both cases. Typo perhaps?
rocklander wrote:hairsplitting and semantics aren't exactly the same thing though.. we may need two contests for that.
Mictester,
Thanks for sharing your circuit. I recently built a similar switching device but with three FX loops and a bypass. I only have a buffer on the input but I noticed you use a buffer on the output too. I'm using mechanical SPST footswitches to do the on/off switching for each device. I tried using a bipolar supply for Vdd and Vee and found the device could handle up to a +4.5V / -4.5V voltage swing but I was getting bad switching pops even though I used +4.5V to bias the device. Ended up using a single ended +9V supply and seems to perform without any problems even though it doesn't look as good on a scope.
So is there a great deal of benefit having two buffers?
I'm not completely understanding your schematic symbols for the 4066 so I'm not sure what's going on with your biasing.
TT
Thanks for sharing your circuit. I recently built a similar switching device but with three FX loops and a bypass. I only have a buffer on the input but I noticed you use a buffer on the output too. I'm using mechanical SPST footswitches to do the on/off switching for each device. I tried using a bipolar supply for Vdd and Vee and found the device could handle up to a +4.5V / -4.5V voltage swing but I was getting bad switching pops even though I used +4.5V to bias the device. Ended up using a single ended +9V supply and seems to perform without any problems even though it doesn't look as good on a scope.
So is there a great deal of benefit having two buffers?
I'm not completely understanding your schematic symbols for the 4066 so I'm not sure what's going on with your biasing.
TT
- mictester
- Old Solderhand
Information
Yes. I buffer on the way in, so that there's a high input impedance - you can feed almost anything into it. The output buffer just separates the switching circuit from the outside world, and gives a really low output impedance so that you can drive long cables if necessary.tictac wrote:Mictester,
So is there a great deal of benefit having two buffers?
Each gate in the 4066 has an in, out and control pin. I bias the audio path at ½ rail. Then there's little chance of introducing switching clicks and pops.tictac wrote:I'm not completely understanding your schematic symbols for the 4066 so I'm not sure what's going on with your biasing.
TT
"Why is it humming?" "Because it doesn't know the words!"
Hmmm, may have to rethink my circuit to add that 2nd buffer, not alot of room on my board though...Yes. I buffer on the way in, so that there's a high input impedance - you can feed almost anything into it. The output buffer just separates the switching circuit from the outside world, and gives a really low output impedance so that you can drive long cables if necessary.
Yes I bias my audio path at 4.5v too, and I use a 9V to ground for the Vcc.Each gate in the 4066 has an in, out and control pin. I bias the audio path at ½ rail. Then there's little chance of introducing switching clicks and pops.
I'd sure like to figure out how to use a bipolar supply without getting all that popping noise...
- Hides-His-Eyes
- Tube Twister
Because a tiny difference becomes a real pop. better just to decouple and not risk it.cpm wrote:why AC couple the switches from the buffers (those two 47n caps) ?
both sides are biased at 1/2 V...
Testing, testing, won too fwee
- Ripdivot
- Resistor Ronker
Years ago I built the Craig Anderton Electronic Footswitch and it worked really well. It used a 4016 instead of a 4066 but I think a 4066 would work with the Anderton circuit as well. The only down side was that it used a bi-polar supply.
- Attachments
-
- andertonelectronicfootsro9.gif (10.08 KiB) Viewed 3429 times
- Barcode
- Diode Debunker
It looks rectified to straight DC to me, try it with single supply. Just remove those bottom 3 series diodes and tie negative to ground and see what happens!askwho69 wrote:is there another way to make it as single supply?
- earthtonesaudio
- Transistor Tuner
Almost.Barcode wrote:It looks rectified to straight DC to me, try it with single supply. Just remove those bottom 3 series diodes and tie negative to ground and see what happens!askwho69 wrote:is there another way to make it as single supply?
To make the Anderton one single supply, do what Barcode said, but in addition, tie all points marked "ground" to a virtual ground instead, and AC couple in and out of the effect.
rocklander wrote:hairsplitting and semantics aren't exactly the same thing though.. we may need two contests for that.
- mictester
- Old Solderhand
Information
That'll work, but Anderton was a bit optimistic. I find that there's frequently some break through - particularly with high gain pedals - and so it's good to shunt unused paths to ground. If I do this, I get no breakthrough whatsoever.earthtonesaudio wrote:Almost.Barcode wrote:It looks rectified to straight DC to me, try it with single supply. Just remove those bottom 3 series diodes and tie negative to ground and see what happens!askwho69 wrote:is there another way to make it as single supply?
To make the Anderton one single supply, do what Barcode said, but in addition, tie all points marked "ground" to a virtual ground instead, and AC couple in and out of the effect.
"Why is it humming?" "Because it doesn't know the words!"
- earthtonesaudio
- Transistor Tuner
There really ought to be some (small) series resistance between your input buffer and the switch that shunts the signal to AC ground. Otherwise you're relying on the impedance of the 1uF cap (plus whatever is built into the op-amp's output section) to do your voltage division.mictester wrote:...it's good to shunt unused paths to ground. If I do this, I get no breakthrough whatsoever.
A couple hundred ohms would make this into a much better attenuator while being much kinder to the op-amp.
Same goes for the shunt switch after the "FX out" volume control, but that one is less of a problem because it only comes into play when the effect is at full volume.
rocklander wrote:hairsplitting and semantics aren't exactly the same thing though.. we may need two contests for that.
Hi Mictester, I was just wondering if you still had these transfers. I would love to give them a try.mictester wrote:Hiaskwho69 wrote:hi i've been looking for good electronic switching and found this can iask if you have some PCB layout on this? thanks a lot fors haring
I've got both PCB and Vero layouts for the electronic switching circuit, but haven't got them here in electronic form. I'll scan them in the next day or so.
- mictester
- Old Solderhand
Information
I'll be back at my workshop later in the month - I'll put the layouts up then.jimmybjj wrote:Hi Mictester, I was just wondering if you still had these transfers. I would love to give them a try.
(Current location Boston USA!)
"Why is it humming?" "Because it doesn't know the words!"
Is it next month yet?mictester wrote:I'll be back at my workshop later in the month - I'll put the layouts up then.jimmybjj wrote:Hi Mictester, I was just wondering if you still had these transfers. I would love to give them a try.
(Current location Boston USA!)
- mictester
- Old Solderhand
Information
I'm back! I'll get the relevant bits off the old hard drive and post them this week.
"Why is it humming?" "Because it doesn't know the words!"
- juanro
- Cap Cooler
Hey,
If you don't mind, I'm making a PCB for this.
Here's the schematic done in ExpressSCH I'm working on: A couple of comments:
- I added isolating caps in FX In & FX Out
- I added a series R (low value, perhaps 1K?) before each shunting to VRef switch of the 4066.
- Or may be better: what if it is first series switch, then shunting? So when that part is "off" the series switch is off, the shunting on and no loading is applied to previous stage (buffer or FX)
Let me know what do you think.
Also, I will try to make the PCB as compact as possible, using small footprint caps and vertically mounted R, is that Ok or would you prefer a more spread layout?
Regards,
Juanro
If you don't mind, I'm making a PCB for this.
Here's the schematic done in ExpressSCH I'm working on: A couple of comments:
- I added isolating caps in FX In & FX Out
- I added a series R (low value, perhaps 1K?) before each shunting to VRef switch of the 4066.
- Or may be better: what if it is first series switch, then shunting? So when that part is "off" the series switch is off, the shunting on and no loading is applied to previous stage (buffer or FX)
Let me know what do you think.
Also, I will try to make the PCB as compact as possible, using small footprint caps and vertically mounted R, is that Ok or would you prefer a more spread layout?
Regards,
Juanro
La única verdad es la realidad.