SPST Switch Popping

[dpresley58]

I've built the improved Fetzer valve from Runoffgroove in a small Ratshack project box and use a SPST switch to activate it. Are these switches inherently noisy? I can't seem to get it -not- to pop when I turn it on. I'm using the switch to break ground, if that matters. Article and schem for this project here : http://www.runoffgroove.com/fetzervalve.html


Thanks.

[petemoore]

  There are a number of grounds which could be broken on the Fetzer, most of which would be cause for pop in audio path I would think.
  I find it's difficult to determine where the switch was actually placed.

[dpresley58]

Apologies... should have been more specific. I wired the battery leads in to the board and then cut the ground lead to hook to the switch. Would it have made a difference to have used the positive lead instead?

[petemoore]

  Nope, no real difference in popping.
  That kind of switch assignment will power up your actives, if they're in the signal path [ie not in bypass mode or amp off] you'll hear a pop as the actives fire up.

[dpresley58]

Is there some method (additional circuit precautions / components) using the SPST that would prevent it from popping, or is it simply a fact of life?

[Mark Hammer]

When not in use (i.e., powered) all your capacitors in the circuit drain off remaining charge.  When you power up, the pop you hear is the sound of all those caps going "Gimme!  gimme!".  This is no different than the dull thud you hear from your stereo speakers when powering up your stereo or home theatre.

[dpresley58]

..which would fall into the "just the facts of life" category, I'm assuming. I understand the purpose of a pulldown resistor to drain electrolytics, but from what you're saying this is more or less the opposite effect. The caps are going through the cycles to charge themselves, which is causing the thump.

I'm starting to get the feeling that while there are probably ways to get them to do this gracefully (i.e. more quietly), these methods wouldn't lend themselves to circuit simplicity..? For my own benefit and understanding, what would some of those methods be?

[Mark Hammer]

I certainly can't give you specifics, but one often finds reference to "soft turn-on" or "thumpless turn-on" in audio equipment.  So, it IS feasible, but may well be more complex to implement than the circuit itself.

[dpresley58]

Yep... Started finding references to that in the interim. Thanks for the help.

[dpresley58]

The references I'm finding all relate to easing the jolt to the speakers, and don't really address mitigating the front-end power surge. I think what I'm looking for is a way to ramp up the supply voltage to the circuit over a 1-2 second period. Admittedly, I'm not even sure if such an approach has any merit. Any thoughts?

[dpresley58]

This looks interesting... http://www.allaboutcircuits.com/vol_3/chpt_5/2.html The only problem I'm having with the third diagram is where the article introduces supplying power to the JFET and the circuit. It looks to me as if the positive and negative are joined at the same point. How does that work? Its confusing me.

I'm starting to see that whatever JFET is used will have to be able to supply the maximum current required by the circuit to be switched with it. Is that correct?

[anchovie]

The one connected to the gate of the JFET is supplying voltage rather than power and is simply negative with respect to ground. They won't interfere with each other.

I'm guessing that as you've only got a SPST power switch, there's no actual bypass path for the guitar signal in your box - it's either going through the Fetzer or muted? If that's the case, I'd suggest the simplest solution is to turn your amp on last!

[dpresley58]

This is really an application for my own understanding of FETs as switches. Obsessions can be troublesome. 

Would a MOSFET be more suitable to this?

[darron]

It seems that there is a little bit of confusion here, or maybe it is just me.

As Mark Hammer said, powering up the 'actives' is a recipe for a pop. General all of the components of a pedal are always powered (except LEDs) even when in standby, and the signal is bypassed around the ciruit, but in this case you are turning the circuit off.

My question is the same as anchovie's. Why do you want to turn the power off to the unit but not bypass your signal?

If you want to have silent switching you'd do well with a DPDT (or SPDT, but that would suck tone).

If you are interested in electronic switching then you can check out R.G.'s Millennium Bypass circuit. I've never built it but it's supposed to be reliable and solve lots of headaches. --> http://www.geofex.com/Article_Folders/Millenium/millen.htm

Or if you want to read up on true bypass circuits then you can check this out:
http://www.dazatronyx.com/support/what-is-true-bypass/

Or maybe you are using the circuit as a headphone amp or something or you never want to turn it off except when you pack up your gear? If that's the case then you had may as well use the stereo input jack trick to turn the circuit off so you don't need a switch.

It's late and I feel like I'm over-typing... I'll wait to see what you actually are doing because I'm curious now (: hehe

[dpresley58]

You're right. There probably has been some confusion - sorry for not being more specific.

The device is a small stereo preamp built around J201's that won't be used as a guitar pedal. I housed it in a RatShack project enclosure with a 9v battery.
My thought was to find a simple transistor switch that would ramp power to the circuit to alleviate the pop, which started me to research them. I see where the JFET is preferred over BJT's because it consumes very little power, so it was off to the races to find info on it. The closest I've come is a small article with diagrams that were a bit confusing to me.

http://www.allaboutcircuits.com/vol_3/chpt_5/2.html

It looks like V+ and V- are connected at the same point, and this is the source of my confusion. I must not be holding my mouth just right to see what it's trying to say.

Since space is a consideration, I'd like to keep the part count low as well as find a method that has the smallest impact on battery consumption. The circuit's quiescent is only about 7mA.

Now I'm starting to find other information using MOSFETs as switches, and I've never worked with them. R.G.'s article is a good idea - I'd read it some time back, but before this application came up. Maybe a re-read will shed some light, but in a nutshell, that's where I am with this. I got a BJT switch working this morning, so I'd like to see what the FET switching is all about.

[darron]

So you are using it as a pedal as you'd expect, so that still leaves the question of why you want a switch to turn the power on and off to the unit? Usually people just wire the stereo input socket to turn a pedal off when you disconnect the input jack, but that's only to save power. Also, why doesn't your pedal have a bypass switch to make the effect process your signal or have your original signal go through? Are you sure that you are not getting the two types of switching mixed up? You seem pretty confident in your electronics and construction skills so sorry if I'm way off track on my thinking...

[dpresley58]

No, sir... It will not be used as a guitar effect. I'm running Ipods, small signal Walkman-type stuff through it.

I'm a relative noob at most stuff, although I've built probably a couple dozen different pedals. It's one thing to come up with a layout and burn a board - quite another to get into the theory / application of it, which is really what I'm trying to do with this. Its a learning exercise for transistor switches. The idea behind it is to find out the different varieties available, as well as what might fit the design criteria for this particular build. As stated before, I'd like to arrive at a solution that has a low part count, is quiet, efficient and doesn't consume much.

The thread should've probably had another name, with the direction its taking. I appreciate you hanging in there. 

[darron]

that won't be used as a guitar pedal

ahh man. it was late last night and i read that as will. i should stop sleep foruming...

AMZ has an article on slow starting LEDs to prevent a pop in an already powered circuit: http://www.muzique.com/lab/led.htm

But I personally don't understand how that works while it conflicts with what Mark Hammer points out. Maybe the solution is as simple as adding massive supply filtering (some filtering in the schematic would have been a good idea anyway!). IE a 1000uF capacitor from the +9V input to ground. Can anybody suggest if this will make things better or worse? All of my circuits have at least 220uF and don't really come on with a massive thump when I power them up, but are still sort of noisy to power on. It's just not a problem with stomp boxes though.

[anchovie]

It looks like V+ and V- are connected at the same point, and this is the source of my confusion. I must not be holding my mouth just right to see what it's trying to say.

It's showing two different voltage sources, like two batteries. If you call them V1 and V2, then V1+ is independent of V2-.

[dpresley58]

So, essentially, one could use the 9v routed to both the circuit and the switch, with the source connected to circuit ground and the drain connected to circuit supply.

The 9v lead to the switch is hooked to either supply the gate (switch off) or be removed from the gate (switch on). The circuit itself is constantly powered, but no current is moving while the JFET is pinched off.

Is there a price for this in terms of battery consumption? Will it be a circuit that is constantly sipping at the 9v even when off?