Neovibe Oscillating Hiss

[Stratomaster]

Hello,

I've got my Neovibe up and running, and it sounds alright except for a wooshing of noise at the speed of the oscillation that happens even when the effect is switched off.  My guess is that something weird is happening with the way I am powering it.  I am running a 1Spot into it, and I incorporated the circuit from a Godlyke Powerpump set to 18V to step up the voltage to the desired voltage.  So at the DC jack there is 9V that is tapped off to an opamp buffer and the status LED and to the power pump.  It is hard to describe in words, so i have liked to a mini-flow chart I made.  When I connect power only to the Neovibe, I don't get this effect, but when other pedals are attached to the 1spot, it happens all the time.  Any suggestions?

http://web.mit.edu/omj/www/NEOVIBE/neovibepsissue.JPG

[R.G.]

I would suspect the connection of the Godlyke Power Pump to the 1Spot. Both are switching power units, and it's common for the interconnection of switching units to have some sum/difference interaction, if not outright radiation. Internal devices can amplify this as well. I think it's worth digging there.

[Stratomaster]

Thanks for the response R.G.  I'm thinking something similar, but I can't get over the fact that it doesn't happen when only the Neovibe is plugged in.  Only when the power is shared w/ other pedals does it act up.  And yes, I did wire it as suggested in one of your previous posts about sharing power supplies w/ the neovibe.  Thanks again.

[frequencycentral]

Try a 22uf (10uf should do) cap and a 0.1 uf cap across the power rails where the power enters each and every seperate circuit. This is called 'decoupling'.

[Stratomaster]

Thanks also, but again... the problem is only happening when the Neovibe is sharing power from the 1spot with other pedals.  It is not acting up when it is the only thing plugged in.  This leads me to believe that there is no issue with coupling between the circuits in the Neovibe.  I could be wrong though.

[axg20202]

A long shot, but could it be that you are approaching the limit of the power supply max current output when you chain it with a bunch of pedals in addition to the Neovibe? Might be worth ruling out - the Neovibe is a fairly hungry pedal because of the need to light that filament bulb. Is the powerpump you speak of capable of supplying enough current at 18V? I would try powering the Neovibe from a dedicated 18V power supply, rather than a stepped up 9v supply - pound to a penny it solves the problem.

[Stratomaster]

The 1spot claims to be capable of outputting 1.7A of current at 9V.  The Godlyke powerpump can output 440 mA at 18 V.  Even at half these values, there is no danger of approaching the limit load of these supplies.  Getting a dedicated adapter would defeat the purpose of the 1spot, make my new pedal board plans not work out so well, and royally piss me off.  If the mini-dejavibe can be made to work in this setup, so can my Neovibe build. 

[axg20202]

Fair enough. All I'm suggesting is that you try a second power source to help track down the problem.

[R.G.]

The 1spot claims to be capable of outputting 1.7A of current at 9V.  The Godlyke powerpump can output 440 mA at 18 V.  Even at half these values, there is no danger of approaching the limit load of these supplies.  Getting a dedicated adapter would defeat the purpose of the 1spot, make my new pedal board plans not work out so well, and royally piss me off.  If the mini-dejavibe can be made to work in this setup, so can my Neovibe build. 

Mother Nature has always shown me a callous disregard for my feelings. She doesn't seem to care whether I'm pissed off or not. Your mileage may vary.

Here's a question - does your Neovibe circuit still have the rectifier and filters in it?

[Stratomaster]

Mother Nature has always shown me a callous disregard for my feelings. She doesn't seem to care whether I'm pissed off or not. Your mileage may vary.

Here's a question - does your Neovibe circuit still have the rectifier and filters in it?

I hear that!  Yes.  Aside from putting in a switchable buffer before the audio path and an LED in series with the lamp, the Neovibe circuit is wired as stated in your write-up. 

[R.G.]

The Neovibe is somewhat unique in that the power input is designed to take in AC voltages, rectify them to DC and then use that DC. In spite of warnings and many how-to-fix-my-Neovibe threads here and across the net, people still feed in DC (which works OK) and are then confused when they have grounding problems by grounding the DC to other pedals (which does not work).

Unless the Godlyke Whachamacallit isolates your 18V for the Neovibe, you are trying to make the 1Spot negative be two different voltages - one to all the pedals and another inside the Neovibe.

Does the Godlyke put out DC?
Is the Godlyke widgie isolating?

[Stratomaster]

R.G.,
The Godlyke does put out DC (range of 12V to 24V).  What do you mean by isolating?  I figured something like that was happening, though it is hard to believe since I am splitting power at the DC jack as illustrated in my diagram.  The ground side of the Godlyke pump is only used as a ground for the Neovibe circuit.  The hot side of the Godlyke output is going to pad J, and the ground side is going pad O.  The input & output jacks, buffer, and on/off LED are grounded to the 9V side.  I'm really stumped because it all should work, but doesn't.  It's fine w/o any other pedals attached, so grounding is correct at least within the pedal. 

If I absotively, posolutely need to get a dedicated power supply to get it to work, I will.  But that will be my last resort.  Thanks again for all your help.

If it's a grounding issue, why would it happen at the frequency of oscillation?  I can change the frequency of the noise by rocking the pedal.

[swt]

i've built three and have the same issue with all of them. Always thought it's a ground loop, but also, i don't have a metal isolator for the bulb and cells, and maybe, the transistors get noise from the bulb, so this is the next thing to try. Will report....

[axg20202]

i've built three and have the same issue with all of them. Always thought it's a ground loop, but also, i don't have a metal isolator for the bulb and cells, and maybe, the transistors get noise from the bulb, so this is the next thing to try. Will report....

Can't hurt to try, but I doubt it's the cure. Many have made Neovibe's with plastic film cannisters and don't get this noise.

[R.G.]

The Godlyke does put out DC (range of 12V to 24V).  What do you mean by isolating? 

By isolating I mean that none of the electrical pins show DC continuity between input and output; in particular, that the ground pins on input are not connected to the ground pins on output. I doubt it is, but it could happen.

I figured something like that was happening, though it is hard to believe since I am splitting power at the DC jack as illustrated in my diagram.  The ground side of the Godlyke pump is only used as a ground for the Neovibe circuit.  The hot side of the Godlyke output is going to pad J, and the ground side is going pad O.  The input & output jacks, buffer, and on/off LED are grounded to the 9V side.  I'm really stumped because it all should work, but doesn't.  It's fine w/o any other pedals attached, so grounding is correct at least within the pedal. 

OK.

If it's a grounding issue, why would it happen at the frequency of oscillation?  I can change the frequency of the noise by rocking the pedal.

At least part of what's going on is related to the ultrasonic or RF oscillations inside the charge pump and/or 1Spot. Without some fairly detailed work on the setup with at least a good oscilloscope, it's not possible to do other than guess about causes.

The 1Spot has very little ripple on its output, and what it does have is well above the audio range. But you're adding to that a power switching converter - the Godlyke - that pulls big pulses of current at its own switching frequency. Both of those frequencies may and probably do vary with loading. Any time two oscillators get funneled together into a non-linear process (pedals count here), you get the generation of sum-and-difference frequencies in the well-known aliasing process. That is definitely going on; the question is how does it get into the audio path?

That it varies with LFO is not too amazing - that light bulb causes a big load, and the result of that is varying current drain on the Godlyke. A characteristic of up-converters is that they eat input current in rough ratio to their output stepup. So if the Godlyke is putting out 50ma at 18V, it will need at least 100ma of 9V, probably more, to work. So there's a big loading effect on the 1Spot as that varies.

It's possible that only certain loading conditions cause it, as the other pedals may push the 1Spot into a frequency or current range where the Godlyke is sensitive.
It's possible that the Neovibe isn't doing this at all, but that some other pedal is sensitive to what it gets on its power supply at a particular frequency, and when it is
hooked up, the 1Spot/Godlyke combination feeds it a ripple that it's sensitive to.
It's also possible that the wiring/cabling loops are picking up a nearby radio transmitter and that the combination of the radio station plus beating between 1Spot and Godlyke cause that to be audible.

RF processes get to being black magic. For instance, that 100uF cap you used to bypass power - it likely does not look like a capacitor at frequencies over 1MHz or so; after that, it looks like an inductor, due to the winding of its film innards. All components eventually look  inductive as frequency increases. They may be self-resonant at some frequency (or multiples!) as well. A wire can look like a resistor, a capacitor, an inductor, or and antenna, depending on the frequency and mechanical dimensions.

Try powering the Neovibe with the 1Spot and Godlyke, but don't hook audio signal through it - only through your other pedals. See if that still causes a sound.

[Stratomaster]

With the Neovibe powered but not hooked up to the audio path, the noise is still audible, but not nearly as noticeable as with the Neovibe in the audio path.  You are kicking ass, R.G.  The Godlyke is non-isolating, as there is continuity between input and output ground connections.  Also, the phenomenon happens at the same volume regardless of whether the pedal is bypassed or active.  Is there any way to fix it?  I will get a 2nd power supply if absolutely necessary.  Thanks.

[fpaul]

What bulb are you using?  Mine was doing something similar when I was using a 1.5volt bulb.  The bulb blew and I replaced it with a 12volt.  The noise problem went away.  I think the 1.5volt was just too bright; it was like hearing it turn on and off.  It sounds great now and I'm actually playing my squier strat quite a lot now. 

[R.G.]

That was useful. It tells us that the whine is partly from the audio circuits in the Neovibe picking up whine, but also partly because the Godlyke is either (a) radiating the noise as bursts of RF that the other pedals pick up or (b) wiggling the ground lead around by sucking bursts of current big enough to be picked up by the other pedals. With the Neovibe out of the audio path, that's about the only two things that make sense.

It would be very interesting to remove one pedal at a time from the audio path, like the Neovibe. That would tell you if one pedal was particularly sensitive.

But I'm guessing it's (a). That's a tough one.

It *may* be possible to put a ferrite core inductor in each of the + and - leads to the Godlyke and a few capacitors after the chokes to prevent the noise from spreading back into the rest of the chain. Likewise, it *may* be possible to put the similar chokes and filter caps after the Godlyke to keep the noise out or the Neovibe, or the front end filter may kill off the interactions between the 1Spot and the pump enough. The caps will need to be something like 1000uF paralleled with 0.1uF ceramic paralleled with 0.01uF ceramic.

I've run into something similar with a few pedals whining with the 1Spot. Near as I can tell, it's always a pedal with an internal switching power supply and higher current drain, very much like this. In the electronics industry at large, one would never voluntarily design two power switching power supplies in a chain with non-synchronized switching frequencies, for this very reason. The two different frequencies generate sum-and-difference frequencies, the difference frequency being down in the audio range.

At this point, you need to decide whether you want to satisfy your curiosity and try the filters or put a linear wall wart on the Neovibe, which should clean this up too.

fpaul's comment is also interesting. That would seem to verify that it's the bigger current drains causing heavier switching currents that's involved. By all means, try the higher voltage bulb if that can work for you.