ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

John Hollis Ultra Flanger

11/27/2005 4:04 AM
Ed	John Hollis Ultra Flanger I just finished building the Ultra Flanger pedal, using the schematic and PCB at geofex.com (http://www.geofex.com/PCB_layouts/Layouts/ultrafln.pdf). The pedal works brilliantly apart from a clicking noise in time with the LFO. I decided to have a look and found that there had been a similar problem with the Zombie Chorus pedal which John had designed, there is a post in the hammer.ampage.org list (http://ampage.org/cgi-bin/hammer/index.cgi?cmd=lt&xid=&fid=&ex=&pg=3) about fixing this but no mention of the problem with the Ultra Flanger. I have tried debugging it using the same method as with the zombie chorus with two pots, but the clicking can't be improved. Does anybody have any suggestions or idea? Many Thanks.

11/27/2005 4:04 AM

John Hollis Ultra Flanger

I just finished building the Ultra Flanger pedal, using the schematic and PCB at geofex.com (http://www.geofex.com/PCB_layouts/Layouts/ultrafln.pdf). The pedal works brilliantly apart from a clicking noise in time with the LFO. I decided to have a look and found that there had been a similar problem with the Zombie Chorus pedal which John had designed, there is a post in the hammer.ampage.org list (http://ampage.org/cgi-bin/hammer/index.cgi?cmd=lt&xid=&fid=&ex=&pg=3) about fixing this but no mention of the problem with the Ultra Flanger.

I have tried debugging it using the same method as with the zombie chorus with two pots, but the clicking can't be improved.

Does anybody have any suggestions or idea?

Many Thanks.

11/27/2005 10:11 AM
Mark Hammer	The issue of ticking LFOs is one that simply will not go away. So many designs attempt to be compact, simple, and inexpensive by crossing their fingers and omitting decoupling. There is extensive discussion of deticking all manner of modulated effects over at the diystompbox forum that you can find by looking through the archive, but the quick and dirty version is this: Appliances in your home share the same 120VAC source. If amotor gets turned on somewhere else in the house, the alternating draw on the current will be reflected in audio appliances elsewhere in the house, and you'll hear interference...UNLESS the various devices regulate and smooth the power they receive so that minute fluctuations are effectively sanded down by the power supply. Fortunately, most appliances these days have regulated power supplies, but let me tell you that in the old days, if mom decided to sew while the big game was on TV, all hell would break loose! In a stompbox, every semiconductor in it is a bit like an appliance in the house, except that instead of sharing 120VAC, they share a battery. An LFO works a bit like an AC motor in that it makes repetitive momentary huge draws on the current from the battery. You may THINK the LFO producesd a triangle wave, but it starts with a square wave that gets lag added to it to make a triangle, and those sudden current draw surges required by a square wave are what makes the ticking. To fix the ticking, you need to "regulate" the supplies of the individual chips. That is done in several ways. One is to insert a very low value resistor in series between 9v+ and the V+ pin of the chip, and a medium value cap to ground from there. That cap hangs onto a bit of current to tide the chip over for the milliseconds required. Think of it as a teeny tiny UPS. The other tactic to apply (in addition, not instead), is to decouple the Vb (Vref) as well by running a a separate voltage divider and cap to ground for each chip so that sudden draws on Vref from chip A have no impact on chip B, because they each have a few milliseconds of reserve power to tide them over. That's not all, but is a big part of deticking pedals. Hope it makes sense. Gotta run right now, but I'll check back in later.

11/27/2005 10:11 AM

Mark Hammer

The issue of ticking LFOs is one that simply will not go away. So many designs attempt to be compact, simple, and inexpensive by crossing their fingers and omitting decoupling. There is extensive discussion of deticking all manner of modulated effects over at the diystompbox forum that you can find by looking through the archive, but the quick and dirty version is this:

Appliances in your home share the same 120VAC source. If amotor gets turned on somewhere else in the house, the alternating draw on the current will be reflected in audio appliances elsewhere in the house, and you'll hear interference...UNLESS the various devices regulate and smooth the power they receive so that minute fluctuations are effectively sanded down by the power supply. Fortunately, most appliances these days have regulated power supplies, but let me tell you that in the old days, if mom decided to sew while the big game was on TV, all hell would break loose!

In a stompbox, every semiconductor in it is a bit like an appliance in the house, except that instead of sharing 120VAC, they share a battery. An LFO works a bit like an AC motor in that it makes repetitive momentary huge draws on the current from the battery. You may THINK the LFO producesd a triangle wave, but it starts with a square wave that gets lag added to it to make a triangle, and those sudden current draw surges required by a square wave are what makes the ticking.

To fix the ticking, you need to "regulate" the supplies of the individual chips. That is done in several ways. One is to insert a very low value resistor in series between 9v+ and the V+ pin of the chip, and a medium value cap to ground from there. That cap hangs onto a bit of current to tide the chip over for the milliseconds required. Think of it as a teeny tiny UPS. The other tactic to apply (in addition, not instead), is to decouple the Vb (Vref) as well by running a a separate voltage divider and cap to ground for each chip so that sudden draws on Vref from chip A have no impact on chip B, because they each have a few milliseconds of reserve power to tide them over.

That's not all, but is a big part of deticking pedals. Hope it makes sense. Gotta run right now, but I'll check back in later.

11/27/2005 11:40 AM
Ed	Thanks for the tips and explanation! I'll have a look on the diystompbox forum and see what's going on there. Once again many thanks! If I get stuck I'll post back here again

11/27/2005 2:20 PM
Mark Hammer	You're welcome. My advice is to post there if you experience difficulties. This place is tops for amps and pickups (and political arguments ), but if it is pedal-related, the stompboxforum is THE best resource on the web, bar none.

11/27/2005 2:57 PM
Ed	I'll definately lurk in the stompboxforum for a bit methinks. I'll try the resistor and cap mod you suggested, what sort of values do you think would be required to do the job? Regards

11/27/2005 5:49 PM
anon	Hmmmmmmmmmmmmmmmmmmmmmmmmmm. Well, maybe not. Appliances often have motors which have brushes which in turn cause sparks which in turn cause RF generation which in turn can be picked up by audio devices as noise. Don't need to be on the same AC line for RF induced noise to be picked up by an audio device.

11/28/2005 8:56 AM
Mark Hammer	Well yeah. It is true that lines don't HAVE to be shared to create noise. The basic point being made, however, was not to cover all forms of noise, but to point out that because they sit on the same PCB, we often forget that chips behave like multiple devices sharing the same power lines and sudden current demands by one device can have an impact on other devices sharing the same line unless precautions are taken. When it comes to LFOs, many builders/DIY-ers are unaware that the standard 2 op-amp LFO produces a square wave whichemands lots of current every time the signal/wave swings in the other direction. That's why the tick gets faster when you adjust "rate". The general consensus about John's designs is that they are very clever and elegant, but occasionally "lean" to the point of inviting problems. His Zombie chorus is a nice cheap chorus project, but by the time you add in all the things needed to improve this and that, you end up with the equivalent of a more complex commercial equivalent. That's not a slag, just a reminder that you get what you pay for and you get what you build.

11/28/2005 8:56 AM

Mark Hammer

Well yeah. It is true that lines don't HAVE to be shared to create noise. The basic point being made, however, was not to cover all forms of noise, but to point out that because they sit on the same PCB, we often forget that chips *behave* like multiple devices sharing the same power lines and sudden current demands by one device can have an impact on other devices sharing the same line unless precautions are taken. When it comes to LFOs, many builders/DIY-ers are unaware that the standard 2 op-amp LFO produces a square wave whichemands lots of current every time the signal/wave swings in the other direction. That's why the tick gets faster when you adjust "rate".

The general consensus about John's designs is that they are very clever and elegant, but occasionally "lean" to the point of inviting problems. His Zombie chorus is a nice cheap chorus project, but by the time you add in all the things needed to improve this and that, you end up with the equivalent of a more complex commercial equivalent. That's not a slag, just a reminder that you get what you pay for and you get what you build.

Page 1 of 1