Author Topic: Hollis "ultimate flanger"  (Read 4871 times)

ethrbunny

Hollis "ultimate flanger"
« on: October 16, 2004, 11:18:55 AM »
I typed all the parts into mouser this morning - looks like with a good box and such it will be 70+$ for the parts. Before I dive into this has anyone built it?

I can ( :roll: ) buy a reasonable flanger for this price. Im hoping this will have better tone though. Any opinions?
--- Dharma Desired
"Life on the steep part of the learning curve"

Thomas P.

Hollis "ultimate flanger"
« Reply #1 on: October 16, 2004, 11:42:29 AM »
Well that seems to be realistic - but wee don't do it to to save money, do we?!
Anyway I didn't built it but I heared it has some 'ticking' problems like the zombie. Unfontunately I never heard of a solution. Maybe it wont be a problem as it never was with my zombie. Some go nuts with it others (including me) never heared it.

Maybe you could leave the box out and buy some cheap (but reasonable) potis to give it a try and box it if it is worth it (I have a whole bunch of pots for doing this).
god said...
∇ ⋅ D = ρ
∇ x E = - ∂B/∂t
∇ ⋅ B = 0
∇ x H = ∂D/∂t + j
...and then there was light

Aharon

Re: Hollis "ultimate flanger"
« Reply #2 on: October 16, 2004, 02:53:01 PM »
Quote from: ethrbunny
I typed all the parts into mouser this morning - looks like with a good box and such it will be 70+$ for the parts. Before I dive into this has anyone built it?

I can ( :roll: ) buy a reasonable flanger for this price. Im hoping this will have better tone though. Any opinions?




Don't even go there.Build the Tonepad FL301 instead.
http://www.tonepad.com/project.asp?id=26
Aharon
Aharon

gorohon

Hollis "ultimate flanger"
« Reply #3 on: October 16, 2004, 06:55:34 PM »
While I like the good sounds that come from my Ultra Flanger, it has plenty of swishy hiss and ticking, making it more of a project for project's sake than a usefull, quiet effect.  Build something else.  I'm scavenging my parts from mine to make something else.
"Come on in...I've got caaandy!" H.S.

ethrbunny

Hollis "ultimate flanger"
« Reply #4 on: October 16, 2004, 09:13:50 PM »
Hmm. Ok. I guess thats a big thumbs down.  :?

Glad I asked first.

Is there any chance that the ticking is from a poorly wired bypass circuit? Ive seen other postings that point to this issue.
--- Dharma Desired
"Life on the steep part of the learning curve"

Thomas P.

Hollis "ultimate flanger"
« Reply #5 on: October 17, 2004, 04:50:08 AM »
Quote from: ethrbunny
Hmm. Ok. I guess thats a big thumbs down.  :?

Glad I asked first.

Is there any chance that the ticking is from a poorly wired bypass circuit? Ive seen other postings that point to this issue.


I don't think so. I believe what's called 'ticking' is the LFO bleeding into the audio signal.
god said...
∇ ⋅ D = ρ
∇ x E = - ∂B/∂t
∇ ⋅ B = 0
∇ x H = ∂D/∂t + j
...and then there was light

David

What about Hammer's "de-tick" mod for the Zombie?
« Reply #6 on: October 17, 2004, 07:15:13 AM »
Tomboy:

Last year, Mark Hammer posted an article at his site about a mod he made to the Zombie.  He created a new bias voltage source and got rid of the tick.  Has anyone tried something similar with this beast?

Thomas P.

Re: What about Hammer's "de-tick" mod for the Zomb
« Reply #7 on: October 17, 2004, 09:13:46 AM »
Quote from: David
Tomboy:

Last year, Mark Hammer posted an article at his site about a mod he made to the Zombie.  He created a new bias voltage source and got rid of the tick.  Has anyone tried something similar with this beast?


I remember that there was a thread about it. I think the conclusion was that the bias of the LFO is designed not to be Vcc/2 so the 'usual' trick can't be used here. Otherewise I can't see the use of a bias like this.
god said...
∇ ⋅ D = ρ
∇ x E = - ∂B/∂t
∇ ⋅ B = 0
∇ x H = ∂D/∂t + j
...and then there was light

ethrbunny

Hollis "ultimate flanger"
« Reply #8 on: October 17, 2004, 07:57:50 PM »
Surely this must be a known problem. I can't imagine Hollis posting a schematic that would make a ticking sound.. Is there anyway to contact the person that runs the geofex.com site?
--- Dharma Desired
"Life on the steep part of the learning curve"

Michael Allen

Hollis "ultimate flanger"
« Reply #9 on: October 17, 2004, 08:04:08 PM »
I remember that John Hollis did his on perf board and he said he had no ticking problems. The reports of ticking are with the Geo board i believe.

You can post here and RG will most likely see it. He's always around blowing my mind....

StephenGiles

Hollis "ultimate flanger"
« Reply #10 on: October 18, 2004, 02:23:45 PM »
I had ticking on stripboard.
Stephen
"I want my meat burned, like St Joan. Bring me pickles and vicious mustards to pierce the tongue like Cardigan's Lancers.".

ethrbunny

Hollis "ultimate flanger"
« Reply #11 on: October 18, 2004, 02:37:37 PM »
Thats a shame that the layout / design has issues. Id really like to try this project. Its well documented with a nice parts list (that I've already typed into my Mouser account...  :roll: )
--- Dharma Desired
"Life on the steep part of the learning curve"

Mark Hammer

Hollis "ultimate flanger"
« Reply #12 on: October 18, 2004, 03:58:10 PM »
The ticking at the LFO rate is a product of either problematic layout or poor decoupling, or both.

Although the LFO waveform you use for sweeping is triangular, in fact  that same LFO produces a square wave also.  As the oscillator suddenly swings high then low, high, low, high, etc., there is a big surge in current consumption that creates glitches on the power lines.  To some extent such glitches can occur if the LFO is biased asymmetrically, but even with a flawlessly symmetrical power supply (i.e., *exactly* +/-4.5v), those blips still occur.

In particular, if the bias voltage that provides the Vref is simultaneously feeding a number of different devices (op-amps) sudden current draws from one of them will create audible consequences for the others.

The "decoupling" mentioned above involves smoothing out the Vref /Vb provided to each op-amp by having it come from its own voltage divider with its own smoothing cap.  In the pdf file for the Ultraflanger at GEOFEX, you can see that there are essentially 2 main Vref/Vb sources, labelled points 1 and 2.  Vb 1 goes to a few places, and Vb 2 goes to a few others.  Though this represents an improvement over the Zombie, it doesn't necessarily go far enough.  You might consider upping C7 to maybe 22uf, and might also consider sticking a 4.7uf - 10uf cap in parallel with the 3 series diodes going to ground.

The idea here is that each of these caps does two things.  One is that it stores current such that any sudden surges in one part of the circuit are easily weathered by other parts of the circuit because of the "reserve supply" stored up in the caps (think of it as an "uninterruptible power supply" guaranteed to see you over a few msec).  The other is that it results in a more sluggish response on the Vb line to any sudden current draws elsewhere in the circuit.  In this respect it acts a bit like a severe treble cut (and I mean *severe*) for any audio hash coming on the Vb line.

Try those and see.

You may have some difficulty reaching RG at the moment.  I do believe he is presently on a continent other than his home base.

chumpito

Hollis "ultimate flanger"
« Reply #13 on: October 18, 2004, 07:23:13 PM »
So... does that mean that if the oscillator section is run from a different power supply than the rest of the circuit that would solve the ticking?  What about a total redesign of the oscillator circuit?

puretube

Hollis "ultimate flanger"
« Reply #14 on: October 19, 2004, 02:38:01 AM »
"improper layout" includes long wires to the speed- (rate-) pot running near the audio-section....

Mark Hammer

Hollis "ultimate flanger"
« Reply #15 on: October 19, 2004, 09:22:35 AM »
Quote from: chumpito
So... does that mean that if the oscillator section is run from a different power supply than the rest of the circuit that would solve the ticking?  What about a total redesign of the oscillator circuit?


No need for a total redesign.  Besides, unless you have two entirely distinct battery sources for each of the sections, it's all run from the same power source anyways.  The decoupling simply makes it the case that as the different parts of the circuit share the same source, they don't step on each others' toes.

There aren't many of us here to remember such a time, but 30-40 years ago, all the appliances in your home used to have bigger current draws and less on-board regulation.  Given that they all shared the same power coming to your home, having the motor of your mom's sewing maching going on and off in one part of the house, or the compressor motor on the fridge go on in another part of the house, would show up as power line glitches and "ticking" in your record player, or radio, or TV in another part of the house.  Arguments would erupt when people trying to watch the critical plays in a sporting event were competing with folks trying to get those pants hemmed for school tomorrow.  The reason was that when one thing sucked from the same current bottle on its own straw, the other things drinking from the same bottle were momentarily starved.  Of course now that so many appliances consume less current, operate at much higher speeds, and have better on-board regulation, it means that you can listen to music in one room, while someone grinds coffee in another and someone vacuums in a third, and none of the appliances will produce fluctuations in the power line that will show up in the amplified sound.  Of course the noise from the coffee grinder and vacuum might, but that's another story.

The decoupling allows the bias voltage relied upon by all those various semiconductors to remain fairly stable for themselves despite whatever happens elsewhere in the circuit.