Messages

Old thread, but I figured I'd try to offer some assistance. Rather than trying to shoehorn a compatible foot-pedal tunable pot into the phase LFO design, we can modify the LFO to work with a standard value potentiometer.

The LFO can be replaced with a Voltage Controlled (Oscillator), which works just as it's name implies. You apply a voltage to the oscillator and it...oscillates. The voltage input varies the speed. So you can use a standard voltage divider!

Now, where can we find a VCO Schematic? Synth forums? Sure. But then we've gotta modify it so that the output of the VCO matches the voltage swing of the phaser's inherent LFO. Then we've gotta remove those LFO components from the board and shoehorn this bastard in. Do you have any idea how many sleepless nights it took me to research the subject? An embarrassing amount...

Oh wait. RG Keen already did that? Without destroying the pedal's LFO? And we can add a few components to a perf/vero daughterboard to make this all work together? Awesome!

http://www.geofex.com/FX_images/p90ramp.pdf

Please note that this is for a leslie style ramp. You footswitch between 2 voltages, for 2 speeds. The 100K Pot and 100uF cap are there to slow down the transition between voltages. So don't include the switch, the second, pot, or the 100k pot and 100uF cap combo. Just have your "wah pot" voltage divider go straight to the base of that transistor.

NOTE! I believe the nom nom's LFO IC is shared with the input buffer. For this design, you cooooould replace IC with the IC that they call out in the document above, and just provide another IC to take care of the other IC required for the VCO.

Bonus points if you can figure out how to add a switch so that the wah rocker controls the sweep like a standard wah pedal, instead of just controlling the speed

Ok, so I finished the schematic.  If someone could check it and make sure I didnt make any mistakes, it would be much appreciated.  Also, any concerns? I used the Pork Barrel part numbers and schematic since most people on this forum, Im assuming, would be most familiar with it.

Thanks in advance!
Matt

You're over thinking this.

In this scheme that you've devised, you don't need to switch the power to the LFO's, nor should you. As a rule of thumb, switch the ground, not the hot. Look at how LED's are switched, you toggle the ground.

That said, switching power to the LFO will likely cause a nasty pop. Try this. Set the pedal up and strum a chord through it. While the chord rings out, unplug the power to the pedal, then plug it back in. What do you hear? POP. You're likely to incur the same effect if you switch the power to your LFO.

Additionally, LFO's require a "moment" to start oscillating once connected to power. So switching the power to them will not give you a seamless switch, which I'm assuming is one of your design criteria.

In your scheme, you could simply switch the output from the LFO. I don't see a reason why having 2 LFO's oscillating, with only one being utilized would be an issue. Noise? Maybe, but that can be mitigated. Would the noise from another LFO be more offensive than the noise of a pop? Probably not.

But wait. Why do we have 2 LFO's anyway? To switch speed, range, depth, AND LED's on the same 3PDT switch? Hmmmm, I think we can devise something clever. To have 2 dedicated LFO's for this task seems...inelegant.

Ok, so what determines the speed? A voltage divider. We have a pot that is simultaneously increasing the shunt resistance while decreasing the series resistance as we turn the Rate Knob up. Now, what if we were to switch just the shunt resistance, but keep the series resistance fixed? It's still a voltage divider, right? Maybe we use a log or reverse log taper pot instead of linear to compensate for the change in the resistance taper.

Ok, so that's 1 "row" from our 3PDT. Now depth? Another voltage divider? See above.

2 Rows gone. 3rd Probably has to go to our setting indicator LED. But...

But this cap...We're out of switch lugs, so let's think for a minute. A user above suggested that the CE-2 doesn't get quite fast enough to get into Leslie territory. Fair. Decrease the cap size, and we get a speed that's more appropriate for that Leslie swirl. This affects the whole range of speeds available on our Rate Knob, so bumping the range up reduces our lowest available speeds. But...will they necessarily be missed? Sure, you get a nice doubling effect when the rate is super low, but are you honestly going to be switching between a super slow chorus up to a much more rapid Leslie-esque rate? If no, it might simplify things greatly if you can possibly find a cap value that shifts the range up to where you want it, but doesn't sacrifice low speeds that are available to you when footswitched. Because remember, if you want those sounds to be available, you can always throw that cap on a toggle switch and save on precious 3PDT Real estate. If you're REALLY insistent on having the whole range available to you, then get a 4PDT.

This solution may not be for you, but it is certainly something to experiment with.

Bonus! Futz with the value of the 33pF capacitor on the clock to mess with the delay time. You start to creep into flanger territory if you reduce the delay time, which I think is pretty characteristic of a real Leslie.

Good luck!

Explanation is simple: C6 was used on rev.1 to eliminate DC on the Cut pot. Although the original design didn't have an scratchy type noise on the Cut pot when you turned it I thought it was better to just remove it altogether. The added 1M resistor is to provide a small discharge path for C7 just in case any voltage sat on it when the Sat switch is flipped to ground (unlikely to happen in this case though). This section is probably a little over-engineered but I thought that the couple extra parts was worth the peace of mind of no one having any possibility of a noisy Cut control or pop when the switch is engaged.

Brilliant! Thanks for the description. That pretty much answered everything.

Now if we were to look at the Green Bean:
http://www.madbeanpedals.com/projects/GreenBean/docs/GreenBean2014.pdf

I'm wondering why R9 is needed to re-establish the bias voltage, which is already set to VB by the previous stage.

This is an open question to Bean, that I figured could benefit everyone. Alternatively, I would appreciate anyone else's input. I will be using the links below as references to my questions.

Original Sparkplug: http://www.madbeanpedals.com/projects/Sparkplug/Sparkplug.pdf
Sparkplug Rev.1: http://madbeanpedals.com/projects/Sparkplug/Sparkplug_rev1.pdf

I notice that after the output of Op amp one of the Original Sparkplug, there is no cap and the diodes are referenced to VB for hard clipping. Makes sense. We're biased at 4.5V, so biasing the diodes at that makes sense. In Revision 1, there is a 1UF cap (removing that 4.5V DC bias) and subsequently the diodes are referenced to ground. Also makes sense. Why the change? The only seeming disadvantages of the additional capacitor are the addition of a new part and the inherent filter of the cap in the signal path (even if 1UF is more than enough to pass all the signals). Can you please comment on the reason for the change?

Secondly, the method of bypassing the hard clipping has changed. Previously, you would only make/break the ground/VB connection with an SPDT. Now, you're shorting the 1M resistor to provide a reference. What is the added benefit of this arrangement? It should be noted that this is somewhat similar to how you introduce the bypass capacitor to the gain stage in your Lavache Design (I believe a deviation from the original design), by shorting a large 3.3M Resistor.

Thank you for the help, and I hope this offers some insight to others.

I had a thought as I reading the schematic of the Low Rider, in preparation for my build. It seems that the 2 octave down section is taken off the output of the 1 octave down.


What if we were to instead put the "2 octave down" section in parallel with the 1 octave down? We could also put a high pass filter in front of one section, and a low pass in front of the other.  They can be tuned to have the same corner frequency, a bit of overlap, or even a slight gap. Experimentation is in order.

Of course the mixing resistors (R46 and R55) can be adjusted. This mod can easily be implemented by lifting  one end of C14, and inserting the filter right after the capacitor. C18 however needs to be connection the input end of C14, then having the filter succeed it just as we did in the other section.

Essentially, this mod puts the octave down sections in parallel, affection different parts of the fret board. Depending in the filtering, this could make power chords or at least 5ths track better. Obviously the filter isn't going to track where we are perfectly. But for extra hacking points, you can have the filter(s) be controllable via external pots or even rig up a simple passive crossover.

Any thoughts?

I'm working on a pcb for pairing up two 3008s to act like one 3005.  I'll let you guys know how it goes.  If it works out I can make a batch and share.

Is it going to be different in design than what is included in this project file of the aquaboy?

http://www.madbeanpedals.com/forum/index.php?action=dlattach;topic=5521.0;attach=4149

The datasheet on MN3101 states that it is capable of driving 2x MN3005's. I'm afraid that I don't think you can use double delay boards on an Aquaboy DX to get 4x MN3005's.

However, the datasheet (linked below) states the following: "BBD Direct driving capability of up to two MN3005s (equivalent to 8192 stages)."

http://pdf1.alldatasheet.com/datasheet-pdf/view/14238/PANASONIC/MN3101.html


What one might be able to experiment with is driving 4x MN3008's to get the same delay time as 2xMN3005's. That opens up the possibility of having the improved fidelity and headroom of MN300X BBD's versus MN320X, without having to gamble on buying MN3005's on ebay.



Also, I wonder what the tonal differences would be between 1x MN3005 vs 2x MN3008? I definitely plan on testing that out with MN3205's and MN3208's, (can't find any MN3005's).