Create project Need help?
Show more (3)

Enhanced -48dB/Oct. Ladder-VCF

Status: Proposal
August 29, 2013
Second resonance 03 - 432 Hz - 24dB.jpg

A real new Part on Bob Moog´s classical Ladder Filter Concept !!

(04. Nov. 2015 - The project is completely updated)

This project shows a Ladder-Filter-Module with enhanced abilities on widely selective characteristics, with also a very good working Q-Damping-Compensation on the current controlled Emphasis-Parameter, guaranteed without any losses and - like this way - almost unique for an addition to the known Moog-Concept.
The circuit adds to a VCF-Module on experimental purposes up to 10-15 possible Switches &/or analog Controllers, chooseable on electronic Resonance/Emphasis-Modulation, Light to Sound Control for interesting Liveacts, LDRs, Solarcells, Phototransistors for the ladder/resonator & other special el. parts etc. and always accounting on maybe crazy ideas.
A second controlling module for this project part will follow as soon as possible, as a separate project-contribution (planned as an "Emphasis-Scanner-Option for the Resonator").

The main difference between this concept, compared to the classical Moog-Ladder, is that it can get in selective resonances beneath the main filter-frequency, what is hereby configureable with a second ladderlike construction, the so called "Resonator-Cell", which gets the currentpeaks mirrored from the top of the Ladder, respectively from the first of allways four poles in the filter, where the productwaves origin.
As a nice addition, the module features separate signals and outputs for -12 dB High-Pass and -6 dB Band-Pass, completely available in parallel to the Low-Pass Main-Characteristics.
Please note that the described "Resonator" has nothing to do with the known "Poly-Moog-Resonators", hereby maybe confusing.

If you choose to built that project-circuit for your experimental diy-usage, please do me that favour and implement this only with the recommended current-mirrors BCM62B (or better BCV62A) and like in the shown doubled combination, (those parts are available by Mouser, Conrad, Reichelt, Farnell, Digikey) because only these will lead to the suggested functions with practical usage abilities, due to the needed temperature coupling in that parts. To do this with standard-transistors (expl. BC557 or sim.) is of course possible, but this will not fix the interests and downgrade the main performance. A maybe outsourcing workshop for those kind of mirrors - with respect to analog technology - and their advanced possibilities is planned upon incoming feedback, so to speak, with enough (suspected) interest on the item and by the community, and therefore following later on...

Because all Ladder-Filters that are based on Bob Moog`s patented principle are in the end anyhow always "Diode-Ladders" (meant also on the wellknown prebiased Transistor-Ladders, please refer therefore also to the original of the US-Patent No. 3475623 and its descriptions), this project is also called like that, a "Diode-Ladder".
On these facts I will present here several variations of modules (most of them coming later as variations of the layouts for mixed and Full-SMDs) of Mr. Moog`s great and unbeatable circuit on a new kind of added technique since up today, where these told variations are preferably splitted for Bass-, Middle- and Treble-Frequencies and therefore possible usages, using regular diodes, then transistors which got configured as diodes and on top of all: Prebiased, highly sensitive Transistors but with almost not too much hFE (well-matched lower C-Types may get recommended).

Another feature as part of the project I present here for maybe also otherwise universal usage, are the selective so-called "Voltage Addressed Switches", for a possible interchange between the external modulation-inputs (frequency-mod. or emphasis-mod.) and/or audio-inputs. Additionally these switches may get again voltage-modulated from externally sources, for really "tricky" usages!
By fact, these very useful function represents a 3-level-selector based on CMOS-Analog-Switches and depending on the applied values to a voltage-input (3-Bit A/D, see therefore the main schematic sheet no. 2).

How the circuit works:

The main function of the module is alltogether a standard Moog-Ladder (selected as "main feedback-loop with the resonator disabled"), with the extension of electronic Q-Control by the usage of OTAs and an implemented "Q-Damping-Cancellation", where the innovation and main difference to such formerly known concepts, is the fact that the audio-signal, as well as the feedback-loop, gets inserted on both sides (!) of the input-difference-amplifier and then accessing the ladder, also as the ladder - meant from its principle - is a truely symmetrical unit and therefore with an inverted signal for each insertion gives the best results.
Herefore the Dual-OTAs LM13700 or NJM13700 are a quite good choice, because they come almost as very closely matched parts in one package. The only critical data where those shurely will vary, and for to do any adjustments, is the offset and the overall current product, which may differ a lot between several ICs.

If the user selects the resonator, the feedback-signal-path is different and something really new, where the whole current-product from the ladder gets mirrored into a second ladderlike construction, where again a frequency- and amplitude-selection may pass, and which is completely free in configuration, of course reproducing a resonance within a feedback-loop.
Here you can implement what you like, capacitors, inductors - also maybe "electronic lossless inductors" are highly recommended (based on OTAs, and those will get discribed in a following project and again extending the concept).

For a prepatched useful situation at start-up, the module comes here with 100 pF for instance at all caps, what results in an almost not really different resonance-characteristic, compared to the standard feedback-loop, but here you may vary and try your own experiments. For changing those values, it could also get interesting to expand this section directly to the front panel, where the user may insert parts externally in useful sockets (maybe banana-jacks) and so on, could also of course all get possible...
This resonator´s configuration is maybe not that much for playing solis (but ?? - who knows), but more for getting strange noises and unusual sounds, fully dependable on the users choice and always staying experimental.
As this circuit-part is really new, time will tell and for the first fully featured series-modules I will order the first readymade PCBs for myself soon next days (costs...), - but until now all theory, little breadboarding and prototyping is done. A report from a friends lab is in progress and will follow by promise, and as I hope so, also coming as a youtube-video-demo for the series modules.

Additional information to the used and recommended TLE2037A "Excalibur-Opamps":
These are "uncompenated" and are a first choice replacement for the very expensive Burr-Brown OPA637. They need to be driven with a minimum amplification of 5 or more, to work in a stable condition. They will drive around a quiescent current of 5 mA for each used Opamp, so the module gets of course into higher power category if set up completely.

The main module is presented as several versions with the following features appropriating to all units:

1.) Heated Transistor-Section - for compensating a possible detuning on environmental temperature changes. This circuit is based on the LM3046 Transistor-Array, manufactured by Texas Instruments and these devices are quite valueable. This section is like all the others, still in an experimental condition and will ask for each module on a longer time of observation to work properly, because all such arrays differ a little. The presence of a good temperature-meter, or maybe a heat sensing camera, would be recommendable herefore. Critical values are indicated in the schematics.

2a.) Three main controls for setting the Filter´s Frequency. Tune, Modulation (2 Pots possible) & Envelope.

2b.) Two main controls for regulating the Emphasis, plus one additional for possible Envenlope-Control with external potentiometer (not implemented as on-board-position).

3.) Two additional Instrumental-Amps and Outputs for Highpass- and Bandpass-Function, for well-used diy´ers easily interchangeable to other possible signal outputs (maybe 3rd or 4rth order pole-grips, etc.). This means at all, three Audio-Outputs.

4.) Eight possible signal-inputs and two interchangeable modulation-inputs, also partly controlled by voltage addressed switches, where each pot may act for several selectable input-functions.

5.) Feedback-Loop with selection by CMOS-Switches for either Standard-Emphasis, or extended control with Resonator-Usage (Current-Mirror and therefore very fast action in that loop, is the most benefit).

6.) Multiple-Systems Patchbay-Connector for individual configuration adaption. Here the user may finally patch his own connections with possible single wires underneath the module, to fit into several possible brands, like the Doepfer-Bus, etc. and easily connecting into them and making the module versatile.

Adjustments (& Specials) without trimpots (for most cases not critical):

- Adjust R55, R66 for proper Q-Damping-Cancellation.

- Adjust R107, R108 & R129 for proper Main-Loop-Emphasis.

- All GE-Diodes may also vary as Schottky´s.

- For the first start-up it´s recommended to insert R190 with 220 Ohms, then change it to 100 Ohms and if the thing gets not more temperature than around 45 degrees celsius, finally replace it with Zero-Ohms. Optimized adjusted temp of heating is around 50 degrees celsius and may change widely on different manufacturers for the arrays, especially also on the recommended UL1111. Those are very good and come quite often with very close matching data, but also somtimes they may come with true "rascals". Finally, and in situation if nothing helps, vary very carefully R189 downwards until said temperatures, where I would personally say 55-60 degrees is almost too much.

The module runs on +/- 15 Volts and features max. five possible drills for mounting the units into individual mechanics, therefore giving high stability.

You may download the versions of the modules in folders indicated as "1" or "2" with completed Eagle-Files V-7.4 (all passives almost as 1206-SMDs) inclusive Electina-10-Files and all material as PDF´s, but without any simulation-diags, where Version 2 comes with all parts completely implemented as not too small SMDs.
Each folder holds also a sub-folder with prepared files for giving them directly into PCB-Manufacuring (optimized for Eurocircuits Design Rules), with the vias at 0.8mm diameter and with 0.4mm drills. Support for Non-Eagle-Owners is therefore also possible of course, if you refer to these PDFs, which are also presented as JPEGs.
Additionally I may serve - if needed - complete material sets, with especially selected Transistor-Arrays and as shurely needed key-parts complete sets with best-matched BC547´s (also for +/-12V-Versions contact jo2030 aahhtt gmx dot net and please no spam) and also with recommended Audio-OTAs NE5517D by Philips.

Have much fun with this thing and a nice day.

Read the full post
Show less
48dB-Ladder Module Version 1
Example 2nd Resonance shifted 11.09.2015
Example twice Resonance 11.09.2015
Another Snapshot varying the Resonator
1 kHz Sinewave converts into strangest Results 11.09.2015
High Pass Signal origins the same at 4 Hz inputs
2 kHz Resonance Nyquist - Allmost a donut
Resonance regular
Diode-Ladder-Square-Wave-Product - Plastic
Diodes 1N4151 - Sim-Frequency-Response
Variable Caps A
Variable Caps B
OTA out of SOA
Crystal Nyquist
Variable Caps Nyquist
48 dB Transistors diff. biased
Possible Waveforms - Tina Simulation
My Laptop has a Problem ????
Input is a triangle
And I wonder...
Project 24 & 48 dB Ladder 31-08-2013 - The Spear
24 dB - Varied caps in cell - Straight A
24 dB - Varied caps in cell - Straight B
24 dB with OTA as selfmade Makro - Mushroomhead
24 dB with OTA as ideal Makro - The Bow
Different Inputs A 5V-Square - saturation
Different Inputs B 6V-Square
The used Tina-OTA-Makro same as CA3080A Grrrrrr obsolete
Square changed to Triangle 10V Input on Maximum
Sinewave in
Sinewave with a little more Q-Current on the OTA
Sinewave switched to Square Caps 27nF_3nF
Resonator in 4. Cell - The spanned bowspring
Diode Ladder Squarewave-Product 0,5 Vin - 6 Vout
BCM847 Transistors as Diodes - Optimized
48 dB Transistors as Diodes Nyquist- Spanned Bowspring
24 dB Multiple Resonances on Squarewave possible !!!
All Resonators on 1 pF
All Resonators on 1 nF
All R 4 (5 - see circuit ) resonators on 10 nF
1 Kiloohm ch. to 3 Kiloohm - also LDR Phototransistors useable for light!!
Multiple Resonances possible
A little more current on OTA selects another frequency and intenses the first
48dB-Ladder Module Version 2

Loading comments...