30.12.2014 Views

Ultra Flanger

Ultra Flanger

Ultra Flanger

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

<strong>Ultra</strong> <strong>Flanger</strong><br />

Design by John Hollis<br />

The <strong>Ultra</strong> <strong>Flanger</strong>isJohn Hollis' design of avery flexible flanger based on<br />

the MN3007 bucket-brigade analog delay chip.<br />

U1A is an input amplifierand mixer.The inputsignal comes in through C1<br />

and R2 where it meets theRegen feedback signal from oneside of the<br />

Odd/EvenRegen switch.R1 is apulldown resistortoprevent popping when<br />

switched.R3 biases U1A (and through itU3 and U1B)to Vbias1, about4.9V.<br />

U1Ahas aslight gain boost.<br />

The outputfrom U1Agoes to two places: through R5 to the inputto U2 for<br />

delay processingand to the output through R12 to be mixed with the delayed<br />

signalout of U2.The signal level at the inputto U2 is limited byD1and D2 -<br />

these are not fordeliberate distortion.The output ofU2 is available atpins 7<br />

and 8, and this is thenapplied to U1B, which low pass filters thedelayed<br />

signaltoremove thesampling noise caused by U2's analogsampling.The<br />

lowpass filtered signalis mixed withthe drysignal through R13, and goes to<br />

the output jack throughC5.The outputofU1Balso feedsthe Regen control to<br />

determinethe amountof feedback to bemixed with the input.<br />

U5A serves asaSchmitt triggerto determine the min andmax size ofthe<br />

LFO triangle wave. U5Bis an integrator that ramps up and down, controlled<br />

bythe output signalofU5A. Theamount of U5A's output that is applied to<br />

U5B's input is controlled by the Ratecontrol;this determines how fast the<br />

integratorramps.<br />

U5B's outpugoes to the Sweep control,which determines how large a<br />

triangle wave is fed to the highfrequency clock generator, U4. The triangle is<br />

further smoothed by R23/C9.<br />

U4generates abiphase output on pins 2and 3/4 at aspeed determined by<br />

the voltageon its pin 9.This outputis between 50kHz and1MHz. It is buffered<br />

bythe paralleled sections ofU3, andthis drives theclock inputs of U2.<br />

Count Value Designation(s)<br />

5 1N914 D1,D2,D3,D4,D5<br />

2 Red LED1,2<br />

2 DualOpamp U1,U5<br />

1 MN3007 U2<br />

1 CD4049 U3<br />

1 CD4046 U4<br />

1 100R R27<br />

1 470R R21<br />

1 2.7k(2K7) R24<br />

5 10K R12,R13,R15,R17,R5<br />

1 12K R16<br />

1 22K R22<br />

1 39k R24<br />

4 47K R11,R19,R7,R8<br />

1 68K R3<br />

7 100K R10,R14,R18,R2,R23,R6,R9<br />

Count Value Designation(s)<br />

1 150K R4<br />

1 470K R20<br />

1 3.3M (3M3) R26<br />

1 10M R1<br />

1 5Ktrimpot Tr3<br />

2 100Ktrimpot Tr1,Tr2<br />

1 22pF C10<br />

2 0.001uF(1nF) C3,4<br />

1 0.0047uF(4n7) C2<br />

1 0.1uF(100nF) cer C11<br />

2 0.1uF(100nF) C1,8<br />

2 1uFradialelectro C5,C9<br />

1 10uFrad electro C7<br />

1 100uFrad electro C6<br />

6 0.4" wirejumper J1,J2,J3,J4,J5,J6<br />

Revisions History<br />

10/26/01:Firstrelease.<br />

04/07/02:Corrected schematic and layoutper John Hollis Instructions.<br />

Previous Hollis schematic incorrectly connectedC9 and R24<br />

toVbias,not ground.<br />

IN<br />

From<br />

Stomp<br />

Sw<br />

C6<br />

C1<br />

1<br />

R1<br />

+9<br />

Odd/Even<br />

TR3*<br />

See Text!<br />

TR1<br />

TR2<br />

C2<br />

R4<br />

R6<br />

REGEN<br />

+9<br />

1<br />

appx 4.9V +9<br />

R2<br />

1<br />

2<br />

8<br />

C3 6 8<br />

- 1 R5 3<br />

-<br />

U2<br />

7<br />

-<br />

R3<br />

U1a<br />

R8<br />

U1b<br />

3<br />

7<br />

5<br />

+ +<br />

R13 +<br />

4<br />

C5<br />

R11<br />

R15<br />

+9 D1<br />

4 5 6 2<br />

appx 4.9V<br />

appx 4.9V<br />

D2<br />

1<br />

C4<br />

+9<br />

15 14<br />

R14<br />

R17 1<br />

12<br />

1 appx 3.8V<br />

11<br />

1<br />

+9<br />

appx 4.9V 2<br />

10<br />

LED1<br />

50kHz to 1MHz<br />

10<br />

U3<br />

LED2<br />

50kHz to 1MHz<br />

R27<br />

R16 C7<br />

2 3<br />

D3<br />

justunder 9V<br />

4 5<br />

R7<br />

R9<br />

R10<br />

R12<br />

OUT<br />

to<br />

Stomp<br />

SW<br />

+<br />

-<br />

R19<br />

R18<br />

U5a<br />

R20<br />

2 Flips from<br />

2<br />

over7V<br />

down to R21<br />

almost0V<br />

C8<br />

-<br />

U5b<br />

+<br />

R22<br />

Ramps up<br />

and down<br />

around 3.8V<br />

D4<br />

D5<br />

2 2<br />

appx 3.8V<br />

R23<br />

+<br />

C9<br />

-<br />

6<br />

8<br />

7<br />

R24<br />

9<br />

C11<br />

2 3 4 14 16<br />

U4<br />

5 8 11 12 6 7<br />

R25 R26 C10<br />

2<br />

Copyright 2001 R.G.Keen.All rights reserved. No permission forlocal copies ordisplayfrom web sites other than http://www.geofex.com.


<strong>Ultra</strong> <strong>Flanger</strong><br />

Design by John Hollis<br />

Making thecircuit board<br />

I've received som comments that the normal trace<br />

spacing Iuse gets eaten awayabit by printerand<br />

file conversion tolerances, making the pads and<br />

traces abitfatter. Because this is avery busy<br />

board, I've pre-compensatedfor that by making the<br />

majority of the tracesabit thinner.Most arelaid out<br />

as0.020" (20mils)wide.<br />

Tomake thecircuit board,print the pattern to<br />

tonerpaper.Press-N-Peel Blue is recommended,<br />

because the smaller traces will betrickier withother<br />

tonertransfertypes. Thepatternis thecorrectway<br />

'round fortonertransfer printing.<br />

Irecommend printing thepatterndirectly ontothe<br />

paper. Agood way todo this is to printthis sheet<br />

onto paper,then cut out arectangle of PNP Blue<br />

that is slightly larger than the toner section. Tape<br />

this to the previously printed paper sheet; tape<br />

along theleading edge of the tonersection. Now<br />

printthe sheetagain, manually feeding it into your<br />

printer.Resolution and accuracy will be best on an<br />

original print,not aprint thatis then copied.<br />

Follow the instructions for the toner sheet. Clean<br />

the copper board very,very well. Mostfailures are<br />

due to improper cleaning ofthe board. Iron the<br />

sheet onto yourcopperclad blank. If you look atthe<br />

shiny sideof the sheet at agrazingangle,youcan<br />

see the slight indentation where the toner on the<br />

bottom side is adhering to the copper.Do not iron<br />

somuch thatthe pattern spreads. If you foul it up,<br />

justclean the toneroff the copperwith acetone,<br />

then printand iron another. Do any necessary<br />

touch up withaSharpie marker or aRadio Shack<br />

etch resistpen -which Ithink is arelabeled<br />

Sharpie.<br />

Etchin your favoriteetchant. Ilike ferric chloride,<br />

even though it's messy and smells bad.<br />

Drillwith a0.030" drillbit.It helps to indent the<br />

centers of the pads abit withasharp pointe too like<br />

anice pick.<br />

Input<br />

DPDT<br />

Stomp<br />

Sw<br />

Odd/<br />

Even<br />

Regen<br />

100KLog<br />

Output<br />

Rate<br />

100KLog<br />

ToControlline,<br />

Millenium Bypass<br />

Battery<br />

Clip<br />

Sweep<br />

100KLog<br />

Manual<br />

100KLog<br />

The fifth pad from the leftisaground pad. Itcan<br />

beused to ground the outputjack,orjus left open<br />

asin the wiring diagram.<br />

Notes on the layout:<br />

I've addedacouple of things to the design.R27/C11 decouples the digital noise<br />

from U3 and U4 from the analog+9V.You mightbe able to sub in ajumperfor<br />

R27and leave C11 out if you like living dangerously. Trimmer TR3 is there for<br />

tweaking in the analog bias voltage.The pads of TR3 are shorted togetherby<br />

thin traces onthe PCB. If you wantto use TR3,make R15 9.1K,use5Kfor TR3,<br />

and changeR16 to 10K; cutthe shorting traces between the pads ofTR3. If you<br />

justwant it simple, leaveoffTR3, makeR15=10K and R16=12K as per the parts<br />

list.<br />

I've leftthe option for a3.9V zenerinstead ofLED1,2, and D3. If you preferthis,<br />

put the zener in the space on the board forD3, butwith the cathode the other<br />

'way round. LED1 and 2arespaced so thattwo 0.4" long wire jumpersfit neatly<br />

into theirholes.Ifyou put in thejumpers and the zener, thezenercircuitworks as<br />

shown in John's original schematic.<br />

I've included some typical DC voltages as an aid to debugging.<br />

4/8/02:The layoutatleft is the updated onewith the newestfixes from John H.<br />

Copyright 2001 R.G.Keen.All rights reserved. No permission forlocal copies ordisplayfrom web sites other than http://www.geofex.com.


Modsto Fix the PreviousBoardLevel<br />

BoardisshownfromtheCopperSide!!<br />

Copyright 2001 R.G.Keen.All rights reserved. No permission forlocal copies ordisplayfrom web sites other than http://www.geofex.com.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!