AMZ FX ANALOG ALCHEMY GEOFX MEF ROG MEF DIY WIKI

Vintage MXR Flanger problem

Started by Govmnt_Lacky, September 07, 2011, 09:49:21 AM

Previous topic - Next topic
Print
Go Down Pages 1 2 3
User actions

Govmnt_Lacky

#40
September 10, 2011, 04:22:53 PM
Quote from: Fender3D on September 10, 2011, 03:32:45 PM
SAD were used as delay line in older scopes to record waveforms... I recall an older Elektor project doing just this (it was "doable" it had 2 different separated clocks without heterodyning on the same device) and the claim "The sections may be used independently" must allow this operating setup  :icon_exclaim:

This statement hit me like a ton of bricks!!  :o

We have a crap ton of broken/unused oscopes at my work that are just sitting and collecting dust! A lot of Tektronix models and some others. I wonder if some disassembly is in order??  :icon_cool:
A Veteran is someone who, at one point in his or her life, wrote a blank check made payable to The United States of America
for an amount of 'up to and including my life.'

Fender3D

#41
September 10, 2011, 06:13:41 PM
Sadly I guess they might have been used in cheap scopes (Elektor's project was translated Videoscope in italian. it was a scope out of a TV)...
"NOT FLAMMABLE" is not a challenge

Govmnt_Lacky

#42
September 11, 2011, 03:19:33 PM
Quote from: Fender3D on September 10, 2011, 06:13:41 PM
Sadly I guess they might have been used in cheap scopes (Elektor's project was translated Videoscope in italian. it was a scope out of a TV)...

Won't know until I try... right?  ;D
A Veteran is someone who, at one point in his or her life, wrote a blank check made payable to The United States of America
for an amount of 'up to and including my life.'

jdub

#43
September 18, 2011, 04:29:39 PM
OK, so I think I understand how to do the 512-stage mod with a bad-B-side SAD, but I wanna check before I try it.  If I understand correctly, all one is really doing is eliminating the input signal from (in F3D's suggestion) the A-side of the SAD; going from the stock schematic, this entails taking the signal that would normally go into pin 2 via C1 (from junction C1/C12) and rerouting it to enter pin 15 via C6 then taking your output from pins 11/12.  Thus, eliminating input signal from the first half of the SAD.  If I want to do an analogous move with the B-side, it seems all I gotta do is take the output from pins 5/6 and jump that straight to R48 trim, right?  That bypasses the B-side;  my question is whether or not I need to bypass R10.  I assume yes, since in the A-side mod we have nothing between pins 11/12 and R48.  Am I correct in this assumption?  Thanks!
A boy has never wept nor dashed a thousand kim

Fender3D

#44
September 18, 2011, 09:15:23 PM
Mod for a bad B side is somewhat trickier since you must disconnect pin 11 and 12.
You should remove R48 and connect it instead of R10 and R11, then connect R48's wiper to C7.
Just 1 more wire (if there's no room on PCB)
"NOT FLAMMABLE" is not a challenge

Govmnt_Lacky

#45
September 18, 2011, 09:17:31 PM
Quote from: Govmnt_Lacky on September 10, 2011, 04:22:53 PM
Quote from: Fender3D on September 10, 2011, 03:32:45 PM
SAD were used as delay line in older scopes to record waveforms... I recall an older Elektor project doing just this (it was "doable" it had 2 different separated clocks without heterodyning on the same device) and the claim "The sections may be used independently" must allow this operating setup  :icon_exclaim:
We have a crap ton of broken/unused oscopes at my work that are just sitting and collecting dust! A lot of Tektronix models and some others. I wonder if some disassembly is in order??  :icon_cool:

After a few hours of digging...... the scopes came up EMPTY on usable ANYTHING!  :'(
A Veteran is someone who, at one point in his or her life, wrote a blank check made payable to The United States of America
for an amount of 'up to and including my life.'

jdub

#46
September 18, 2011, 09:36:30 PM
QuoteMod for a bad B side is somewhat trickier since you must disconnect pin 11 and 12.

What if I just use a socket with pins 11 & 12 lifted, and run a jumper from 5/6 to the 11/12 trace (that goes to R48)? 

Greg, sorry to hear that the 'scopes didn't pan out... :-\
A boy has never wept nor dashed a thousand kim

oldschoolanalog

#47
September 18, 2011, 10:21:46 PM Last Edit: September 18, 2011, 10:25:05 PM by oldschoolanalog
Paging Kaiser Soze please... :icon_lol:

Are YOU dancing to the Radio? Well, what RU waiting for???
Mystery lounge. No tables, chairs or waiters here. In fact, we're all quite alone.

jdub

#48
September 18, 2011, 11:13:08 PM
Just call me "Verbal" Kint  :D
A boy has never wept nor dashed a thousand kim

jdub

#49
September 20, 2011, 01:53:17 PM
Had a chance last night to try the 512 mod...and it works well!  On the Shredaholic board, I just lifted the end of R10 that connects to pins 5/6 and installed a wire at that pad; then I lifted the pin of R48 that normally connects to pins 11/12 and connected the other end of the wire to that.  I didn't have time to fiddle with C21 & optimally dial in, just a quick adjustment of R48 and R49, so to my ears wasn't quite as intense as before (due to less stages?) but it's close.  Proper calibration might improve this.

Since I'm by no means an expert on flangers, I'm puzzled as to why this works.  Are the full 1024 stages unnecessary?  Why use them, then- better chip availability (SAD1024 vs SAD512(D))?  Because they are there?  To be honest, I've never been quite clear on parallel vs series mode in SAD-using flangers, either.
I understand why my bad SAD won't work in a stock M117 (because the signal is being routed through the bad side in a series config, which the mod bypasses), but why does that same chip work in my Electric Mistress (parallel)? Is it because the signal is split in the chip, and the box will still function with only one side passing through?

Anyway, Federico, consider your mod verified.  Kudos and thanks to you! If by some strange quirk of fate we ever meet, I shall buy you several beers.  ;D
A boy has never wept nor dashed a thousand kim

Govmnt_Lacky

#50
September 20, 2011, 02:06:07 PM
@jdub

Good on ya' for the effort  :icon_wink:

Concerning your question about why it will work in the EM with one bad channel.... Does the Balance trimmer need to be cranked to one side to get the effect? With the balance trimmer fully CW or CCW, this would essentially only let one channel through.

Can you put a little bit more detail into the differences in the effect between the 512 and 1024 stages? Are the controls still as interactive on the 512 setting as on the 1024?

How much trimmer adjustment was needed (if any) to get good flanging with the 512 mod?

So many questions....  ;D
A Veteran is someone who, at one point in his or her life, wrote a blank check made payable to The United States of America
for an amount of 'up to and including my life.'

Fender3D

#51
September 20, 2011, 02:46:06 PM
jdub 1 , Govmnt_Lacky ? ???

:icon_mrgreen:

You welcome  :icon_cool:
Now you just have a little more...  :icon_evil:

I guess the "lesser intensity" deserves some consideration...
Less stages means you have less delay time @ the same clock, less delay time means higher comb filtering.
Intensity should be dictated by Dmax/Dmin ratio, and this should be, more or less, the same.
Nevertheless an higher comb filtering may look like a lesser intensity... but I'm sure you can achieve a better jet sound.

You should check the higher filtering won't play outside guitar frequencies range... lowering clock is the easy way, lowering clock at the same Dmax/Dmin ratio will be the hard way tinkering with caps...  :icon_neutral:
Of course, lowering clock too much will arise clock noise/whining, preventing you to get the stock delay time

About EM, with parallel-multiplex you double the waveform's samples: even samples are managed by one BBD, odds by the other.
If you loose one block, you always have a waveform following the input signal, just not perfect, but a little distortion may be unnoticeable yet the flanging effect might cover it...
"NOT FLAMMABLE" is not a challenge

jdub

#52
September 20, 2011, 03:11:58 PM
@Greg-
Lemme get back to you with more details...I didn't have much time to fiddle with it last night.  First impression was that controls were still interactive, good range of sounds, but a bit less "in your face" (which I guess is just anopther way of saying "less intense"  :-[).  For example, max speed seemed a little slower and boosting regen seemed a little noisy (but my build isn't boxed yet, so take with a grain of salt). As soon as I get a chance (hopefully tonight), I'll play around with Federico's suggestions and see if I can improve it a bit, maybe get a couple clips.

@Federico
Thanks again, dude!   8)
A boy has never wept nor dashed a thousand kim

oldschoolanalog

#53
September 20, 2011, 03:36:31 PM
Re: Series vs. parallel. It's all about effective sampling & clock cancellation. Read that Reticon SAD1024 datasheet I linked to in an earlier post carefully.
Want more intense? Play around w/the feedback loop. If you can't get it to self oscillate then you haven't reached max intensity. Get it to the point where it self oscillates, then back off a bit.
Also, proper BBD bias is very important when it comes to getting the best overall sound/intensity.
BTW, in theory anyway... If you are going to halve the stages used (1024 to 512); you will need to also halve the clock min/max to keep the min & max delay times the same. If the sweep just keeps going up, up, up; you are going to run out of useful sonic information for the flanger to process. W/a clean guitar there really isn't all that much going on up there. However if you run the axe through a nice Foxx Tone Machine OR a synth, watch that baby come alive in the upper part of the sweep. You are still limited by the units LP filtering and the inherent (very) limited bandwidth of the speakers used w/guitar amps.
Food for thought...
Dancing yet? ;D
Mystery lounge. No tables, chairs or waiters here. In fact, we're all quite alone.

jdub

#54
September 20, 2011, 04:02:19 PM
Dig it.  Like I said, I was just trying to see if the thing would work; I didn't have time to focus on it too deeply.  I'll work on it tonight, hopefully...

I actually read through the SAD datasheet a number of times a couple years ago when I first built an EM, cuz I was attempting to use 2 SAD512s wired in parallel to replace 1 1024, which I didn't have at the time (it did work, BTW).  But when I read it then I didn't really understand it, being new to BBDs and electronics in general; I was more interested in quick n dirty info for a specific result...I'll have to read it again as I'll probably get more out of it now...Thanks Dave
A boy has never wept nor dashed a thousand kim

12Bass

#55
September 20, 2011, 08:03:56 PM
An SAD1024 used in parallel-multiplex is essentially the same as a 512-stage BBD in terms of delay time, but with double the system sample rate (because alternating samples from each section are combined at the output); it might be said that parallel-multiplex offers double resolution analog sampling for a given sampling rate.   Using only one half of an SAD1024 in such a configuration would effectively reduce it to a regular single 512-stage device at the nominal system sampling rate.  Finally, removing one half of an SAD1024 in a serial configuration would reduce the delay time by half and therefore shift the range of the flanger up one octave (as mentioned above). 
It is far better to grasp the universe as it really is than to persist in delusion, however satisfying and reassuring. - Carl Sagan

jdub

#56
September 21, 2011, 10:05:24 PM
Dave and Federico, taking to heart your reminders about half the stages/halve the clock, last night I changed my C21 to a 56pF, adjusted the trim to get ~18kH - ~580kH, tweaked bias and distortion trims got it sounding pretty good.  Nice rich sound, full range of controls, maybe still a tad noisy.  Will play with it more when I get a chance...Thanks again.  You guys rule  8)
A boy has never wept nor dashed a thousand kim
Print
Go Up Pages 1 2 3
User actions