Discrete BBD

[bioroids]

Hi!

That thread about early discrete effects got me thinking: why not to build a discrete 20 stage BBD?

Just for experimenting at least, that wouldn't be too hard. If clocked at 10Khz then 20 stages can give 1ms delay right? That can be used in a flanger at least for learning purposes. Also can be used for signal deterioration for those who like that.

The question is, where can one find a schemo for a discrete BBD? I know there are some on MN300x datasheets, but they appear to be too simplistic.

I know this is too DIY

Luck

Miguel

[toneman]

why not??

why not search the web for a descrete BBD??

If U know how they work, U should B able to build one.
At least it doesn't use tubes  (sorry PT)

Hope U have a BIG breadboard.   
Say, mayBe the size of a kitchen table 

staydescrete
but, play loud
tone

[puretube]

OK - here you go:
the first large scale produced BBD was done according to:
F.L.J. Sangster, for Philips;

of course (STAYTUBED... ), that development was based upon:
earlier tube-circuits ,
like this one ,
this one ,
this - (oops: SS...) one ,

and in case one should need a sample & hold circuit in front of it...

And here an idea for "discrete reversed delay" ...

speaking of reverse (and/or scrambled...):
http://www.pat2pdf.org/patents/pat3773977.pdf

[puretube]

not to forget the:

PURE TUBE TAPPED DELAY ...

[bioroids]

Thanks!

Those patents have plenty of info.

I don't know if I will get to build this, but I'd like to know the chances and options.

Luck

Miguel

[Steben]

Please remember though that a flanger effect is reached when the delay is sweeped between 1ms to at least 10ms. But as Mark Hammers notes, with up to 20ms you get finally real deep flanging. To reach 1 to 12ms many commercial boxes use 1024 stage BBD's. Chorus is the same theory but at higher minimum delay let's say 15ms and from there on lesser need for sweep. Useless to say that this explains when how deeper your flanger is (bigger sweep) the more you can make a chorus out of it by lowering the depth (raising the minimum delay). Following?

Classic chorus/flanger: The change in delay is reached with change in clock frequency, which is achieved with a HF oscillator (clock freq) driven by a LF oscillator (driving the clock freq).

[Mark Hammer]

Please remember though that a flanger effect is reached when the delay is sweeped between 1ms to at least 10ms. But as Mark Hammers notes, with up to 20ms you get finally real deep flanging. To reach 1 to 12ms many commercial boxes use 1024 stage BBD's. Chorus is the same theory but at higher minimum delay let's say 15ms and from there on lesser need for sweep. Useless to say that this explains when how deeper your flanger is (bigger sweep) the more you can make a chorus out of it by lowering the depth (raising the minimum delay). Following?

Classic chorus/flanger: The change in delay is reached with change in clock frequency, which is achieved with a HF oscillator (clock freq) driven by a LF oscillator (driving the clock freq).

Whoa, there, pardner.  The best-sounding slow "jet-plane" flanging (and remember that there are a whole lot of other VERY nice sounds made by flangers that do not meet or need these specifications) occurs when three conditions are met: 
a) there is an extremely "wide" sweep (meaning a very large contrast between the minimum and maximum delay times achieved, often expressed as a ratio, such as 20:1 or 30:1)
b) the minimum delay achieved is at a point very close to zero delay or some delay time where any potential notches introduced are inaudible,
c) the sweep cycle (LFO waveform) proceeds in a manner that accentuates that point in the sweep cycle where small differences in the location of notches is easily noticed

There is a fourth condition - the to-be-processed signal is generally wide bandwidth (the closer to white noise, the better) - but this is more a property of where one uses flanging rather than the flanger itself.

Though flangers can go out to 15msec delay and beyond, there is no need to do so to achieve "deep" sounding flange tones.  A sweep from 250usec to 10msec is still a 40:1 sweep ratio and likely to achieve quite satisfying tone, provided the other conditions listed are met.

I'm going to go out on what I think is a fairly sturdy limb here, and claim that once one is out beyond 12msec delay or so, the possibilities for further delay increment to be *perceived* in terms of pitch shifting (vibrato-like wobble) requires that MUCH closer attention be paid to the LFO waveform to reduce undesirable wobble, or at least make it least obvious.

[toneman]

As always, great links PT!! 

And, yes,  of course, "tubes did it first!"

U need a bumper sticker:
"I did it first with tubes!!"


that's assuming U have a car (?) 
Oh, *and* a bumper....   

Mark, U didn't mention "HyperFlanging" this time.
Where the sweep of the flanging is log rather than lin.
Keeps the flange in the low frequencies seemingly longer due to the human ear's log response.
PAIA did it with their "HyperFlanger/Chorus" wayyyy back when.
Wayy back when Curtis chips & SAD1024s were cheap & available.
I believe, Craig Anderton was the "Hyperflanging" originator(?)
U (still)have an article about this on your site...

I've always loved the chorus & tapped delay of the Rockman x100.
The "Boston" chorus, 2 me, is just about the best.
*Especially* in stereo.. 
Not 2 mention the jfet overdrive that definitely *IS*  "Boston".
 

stayhyper(flanged)
tone

[Mark Hammer]

Mark, U didn't mention "HyperFlanging" this time.
Where the sweep of the flanging is log rather than lin.
Keeps the flange in the low frequencies seemingly longer due to the human ear's log response.
PAIA did it with their "HyperFlanger/Chorus" wayyyy back when.
Wayy back when Curtis chips & SAD1024s were cheap & available.
I believe, Craig Anderton was the "Hyperflanging" originator(?)
U (still)have an article about this on your site...

Craig may have coined the term "hyperflanging" but I suspect that the virtues of a log sweep were known well before he coined the term (in the Deluxe EM and even earlier if I'm not mistaken; I'm sure Ton has a patent from 1937 or whatever that claims the same thing  )

All of that stuff *should* be still up on my page, but of late, some things seem to have mysteriously disappeared or have access blocked.  I'mnot the guy to speak to about that.  Steve "T-Boy" Morrison, who maintains Ampage, generously offered me some space and an interface for posting stuff.  I do the posting but none of the software.  I know why some new things initially have failed links and can correct that, but I have no explanation for why some older things that have always been accessible suddenyl show up as not present or as lacking permission.  I also have no way to delete the entry and re-post.

[Paul Perry (Frostwave)]

I believe the first manifestation of what we would consider an electronic BBD was the "Discontinuous Low-frequency Delay Line with Continuously Variable Delay" J.M.L.Janssen, (Holland) printed in Nature vol 169 p148 jan 26 1952. G.A. Philbrick (yeah, the tube op amp mnufacturer) independently invented it on Jan 19 the same year. Philbrick was the one who called it a BBD. Philbrick used 11 12AT7, 10 12AU7 and 6 12AX7 in the proto, with a clock (external) from 1K to 500KHz.

[puretube]

Mark: sorry - this time I can`t go back that far:
http://www.pat2pdf.org/patents/pat4038898.pdf (Norlin, 1975)
http://www.pat2pdf.org/patents/pat4144790.pdf (ARP, 1977)
for non-linear sweeping...

[bioroids]

This is very interesting info, though I never built anything with valves and I'm not gonna start with a tube-bbd that's for sure!

How many years you have to wait before using the patents? 20?

Luck

Miguel

[puretube]

for your own, private, non-commercial, single use: you don`t have to wait...