Simple BBD circuit?

[stfala]

Anybody know of any really simple delay circuits using a BBD? I understand the need anti-aliasing filters, reconstruction filters, clock drivers etc.. But are there any pretty simple circuits in existence where other circuit components are kept to a minimum but a decent output is still obtained?
Just wondering as I'm doing a comparison between three different delay circuits for a uni project and I don't want to get completely bogged down in the one circuit with the timescale I have.
Thanks in advance for any pointers.

[garcho]

John Hollis' Zombie Chorus or Ultra-Flanger

delay circuits

oops! sorry, i guess i skipped over that word when i read the OP

[mth5044]

What BBD do you want to use? How many ms of delay are you looking for? There is a very low part count ADT circuit posted here, I think it was something less than 20 parts.

[stfala]

I'm looking to use an MN3007. I'm quite open to delay time, as long as it's long enough so that it's more of a delay effect than just an echo (does 20mS sound reasonable?). I'm not looking for a ridiculously long delay time either though. A decent middle ground between the two, but ultimately a simple circuit design with a decent output.

[hymenoptera]

Did you mean 200ms?

[mth5044]

20ms is not quite long enough for a delay. The MN3007's datasheet says it can go to 50ms or so, so if you did actually want 20ms, you'd be good, but you're in chorus times there.

Check out reply 17 by marteen for a good start for a very small parts count delay.

http://www.diystompboxes.com/smfforum/index.php?topic=84323.msg710076#msg710076

[stfala]

Thanks a lot, I'll check that out!

[slacker]

John Hollis' Zombie Chorus or Ultra-Flanger

delay circuits

oops! sorry, i guess i skipped over that word when i read the OP

Still not a bad call, if you make the timing cap on the CD4046 bigger, slowing it down, you should be able to get into delay territory. Add a wet dry mix and the ultra flanger circuit is basically a delay.

[armdnrdy]

By reading through the OPs description of delay/echo requirements...

Seems to me he would want to go with a MN3008 circuit.

A single MN3007 isn't going to give you much of a delay...

At the longest delay setting...just a fast slap back delay.

Not very versatile.

[Scruffie]

Schematic for the v3205 EHX Full Double Tracker, 50ms or 100mS settings.

[Mark Hammer]

A pity George Carlin didn't live longer, or else he would have included "simple BBD circuit" along with his collection of other oxymorons, like "jumbo shrimp", "tight slacks", military intelligence" and "government efficiency".

In general, complexity in any BBD circuit stems largely from what is involved in extending the delay time out to the maximum a chip can comfortably handle.  If one is aiming for 20ms from a 2048 or 4096-stage BBD, then the clock frequency will be up high enough that all the lowpass filtering, companding, etc., normally needed to achieve more delay time without hideously audible clock noise, will be maybe not entirely unnecessary, but at least optional.

The other thing is that, unlike something like a PT2399-based delay, the clock signal has to get from somewhere to somewhere else, and that begets attention to layout.  For the all-in-one digital circuits, the clock may well be something that warrants attention, but it won't require attention to layout.

If you are in possession of an MN3007 and wish to make good use of it, a flanger or chorus is likely your best alternative, keeping in mind that the shorter the delay time aimed for, the higher the clock frequency, and the less lowpass filtering needed.

[allesz]

Mr. Hammer is right: bbds are hard use for self tought diyers, I tryed to use them but always failed.
But marteen's schem gave me the wish to try again.
I guess I will loose some sleep, and patience, on the breadboard again.

[allesz]

Hallo, I have a question: 3208 and 3205 will need a 3102 clok chip in marteen schem?

[Fender3D]

3101 is for 30xx series, whereas 3102 is for 32xx

[allesz]

Thanks mate (grazie?).
I will place an order for the parts... wish me luck.
Beside the basic adt, I was thinking that I shoul try also the eh double traking with the 3102 and a 3205.

[Mark Hammer]

3101 is for 30xx series, whereas 3102 is for 32xx

The circuit shown use a 4047 to supply the high frequency clock.  The 3102/3101 were produced by Panasonic to provide a convenient and small package for generating complementary clock pulses.  Many other circuits and chips can do the same thing.  So, while you are correct about which clock generator chip goes with which BBD, in this case, neither are used.

[Fender3D]

Thanks mate (grazie?).

you're welcome (prego) 

and... buona fortuna!!

[stfala]

Thanks again everyone for your input.
Think i'll go for an MN3008 after reading through all your replies. The actual project brief is to compare three different kinds of delay, so I'm building this BBD based circuit as well and a circuit using the PT2399 and one using a programmed PIC.
I think i'll try and do something close to the application circuit on the 3008's datasheet (for the sake of simplicity), and aim for 100ms delay time.
Can anyone tell me if the internal topology of the MN3008 is similar to the MN3007 (i.e. is is just two 3007's cascaded together)? There is a lot of useful stuff concerning the 3007 on the Electrosmash site but not much regarding the 3008.
I may have more questions as this project progresses.
Cheers.

[samhay]

If you need a plan B, there are other ways you could consider generating delay.
A low tech method would involve driving a speaker that is some distance from a microphone receiver - you would need about 100 feet for 100ms delay, which while not very practical, could at least look quite impressive.

[duck_arse]

I'd say a reverb spring generated a delay, too.


(or, leave a message on the teacher's answering machine.)