This Week on the Breadboard: a BBD Reverb

[fig]

CoolAudio makes a V3005. Not rare, and not expensive (if you buy 100 at a time).

So we can be looking for a revision then? I’ve got a half-dozen I can throw at it.

 

[Chuck D. Bones]

I can sketch up a schematic, but I only have two MN3005s so you'll be the one breadboarding it.

 

[fig]

If you believe it a worthy effort I would give it a go.

 

[Driz]

I can sketch up a schematic, but I only have two MN3005s so you'll be the one breadboarding it.

Any progress on this? I've got a few 3005s laying around

 

[Chuck D. Bones]

It kinda slipped to the back burner...
I have a couple of things to finish up, then I'll see about drawing up a prelim schematic.

 

[KR Sound]

what about using Cabintech’s beta for this?

https://forum.pedalpcb.com/threads/digital-bbd-concept-looking-for-comments-and-beta-testers.18909/

 

[Chuck D. Bones]

I'd sooner use an FV-1.

 

[passtheducky]

what about using Cabintech’s beta for this?

It would be kind of silly to take a whole dsp that has the capability to run a complete algorithm and then build a primitive reverb out of it by treating it as a dumb delay line. Analog feedback is cool for delays, not so cool for reverbs.

 

[Rpschultz13]

I've been chasing this analog bbd reverb lately. I agree with some others, use the 3005 instead of the 3011? How easy would it be to tweak the circuit for that?

 

[owlexifry]

I've been chasing this analog bbd reverb lately. I agree with some others, use the 3005 instead of the 3011? How easy would it be to tweak the circuit for that?

+1

if this is better than belton brick - 100% interested in a multi mn3005 option

 

[Cabintech]

I did a back-of-the-envelope feasibility check on using multiple MN300X chips to reproduce the multiple taps points of the MN3011. From what I could tell this is as close as you can get to the same tap points with current production BBDs:

Number of Stages Comparison
	MN3011	CT3011
Tap	SumS	S1	S2	S3	S4	SumS	Diff	%Diff
1​	396​	256​	128​			384​	-12​	-3.0%​
2​	662​	256​				640​	-22​	-3.3%​
3​	1194​	256​	256​			1152​	-42​	-3.5%​
4​	1726​	256​	256​			1664​	-62​	-3.6%​
5​	2790​	1024​				2688​	-102​	-3.7%​
6​	3328​	256​	256​			3200​	-128​	-3.8%​

So this would be a 256 stage BBD (MN3009) followed by another MN3009 running at 2X speed, then 5 more MN3009's, an MN3005, and 2 more MN3009. You can scale the overall clock speed and get quite close to the MN3011 delay times if you run the overall clock about 3% faster:

Msec Delay Time Comparison
	MN3011	CT3011
	10KHz Clk	9.6KHz Clk
Tap	Delay ms	S1	S2	S3	S4	Delay ms	Diff ms	%Diff
1​	19.80​	13.33​	6.67​			20.00​	0.20​	1.0%​
2​	33.10​	13.33​				33.33​	0.23​	0.7%​
3​	59.70​	13.33​	13.33​			60.00​	0.30​	0.5%​
4​	86.30​	13.33​	13.33​			86.67​	0.37​	0.4%​
5​	139.50​	53.33​				140.00​	0.50​	0.4%​
6​	166.40​	13.33​	13.33​			166.67​	0.27​	0.2%​

The cost for (9) MN3009 and (1) MN3005 is over USD 100...

 

[Rpschultz13]

There are a handful of DIY reverb pedals based on the PT2399. How does the PT2399 compare to the MN3005/3009? I realize that the PT is digital and the MN3xxx.

 

[Chuck D. Bones]

PT2399 has a longer delay than MN3005 & MN3009 BBDs. This makes the PT2399 more appropriate for echo and the BBDs more appropriate for chorus and flanger applications. Their delay ranges overlap in the reverb arena.

The PT2399 contains analog filters and a VCO clock gen. It's pretty-much self-contained. The BBDs require additional hardware to provide the sample clock and filtering.

They both get noisier and more distorted at longer delay times.

Read the datasheets for more specifics.