Madbean Total Recall: MN3005s and MN3008s

[spacekid71]

Hi everyone,

I am looking into the Madbean Total Recall build and I need someone to verify if
I understand the BBD requirements correctly:

Total Recall: 2x MN3005s

Total Recall + TRXPN: 4x MN3008s

Or is it possible to mix the MN3005s and the MN3008s...

Cheers,

Mart

[armdnrdy]

Post a schematic please.

[spacekid71]

The project pdf is located at http://www.madbeanpedals.com/projects/TotalRecall/Total%20Recall.pdf

[Mark Hammer]

MN3008s and MN3005s can always be used interchangeably.

However....

1) Like any BBD, the optimum bias setting might be different.
2) The min/max delay-time produced from 2048 vs 4096 stages will be different.

If one uses a quarter of 3008s to do what a duet of 3005s do, then there should be no need to adjust the filtering or clock.  If one were to use, say, a 3005 and 3008 (because you didn't have a second 3005), then you'd get less overall delay time.  If you tweaked the clock to get the same delay time as 8192 stages, you'd likely need to re-tailor the lowpass filtering.

But apart from those caveats, nothing is going to blow up or fry.  Plugging in an MN3005 and MN3208 to the same circuit, on the other hand....you don't wanna do that.

[spacekid71]

So if I understand correctly, if I would go for the Total Recall + expansion board, I would use 4 MN3008s.

Alternatively could I use 4 MN3005s with the circuits being designed as they are?

Total Recall: http://www.madbeanpedals.com/projects/TotalRecall/Total%20Recall.pdf

TRXPN: http://www.madbeanpedals.com/projects/TotalRecall/TRXPN.pdf

[Beo]

I know that Madbean does NOT recommend 4 MN3005's, and in all the forum posts and build reports, I haven't seen a single post where someone has tried it. Could work, but not verified. Fork out the money for 4 MN3005's and give it a try!

[spacekid71]

So I guess that with the expansion board the only option would be to use 4 MN3008s right?

[armdnrdy]

MN3008s are less expensive and easier to source than the MN3005s.

That is why you are given the option.

As Mark pointed out...the filtering would have to be redesigned to use any more than two MN3005s....even then...the repeats would probably sound horrendously distorted.

[Mark Hammer]

The Total Recall does not have the buffering that would be needed to drive/feed more than 8192 stages of delay, which is why a quartet of 3005s is contra-indicated.  Just HOW one reaches the maximum of 8192 stages is your own business (4 x 3008, 2 x 3005, 8 x 3007, (1 x 3005)+(2 x 3008), etc. )
The rationale for the limit is explained here:  https://www.premierguitar.com/articles/25035-behind-the-bucket-brigade?page=1

[spacekid71]

Thanks very much Mark!

[illuminatiNPS]

Mark, what would you consider to be the best source of info on BBD technology. Not finding much in the old Art of electronics book and want to study the topic more.

[Mark Hammer]

Panasonic published a BBD manual 30+ years ago.  Unfortunately, it's 16meg and exceeds my e-mail privileges, or else I'd mail you a copy.  Perhaps somebody knows the link where I got it from?

[illuminatiNPS]

Wouldn't happen to be this would it?? http://www.sdiy.org/BBDHell/BBD-Manual.pdf

[bean]

The way the expansion board works is it inserts between BBD stages 1 and 2. This pushes the gain recovery between BBD3 and 4 when using 4xMN3008. This is maybe less ideal than inserting it between BBD2 and 3 but it's the compromise that makes an expansion board possible. So, it's not impossible to use 4xMN3005 or a mixture of MN3008 and 3005 to push the maximum delay beyond 550ms, it's a question of acceptable results as others have pointed out. I haven't tried it myself because the two listed configurations is already complicated enough for most people who are going to build it (IMO).  Of course, someone could always design a larger expansion board that includes more gain recovery on it to make 4xMN3005 work-able. I never considered it; it would be a pretty large board and I think it would be a bit of a nightmare to deal with in terms of support 

[ElectricDruid]

I tried using 4 x MN3005s once, attempting to build a analog tap tempo delay. Not trying that again! 

Firstly I had to overcome the clock drive issue, which meant adding lots of buffering. I used some CMOS buffer chip, IIRC. That part was ok. But once you're done, you've got something which is like a dark, noisy old BBD with plenty of distortion, except at least four times as noisy, four times as distorted, and still dark as hell.
For me, it was a waste of time. For long delays, I went digital. Analog is cool for short stuff though; chorus, flangers, slapback, etc etc.

Tom

[Mark Hammer]

Current issue Deluxe Memory Man 1100TT uses a quartet of Xvive MN3005s for over a second of delay (hence the 1100).  I have no idea what they did to clear the 8192-stage hurdle.

[StephenGiles]

Here's how Yamaha used 4 x 3005 in the E1010, which I built years ago!
https://www.dropbox.com/sh/if5q3z2na5v66ga/AADL4IWxMzqA0UBXknNzeDeia?dl=0

[ElectricDruid]

Brave man, Stephen!

I used the E1010 schematic for guidance when I was messing about with the idea. The amount of "extra glue" it all needs puts me off rather.

Tom