Possible BBD score!!!

[Govmnt_Lacky]

So I was digging through some items that were marked as trash here at work (Yes... I was bored and going through someone's trash, so what of it?!) and I think I may have stumbled upon a small jackpot. I found these goodies:

Reticon R5106 (10 of them)
Reticon R5107 (12 of them)
Reticon R5108 (3 of them)

They were all stored in static prevention packaging and appear like they were NEVER used 

Any ideas on what to do with them? From what I have read and researched, they were used as SAD512D and even SAD1024 replacements 

Any input is appreciated!!!

[Scruffie]

All I know is the R5106 is used in reissue Deluxe Electric Mistresses...

[Fender3D]

MXR small flanger and small chorus...
tiny and effective...

[theehman]

MXR small flanger and small chorus...
tiny and effective...

And noisier than the SAD512.

[Mark Hammer]

RD5106 = 256 stages
RD5107 = 512 stages
RD5108 = 1024 stages

The chips only require a single-phase clock input, providing the flip-flop action internally, and offer only ONE output rather than the complementary outputs we normally expect from BBDs.  Note that the 8-pin SAD512D is almost the same as the RD5107, but where the 5107 shows N/C for pin 3, the SAD512D shows a complementary output that has to be level balanced against the other output.  Perhaps that task is accomplished less effectively inside the 5107 than it is outside the 512, leading to the noise Ron mentions.

While they can functionally replace an MN3009 (256 stages), MN3004 (512 stages), and MN3007 (1024 stages), they are NOT pin-for-pin compatible, and will need additional circuit modifications to use as a sub for an MN3xxx chip.

Still, great score.  I'm jealous.

[jdub]

Still, great score.  I'm jealous.

Me too! Congrats, man.

[Govmnt_Lacky]

Thanks to all for the inputs.

I can only ASSume that these are functional as they were packaged and protected quite well however, it would still be nice to be able to test them without building 3 seperate circuits and/or designing a daughter board to plug into an existing effect 

I guess I will hold on to these in case someone needs them for repairs 

[Govmnt_Lacky]

After another "careful" look at the parts, it appears that the quantities are different:

Reticon R5106 (3 of them)
Reticon R5107 (13 of them)
Reticon R5108 (9 of them)

[Mark Hammer]

I think you need to figure out how to adapt the A/DA Flanger to the RD5107!  512 stages is just right for that beast.

[Govmnt_Lacky]

I think you need to figure out how to adapt the A/DA Flanger to the RD5107!  512 stages is just right for that beast.

I thought the A/DA worked off of 1024 stages??

SAD1024 or MN3007. Am I wrong?

[Scruffie]

I think you need to figure out how to adapt the A/DA Flanger to the RD5107!  512 stages is just right for that beast.

I thought the A/DA worked off of 1024 stages??

SAD1024 or MN3007. Am I wrong?

When it uses the SAD1024, it works in parllell (I think?) for better fidelity so it's using 2 x 512 seperatley.

The MN3007 has the 4049 buffer to get it to go to clock up high enough to approximate the 512 stages.

[Govmnt_Lacky]

Well, unfortunately I am ashamed to say that I do not posess the knowledge OR the available time to figure out how to retrofit these in place of other BBDs.
If someone is quite smarter than I, and would like to take a run at it.... I would be willing to donate 1 or 2 pieces for the effort.

Let me know. I would definitely be interested in an A/DA adapter or even if someone has a link to an MXR or DOD pcb layout I would be most grateful 

[Earthscum]

10 if 5106=256 then goto 20
20 if 5106=single clock and single output then build a chorus

[Mark Hammer]

As Scruffie notes, the MN3010 and the SAD1024 both possess two functionally independent 512-stage sections.

Normally, a a BBD would have two complementary paths internally, that I often refer to as "tick" and "tock", since they are both clocked by alternating pulses to be doing the opposite of what the other path is doing at any given moment.  When they are mixed in perfect balance, two things happen at the output.  First, much like the two sides of a zipper, the alternating samples are stitched together to form a complete seamless depiction of the input signal.  Second, the opposite-phase clock pulses are combined in a manner that lets them cancel each other out.  It's actually not unlike how humbucker pickups work: each path gather signal you want and signal you don't want, with the wanted signal combining nicely and the unwanted signal cancelling.  Since there is no guarantee that what comes out those two outputs is perfectly matched in terms of level, some manufacturers will use a trimpot, rather than same-value resistors, to tweak it to perfect balance and maximum clock-noise cancellation.

In the case of the MN3010 and SAD1024, what I just described is done in duplicate, such that there are two tick and two tock paths running together.  In principle, this shouldn't really improve noise levels all that much, but may improve signal output levels.

What is relevant for you is that each chip is used as if it was a 512-stage BBD.  So, if you replace the MN3010 or SAD1024 with a 512 stage device, like the RD5107, the delay range achieved (important for the way it sweeps up so high) is the same, without having to change very much about the modulation and clock circuit.

The 4013 flip-flop that normally provides the tick and tock pulses only has to provide either a tick OR a tock to the RD5107 clock input.  The 5107 will provide its own tock in response to a tick.  The two complementary signal outputs, that would normally be present in each half of a 3010 or 1024, are combined internally, so that you wouldn't need to use the 4 mixing resistors you see when there are two parallel tick and two parallel tock paths to be mixed together outside the chip.

So, in many respects, using an RD5107 in an A/DA flanger is simpler than using an MN3010 or SAD1024.  The tricky part is knowing what to leave out.

[Mark Hammer]

Okay, since this thread will attract BBD-o-philes, allow me to draw your collective attention to some info that came my way today from a gracious forum member (on another forum I think), that kind of blew me away.  Any of you folks ever run into any of these babies in your travels or scan through parts bins?  Certainly if you ever find datasheets for them, let me know.  There may be an outside chance that some of these could be useful pin-for-pin replacements in pedals we all know and love.


TCA350Z 1 x 185 cells, FET, -22 V supply, TTL-compatible symmetric 250 kHz clock, DIL-8, ITT

TDA1096 2 x 256 cells, 2 x 5 to 500 kHz TTL-compatible symmetric clock, DIL-16, Philips

TDA1097 1 x 1536 cells, P-channel enhancement MOSFET, -18 V supply voltage, 5 to 100 kHz clock symmetrical, DIL-8, Philips

TDA2105 1 x 4096 cells, P-channel MOSFET tetrode, -18 V supply voltage, symmetrical clock 100 kHz, 12 kHz bandwidth, 1.2 V maximum signal, 75 dB signal / noise ratio, distortion 0.5%, 2.800 pF clock tax, special DIL-16, Philips

TDA2107 1 x 1024 cells, P-channel MOSFET tetrode, -18 V supply voltage, symmetrical clock 100 kHz, 12 kHz bandwidth, 1.5 V maximum signal, 80 dB signal / noise ratio, 0.5% distortion, 700 pF clock tax, DIL-8, Philips

TDA2108 1 x 2048 cells, P-channel MOSFET tetrode, -18 V supply voltage, symmetrical clock 100 kHz, 12 kHz bandwidth, 1.5 V maximum signal, 78 dB signal / noise ratio, distortion 0.5%, 1,400 pF clock tax, special DIL-16, Philips

TDA2110 2 x 512 cells, -15 V supply voltage, symmetrical 100 kHz clock, 12 kHz bandwidth, 1.8 V maximum signal, 0.4% distortion, 85 dB signal / noise ratio, 350 pF clock load, DIL-14, Philips

[Govmnt_Lacky]

Thanks for dropping that knowledge Mark.

I was thinking that, as far as the A/DA is concerned, it may be worth trying to swap with moosapotomas' MN3007 version. Since the MN is clocked to simulate a 512 stage chip, why couldn't it be replaced with the R5107 and clocked appropriately?
Now... I know that some parts will have to change and there may be some moving around but, is that a viable idea?

Beyond what I have written above, I would need some help with the "brainier" part of the swap  

Maybe oldschoolanalog will chime in here for some engineering assistance 

[Mark Hammer]

I was thinking that, as far as the A/DA is concerned, it may be worth trying to swap with moosapotomas' MN3007 version. Since the MN is clocked to simulate a 512 stage chip, why couldn't it be replaced with the R5107 and clocked appropriately?

Here is where I may come up to the perimeter of my knowledge.  When the tick (or tock) of the 4047 (mistakenly called a 4013 earlier) hits the clock in pin of the RD5107, does it generate a second tick at the frequency of the 4047, plus another interpolated tock, or is it divided down to generate the tick and tock, requiring the user to double the frequency of the 4047?

[Fender3D]

And noisier than the SAD512.

lol I know these are not the same BBDs as SADs...
But they're easier to manage, and I thought Govmnt_Lacky was looking for faster pedals to build, since my suggestion... those effects are just a few tracks away from SAD models...

Here is where I may come up to the perimeter of my knowledge.  When the tick (or tock) of the 4047 (mistakenly called a 4013 earlier) hits the clock in pin of the RD5107, does it generate a second tick at the frequency of the 4047, plus another interpolated tock, or is it divided down to generate the tick and tock, requiring the user to double the frequency of the 4047?

The clock will be divided by a factor of 2 (flip-flop)
But, and this is the easier management I told above, your clock may also NOT be a "pure" square wave, allowing you to build your favorite oscillator  (no more buffers needed)
Micro flanger use a 4069 pulse wave, Micro chorus a trannies astable ox.

Thanks for dropping that knowledge Mark.

I was thinking that, as far as the A/DA is concerned, it may be worth trying to swap with moosapotomas' MN3007 version. Since the MN is clocked to simulate a 512 stage chip, why couldn't it be replaced with the R5107 and clocked appropriately?
Now... I know that some parts will have to change and there may be some moving around but, is that a viable idea?

Beyond what I have written above, I would need some help with the "brainier" part of the swap  

Maybe oldschoolanalog will chime in here for some engineering assistance  

there will be no difference double clocking a double stages device in order to achieve a half stages device's delay time, provided you'll obtain the needed clock freq. But issues may arise from layout, stray capacitances, chips... (Tomas explained this very well in his 3207 EM thread)

[Scruffie]

Any of you folks ever run into any of these babies in your travels or scan through parts bins?  Certainly if you ever find datasheets for them, let me know. There may be an outside chance that some of these could be useful pin-for-pin replacements in pedals we all know and love.


TCA350Z 1 x 185 cells, FET, -22 V supply, TTL-compatible symmetric 250 kHz clock, DIL-8, ITT

I have 3 x TCA350Z Chips

[Mark Hammer]

What would you do with them?  Would you consider them serviceable for an ultrashort offset delay to yield TZF?