DIYstompboxes.com

(http://i58.tinypic.com/2jbv7sm.jpg)

The schem above was the resul of a discussion started by user stfala here: http://www.diystompboxes.com/smfforum/index.php?topic=112152.0

It give a decent single repeat in the range of 100-150 msec, but I want more: I would like to have some control on the number of the repeats but:

- Connecting the two asterisc on the schem with a sort of level controls does not work.
- I subbed the bjt with another TL072; now I have two inverting buffers, one for the delayed signal and one to (try to) feed that signal back to the bbd input. This idea does not work too  >:( : I don't get more repeats, just distortion.

I can only connect the echo signal back to the input, but this messes quite bad with the dry signal (thought the IC buffer works a little better).

Is there a right way to connect the delayed signal to the bbd without messing with the input?

Do biasing the bbd input (like in the dm2 for example) will help?

To get repeats you really need a mixer at the input to the BBD, similar to the output mixer of the second op-amp. You might get away with a passive resistor mixer there, but it's never the  best solution IMHO.
To add a level control for the repeats, you could add a 500k pot in series with the 47k out of the transistor. That would give variable attenuation of the delayed signal, but not completely turn it off.
If you use the same circuit to mix  repeats into the BBD, another pot in series with signal coming from the transistor emitter would work the same.

You have the BBD input bias at 1/2 Vcc reference (Vb). That could be around 4.5v (if you're using 9v power) and it could be that the particular BBD you have is happy with that, but they do vary on the working bias level from chip to chip. If you put a 10k trimpot in between the 2 10k divider resistors and take Vb from the wiper of the pot, you can set the bias anywhere between 3 & 6 volts to suit the particular chip.

Joining the asterisks was never going to work. The op-amp output is a very low impedance so it will just fight the transistor (which is also low impedance), but like I said earlier, resistors (4.7k? 10k?) added after each output and joined together to BBD pin7 will mix the clean and delayed signal without any fighting. But using another op-amp instead gives another opportunity for a filter cap on its feedback loop should it prove to be needed.

Your circuit is all directly DC coupled. You may just about be getting away with it, but the output op-amp has a gain of about 2. If the output of the transistor has a DC level any different from Vb, then that difference is multiplied by 2, and as it's inverted, the output is pushed down by that amount. It would be advisable to use a capacitor from the emitter to AC couple the output of the transistor which would remove this bias offset. The choice of capacitor value gives you the option (small value cap) to lower the bass response a bit, which can improve clarity when the echoes are dense. Something like 100nF or lower. The same reasoning goes for input repeat mixer if you add one.

Thank you anotherjim.
After some paint work I can show you actually where I am, sorry for the schem "quality"  ::)

(http://i62.tinypic.com/rm4lfc.jpg)

The point is, of course where to connect the question mark point...
Not before the input cap, like we said...
So, before attempting to decouple the 3205 input with caps and then find a way to bias that damned pin, you suggest to put a couple of 4,7K/10K in front of 3205's pin 7 and of the question mark, and join them togheter (of course I will put a volume pot (10K?) between the question mark and the 4,7K/10K resistor; did I understand right?

Btw, where can I buy question marks (only quality items please) in Europe? :icon_lol:

What's easier than the application circuit in datasheet?
Check it out, you may simplify the input and output filters, and use your BJT for high impedance input buffer

Good point bro.... Maybe cutting some cap and resistor. .. At the moment I have the same amount of op amps afterall.
Btw, did someone tried it?

Question for you allesz.

You never really gave your reason for exploring a minimal parts count BBD delay.

To make it as small as possible?

As an exercise?

Do you want it to sound the best that it can? or...are you willing to sacrifice low parts count/size for quality of sound.

The mixer goes AT the input of the BBD.

"Repeats" really means a potentiometer-- if everything were exactly unity, repeat would be infinite; but things are not unity and you don't want infinite repeat.

I don't believe the transistor needs that 220K.

I do agree there should be some DC blocking.

Changes marked with red slashes.

http://oi57.tinypic.com/2zzj0pu.jpg

Thanks for the help people  :icon_redface:

Of course this is more of an exercise: with the same parts count you can build a way better delay with the usual pt2399 chip; but it's just something I had to try, just for the challenge. I mean, today you can buy really good delay (digital or analog) for little money.
Then, if it useable, I will be happy: a single repeat of 100 msec is quite short, but 100 msec with two repeats is not too bad for my simple, unprofessional and nerdy diyer slapback needs.

Thanks for the input PRR, I will give it a try.... but I would like to avoid that kind of output buffer, and the bjt buffer for the delayed signal: inverting buffer (I am a total noob about them) appear to offer more filtering opportunities.... and we need a lot of filtering here. 
Tonite, instead of compulsive breadboarding, I studied the 3205 datasheet and the app schematic, thinking about Fender3d's words....

I was thinking about this signal flow: bjt input buffer (non inverting),then the signal goes to the output and to an IC (inverting) buffer (maybe with some filtering in the feedback loop), then to the 3205 (straight or with a small resistor?). The echoed signal then goes in another IC inverting buffer (heavy filtering do apply here) and from this buffer goes to the output (I could even put a mix control) and to the input of the first inverting buffer (with some kind of pot, of course) to give repeats.

Needs refinement, not sure it'll work as shown but a quick idea I just sketched out to play with;

(http://i898.photobucket.com/albums/ac182/Scruffie_Crow/delay_zpss9bod6of.png)

Has pre & de-emphasis and a sallen-key all set to 1591hZ, the mixing is done at the low pass filter.

OMG! Thank you a lot for the contribution Scruffie!

I started thinking about a BBD project since I saw your EH Full Double Traking conversion for 3205.... others are too complicated to do by myself, unless I buy a kit, of course.

I think I will have to go home, get the family out, and lock me in  :icon_lol:

No problem :)

Something like this version is probably better, the dry wont be filtered so much and it cuts a few parts out, once again it's only a quick 5 minute drawing so no promises it'll work as is.

It's a bit like an electric mistress really with a heavy sallen key on the BBD output.

(http://i898.photobucket.com/albums/ac182/Scruffie_Crow/delay%202_zpstgik0iut.png)

An humble question, since today I will have some breadboard time: does the vr network substitute tha classic two 10k resistors used to bias the IC?

Yes it does, you can have it with 2 x 10k if you want but the trimmer is better, opamps are a bit more forgiving of bias than BBDs.

I can see a few issues with that drawing now, try it first though, it should still work.

Thank you.
Today I tried the first one. No dice.
Started debuggin, tore everithing apart, then realized the battery amp I was using was amlmost dead.
Rebuild everithing. .. right now I have a working input buffer :icon_rolleyes:
Will report my progress.

Finally a little bit of success tonite  8)
I will post the schematic asap.

I upped the "time" control capacitor to 470pF, before it was shorter but not noisy, now at the longest setting it wines a little..... probably I should add some filtering; it's quite dark sounding right now...

I had to put some gain on the op amps, no more plain buffers; somehow the volume was less than unity; now, maybe, I have a little boost, but it's not too big and, btw, I don't like true bypass delays.

The schem in the end is part mine, part PRR, marteen an Scruffie: it borrows and mixes from everyone  :icon_biggrin:

I called it the echopathit because... well, you have to try it.

Of course I would be happy to improve it.... of course I'll need some help.
(http://i61.tinypic.com/f4jua9.gif)

You can ear that the echo is quite dark sounding and, as long as the repeats get longer, a sort of wistling note emerges.... it can be perfected... but it's not too bad (apart for the guitar playing :icon_redface:).

https://soundcloud.com/allesz/echopathic

Eliminated the 470k resistor and 22nf cap from the bjt buffer;  then modified the echo filters: a 10nf in place of the 3,3 nf that now goes to the emitter of the bjt, not to ground.
Then upped the 10k resistor in the time control network to 22k
Did a comparison with my digital delay: it can get aroun 300 msec, and wining only starts at around 200 msec.
I wonder if using 2 3208 would improve noise.

3208 if they're like the 3008 (CBA to check right now) have slightly better specs than the 3205... but having 2 in series will likely negate that.

200mS is what the chip is specified to give, we drag more out with lots of filtering and noise reduction, getting more than that with a single 2 pole low pass filter and a little bit of input filtering and no compander from one chip is a very tall order.

For 300ms you need a clock frequency of ~6.75kHz just for reference, quite audible...

You could try dropping the value of the 4n7 on the transistor to 1n, it'll raise the filter point to 1kHz but give it a steeper roll off... or you could make the caps 22nF & 2n2 and it'll be about the same frequency and steeper.

Thank you. I will try those values asap.
I am thinking about a layout.... no 1950a this time, but even in a 1950b it will be quite tight.

I have the small 3208s... I was thinking about swap the bjt for another tl072: in order to have one more buffer between the two bbds.

One of my first (if not possibly my first?) PCB layouts was the EHX double tracking to fit a 1590B, I think someone did eventually verify it so it should be doable.

If you're going with 2 dual opamps you may as well do input pre-emphasis > sallen key > BBD > BBD > sallen key > de-emphasis, look at the small clone for an example... that'd be a more standard way to do things.

Edit:You'll have trouble with bias though, the BBD output is ~+/-1V (can't remember which, look at the datasheet) so the second BBD will need a trimmer or to tap off the V.Ref... 2 BBDs improperly biased isn't going to be great for headroom.

Buffer after 1st BBD is not needed, better fit a null trimmer on last BBD's output pins.
I'd rather put a buffer/mixer on output, instead of just 2 resistors.

You may set up a differential amplifier @input, +in as high impedance guitar input and -in as return for repeats...

Quote from: Fender3D on October 17, 2015, 03:44:21 PM
Buffer after 1st BBD is not needed, better fit a null trimmer on last BBD's output pins.
I'd rather put a buffer/mixer on output, instead of just 2 resistors.

You may set up a differential amplifier @input, +in as high impedance guitar input and -in as return for repeats...

An inverting buffer between BBDs does offer some benefits... but probably not as much as pre/de-emphasis for noise.

Hey, the filter cap substitution was good: now I have 22n and 4,7n; it works very good, but always olny untill you get to longer time delay, at that point the faint wining is unstoppable. I say unstoppable because it bleeds through even if the echo line is disconnected; I don't knsow if it's a problem that comes from the breadboard or the layout... but it's ok to me: I never wished for something sounding this good (in a nerdy diyer sort of way, of course).

The idea of two 3208 came because, visually speaking, I would like to see five 8 pin bugs in line; of course the circuit will get a little bit more complicated... and I really feel I never did something so big and difficult.

If you look at the GGG ad3208 schematic, you see he did not use buffers, only a null trimmer, a cap, and a bias network. Madbean, on the dirtbag, use an inverting buffer, a bis network, and the null trimmer is only on the last bbd... now I don't find it, but I swear I saw a schematic where the two 3208 (or 3008, can't recall) only had a plain non op amp inverting buffer between them, no bias or capacitors.... but since i can't find it again I admit that my memory may be faulty.

About the null trimmer, I tried a 20k trim, a 50k one, then a couple of 4,7K 1% resistors... I felt no difference at all, so I went with the easier way; but I admit that not having a scope, or other instruments (apart a guitar, of course  :icon_mrgreen:), I can't appreciate the effect of that pot.   

> null trimmer..., I can't appreciate the effect of that pot.

Think how the original "bucket brigade" worked. A row of men passing buckets of water from a lake to a fire. The man at the end threw a bucket of water on the fire, then waited for the next bucket. Water went on the fire only part of the time.

The very simplest BBD would work like that. On odd beats it passed signal, on even beats it moved another bit along. You had output only half the time.

That's not good, and in this case easily fixed. The last stage is duplicated with an extra beat delay on one output.

If CMOS were "perfect", the two outputs would be identical. But CMOS devices vary a lot. One output might be only 9/10 of the other. This is better than nothing but still leaves a glitch. In effect it is clock leakage.

The earliest commercial (and preproduction) BBDs let you mix the two outputs and trim for least clock leakage. I know that most production BBDs have very nearly equal outputs, and the trimmer will be very near midpoint. I think a few later designs omit a trimmer here (perhaps the CMOS process improved so much the two outputs are equal-enough in most cases).

A small update: I am still breadboarding around the damned bbd. It's fun in the end.

I learned something about Sallen Key filters (always a nice thing) and other things like, for example, that breadboard layout is very important on big projects  :icon_rolleyes:

Also I discovered that a big filter on the power rail (470uF, just like in marteen's schematic) works wonder limiting the clock whining bleed through at longer delay settings.

But keeping the project simple makes impossible to reach longer delay settings (and we are talking about 300 msec, when I say long  :icon_lol:).

I am refining the project right now, but I will post something just a little more complicated than marteen's schem that will have a repeat control and it will be quite silent up to around 150/200 msec. Probably nobody will whant to box a thing like that, but it's a funny project to try on the breadboard.

I also made some experiments with a Vb control but the simple two 10K resistot to ground and a smoothing cap really seem to work the same way.
 

Sounds like you have an analogue slapback delay, which plenty of people may find quite appealing.

I agree, but maybe digital delays got to a point of perfection, even emulating analog echoes  :icon_lol:, that there is no point in the exercise.

And a well used pt2399 can give a wonderfull echo with not too much work.

Anyway here is the last "complicated" circuit:
(http://i57.tinypic.com/8yxd3k.gif)

It is verified on the breadboard.
Only watch for * and ** on the schematic: the pot in combination with the cap set the max time delay and, as draw, times are longer and the device quite noisy. Right now my favourite values are 50K pot (just stick a 100K resistor in parallel) and 330pF.
You can keep those values and, simply, don't use the noisy part of the pt's travel.

Actually I am testing an even simpler design that is more noisy; not bbd whining this time, but a little bit of hiss.
I decided I can live with some hiss.

Hera I post my two latest (and probably last) breadboard adventures reports on bbds.

The first is quite hissy, the second is a bit better in this regard  thanks to some advice from mictester on the other forum.

(http://i61.tinypic.com/2jchmhe.gif)

(http://i58.tinypic.com/116i8vd.gif)

I added a bias control, quite useful, but I admit that the coolaudio 3205 and the bl3208 I have biased right even with the standard couple of 10K resistor bias network.
Then I added a 500K trimpot to "tune" the time pot and avoid whining at max delay time: set the 100k pot to max, then turn down the trimmer untill it stops whining: you should get around 170/180 msec of decent delay.

You have your 'repeats' control doing double duty as wet mix as well as controlling the number of repeats. I guess there is a sense of economy there.
The input impedance is quite low. I assume this is not designed to sit at the start of your signal chain?

 :icon_lol: :icon_lol: :icon_lol: Well, the first thing to think about is: do I really need to put this "thing" in my beloved signal chain?
But I admit I did not try to put a buffer, or buffered pedal, before.

You are right samhay: if you turn the repeats to zero you have no repeats at all; this has another side effect, the level of the echo increases as you increase the number of repeats... to be perfect it should work the other way around  :icon_rolleyes:.

The echopatic sounds way better, but the max delay time is almost the same.

 :icon_redface: :icon_redface: :icon_redface:

IMPORTANT UPDATE: I draw the tiime network with a mistake: the 330pF (or 470pF) time cap has to be between pins 5 and 7 of the 3102 chip, not in series with the 22K resistor.

Sorry.

https://soundcloud.com/allesz/echopathetic

Here is the latest super simple schematic, you can liste to it in the recording I posted yesterday, and a pefr layout.

(http://i64.tinypic.com/sy780g.gif)

(http://i67.tinypic.com/2ql4rqp.gif)

Hi,
I guess I could add some discussion here...

I tried the last schematic, and it indeed is a nice short delay. Problem is the high pitch noise at longer times (I tried it with a 1M potentiometer ...)

So after some tests I came up with a small twin T design, which cuts a lot of signal, and with it the noise.

(https://i.imgur.com/zZVYXZql.gif)

I used this tool here to calculate the center frequency
http://sim.okawa-denshi.jp/en/TwinTCRtool.php

to match the frequency I was reading for max. time at 1M: 3kHz! If my math is right, that should be like.... 682ms (1/3k/2*4096)

Of course at that frequency the signal has just a slight resemblance of what it was before. I think what I like is how it fades away anyway, just like a beat that gets quieter.

I built the twin T with a potentiometer so that I could adjust the center frequency to barely match the high pitch.
What also helped was adding a balance at the output of the BBD (not in the schematic). Mine is slightly off center, where the high pitch is also slightly reduced.

The drawback of the twin T approach is that the echo got really quieter. Still something to improve.

My next step is add another TL072 to use as sallen-key filter. A previous test showed a good reduction of the noise until 6kHz without the twin T! That is time enough for a single V3205 IMHO. Let's see if it fits in a 1590a... 

Now something that I was questioning myself: is it really possible to get rid of the clock noise, or is working with higher clock frequencies the only solution, so that I would have to use more BBDs in series?

The compander won't fix the clock noise right? Or does the DM2 and similar designs use lower clock frequencies in the 5k range?

Where the heck is the bias for the BBD?

Hallo tomasha, I was lurking unlogged (as usual) and noticed your reply.

Are you still on the breadboard?

Since I did the last one, long time ago (I boxed a couple), I always wondered if elevating the power rail of the clock chip would allow more noise cancellation.
Never had the chance to try it out... .

I really don't understand why the BBD bias correctly. But this is the part I stole from another forum member (link on the first page).
If I remember right not all double ICs work.

Even a single pole filter on the input of the BBD will help greatly with the quality of the resulting output signal, just tack a cap off the 10k on the input.

The BBD output isn't low impedance, I bet it's not enjoying all that load on it, hence your loss of signal.

You can pull an absolute maximum of about 500mS (realistically 400mS and 300mS if you want decent quality) usable out of a 4096 stage BBD (I've done a second but it was not worth it) and it's absolutely possible to get rid of clock signal... just not with a design like that. If you aimed for 400mS absolute maximum you could get 2.5kHz bandwidth which isn't to be sniffed at.

Companders do help a hair with clock noise through S/N improvements but more with overall hiss. The DM-2 pulls 300mS, it's somewhere in the 7kHz clock frequency range and IIIRC the filters are about 3kHz so just a hair under nyquist.

I would look at 3rd order MFF filters rather than sallen-key if low parts is your aim and for the higher Q factor they can offer.

Quote from: Mark Hammer on June 01, 2019, 05:31:54 PM
Where the heck is the bias for the BBD?

From the op amp output, same as the zombie chorus, crappy way of doing it but apparently the 3205 biases fairly close to mid supply if this works!

Quote from: allesz on June 01, 2019, 05:32:50 PM
Hallo tomasha, I was lurking unlogged (as usual) and noticed your reply.

Are you still on the breadboard?

Since I did the last one, long time ago (I boxed a couple), I always wondered if elevating the power rail of the clock chip would allow more noise cancellation.
Never had the chance to try it out... .

The best way to do it IMO is to run the clock direct from after the diode with a ceramic or tantalum bypass cap directly on its power rail and then r/c filter for the audio.

Hi, nice to see you guys so fast back to the discussion!

I guess that you have to set the TL072 bias so that it works at the BBD desired bias.

A complication in my build, an not shown in the schematic, is that I bought a v3205SD. I never pay attention to the leters, but in this case they were important... It's the 5v version of the chip...For it to work I had to reduce the voltage below 8v.
http://coolaudio.com/docs/COOLAUDIO_V3205SD_DATASHEET.pdf (http://coolaudio.com/docs/COOLAUDIO_V3205SD_DATASHEET.pdf)

Since a voltage drop was required I used a 7805 to get 5v and power the clock and BBD at this level. I'm biasing the BBD separately with a trimmer, cap and 100k resistor.

If the clock is at higher voltages, like 9v, VGG will be off and it won't work, right? I have to test that, but I think it would at least require a special bias for VGG.

The twin-t works, but it's a heavy drop in signal right in the middle of the bandwidth, and needs to be changed for each setting of the clock.

As for the "high-pitched noise", as discussed in another thread here, consider inserting a 5k trimmer between pins 3 and 4 of the BBD, and connecting all the stuff presently connected to those pins to the wiper of that trimmer.  If the circuit has enough lowpass filtering, and/or uses a clock frequency well above the human hearing range, some pedals can afford to skip balancing the two outputs.  But you don't have much filtering, and are aiming for a longish delay time (i.e., low clock frequency) so balancing the two complementary outputs with a trimmer to cancel out the clock noise a little more is probably a good idea.

Ideally, the bias voltage for the BBD is something other than what is used for Vref in the rest of the audio path (Vb in your diagram).  If you look at the circuit for the Zombie chorus, you'll see that it uses a single Vref for the entire circuit, effectively feeding it "through" the op-amp, but it is not simply half V+ (i.e., 4.5V).  I take it that John Hollis used that as a sort of compromise that would come closer to the ideal bias for the BBD, but still provide enough headroom for the dual op-amp.
(http://www.hollis.co.uk/john/zombie.jpg)

The 320X series are advertised to work at 5V which made them appealing for battery but they can actually take 11V (absolute) max as that datasheet states. There are various pluses and minuses at different voltages depending on the BBD type which that datasheet doesn't provide, find the original Panasonic one to see all the specs.

From a quick glance, with higher voltages you get S/N improvements, less gain loss, better frequency response and headroom but at the expense of worse THD.

The MN3102 just has a simple voltage divider internally so the Vgg will adjust with supply.

Edit: But the clock & BBD must be at the same supply voltage or else you will run in to problems.

Quote from: Mark Hammer on June 01, 2019, 06:09:28 PM
Ideally, the bias voltage for the BBD is something other than what is used for Vref in the rest of the audio path (Vb in your diagram).  If you look at the circuit for the Zombie chorus, you'll see that it uses a single Vref for the entire circuit, effectively feeding it "through" the op-amp, but it is not simply half V+ (i.e., 4.5V).  I take it that John Hollis used that as a sort of compromise that would come closer to the ideal bias for the BBD, but still provide enough headroom for the dual op-amp.

The bias varies for the BBD type and for the supply voltage, the datasheet actually confirms it's not too far off half supply for this chip (a little lower for 5V, a little higher for 9V) obviously a trimmer is better as that's just an average but if it's not creating nasty distortion!

Quote from: Scruffie on June 01, 2019, 06:15:16 PM
The 320X series are advertised to work at 5V which made them appealing for battery but they can actually take 11V (absolute) max as that datasheet states. There are various pluses and minuses at different voltages depending on the BBD type which that datasheet doesn't provide, find the original Panasonic one to see all the specs.

From a quick glance, with higher voltages you get S/N improvements, less gain loss, better frequency response and headroom but at the expense of worse THD.

The MN3102 just has a simple voltage divider internally so the Vgg will adjust with supply.

Edit: But the clock & BBD must be at the same supply voltage or else you will run in to problems.

Useful to know.
Again, one needs to appreciate that the MN32xx series was developed to cope with what would have likely been a more common 9V battery supply at that time.  So the recommended 5V voltage was for the purposes of maintaining a stable reliable bias voltage, even as the battery-voltage waned from 9.6V (fresh) down to around 7V.  With few players depending on 9V batteries these days, preferring to use a wall-powered external supply, using higher supply voltages is a realistic option.

QuoteAs for the "high-pitched noise", as discussed in another thread here, consider inserting a 5k trimmer between pins 3 and 4 of the BBD, and connecting all the stuff presently connected to those pins to the wiper of that trimmer.

I tried it with a 10k trimmer and it really helps. The setting is a little off-center and the whinning is considerably reduced in comparison to center position. There is a narrow band where I could hear a drop in the noise.

QuoteThe 320X series are advertised to work at 5V which made them appealing for battery but they can actually take 11V (absolute) max as that datasheet states. There are various pluses and minuses at different voltages depending on the BBD type which that datasheet doesn't provide, find the original Panasonic one to see all the specs.

good to know, will check that.

I completely changed the circuit today. First I introduced another tl072, both sides working as 3rd order Sallen-key filters, at a frequency around 3.5kHz.
Just realized that boss also did that, but with 10k resistors. In mine I put 10k, 4.7nF, 100k, 2.2nF and 30k 330pF, kind of lossy.

The second opamp of the first IC is now the de-emphasis stage, with similar frequency response as the first stage (2.2n and 4.7k). It's also the new output (need to finish this schematic). I kept the 33k 15n low pass filter between the anti-aliasing and the de-emphasis stage, at the place of the 47k mix resistor.

What is interesting is the gain of 20 for frequencies above 10kHz, was that really the intention? Boss has a much lower gain (5 times) and a lower frequency corner of approx. 2kHz.

(https://i.imgur.com/lTrJ3Jol.png)

It looks much more complicated than previously, but it helped. I'm one compander away from the dm2 apparently, but it's at my limit of ICs for a 1590a (equivalent surface area of 5 Tl072s).

An experiment that I did was test 5v supply vs 9v supply for the v3205sd and the clock. I measured the output signal after the anti-aliasing filter, while the input was fed with a 600Hz sine wave. I had to adjust the bias for both cases, but I could get 2 times more signal out of the BBD with the 5v supply than with 9v. I used a cheap oscilloscope, and completely forgot to check the RMS amplitude at the input of the BBD...

When I tested it with my guitar the 9v configuration sounded distorted and quieter, it was also more difficult do adjust the bias of the bbd with the 20k trimpot. The "usable" range of the bias was narrower than with 5v.

Good news is that I got longer times without the clock noise, but it is sounding much darker. I'm still planning to work on the emphasis stage and check the boss values.

Quote from: thomasha on June 02, 2019, 01:30:54 PM
I tried it with a 10k trimmer and it really helps. The setting is a little off-center and the whinning is considerably reduced in comparison to center position. There is a narrow band where I could hear a drop in the noise.

Yeah, it's a pain to have another trim, but it does provide an advantage if you're using low clock frequencies. If the clock is much higher, the filters generally can deal with it.

Quote
I completely changed the circuit today. First I introduced another tl072, both sides working as 3rd order Sallen-key filters, at a frequency around 3.5kHz.
Just realized that boss also did that, but with 10k resistors. In mine I put 10k, 4.7nF, 100k, 2.2nF and 30k 330pF, kind of lossy.

I checked your values in the tool you mentioned earlier, and I'm not seeing anything lossy. Gain=1, like it should be.
http://sim.okawa-denshi.jp/en/Sallen3tool.php (http://sim.okawa-denshi.jp/en/Sallen3tool.php)
If you're losing a lot of signal through the filter, something's wrong.

Quote
An experiment that I did was test 5v supply vs 9v supply for the v3205sd and the clock. I measured the output signal after the anti-aliasing filter, while the input was fed with a 600Hz sine wave. I had to adjust the bias for both cases, but I could get 2 times more signal out of the BBD with the 5v supply than with 9v. I used a cheap oscilloscope, and completely forgot to check the RMS amplitude at the input of the BBD...

..then go and check! It doesn't make sense that you get more signal out at a lower voltage.
My own experiments with these BBDs at 5V suggested that the maximum input was around 1Vpp at that voltage, so you might be able to get as much as 1.8 or 2Vpp at 9V supply (ooh! Woo! wow! etc ;)).

Quote
When I tested it with my guitar the 9v configuration sounded distorted and quieter, it was also more difficult do adjust the bias of the bbd with the 20k trimpot. The "usable" range of the bias was narrower than with 5v.

Again, this doesn't sound right. Something else is different between the two set-ups.

Quote
Good news is that I got longer times without the clock noise, but it is sounding much darker. I'm still planning to work on the emphasis stage and check the boss values.

More filtering = less clock noise = darker. You can't reduce the high end to get rid of the signal you don't want without also reducing the high end of the signal you do want. As you say, pre-emphasis might help here, to offset the darkening effect of the filters somewhat.

HTH,
Tom

Save yourself some time and hassle, find the schematic for the 'Economy Memory Man', that pulled nearly 300mS out of 3 x SAD1024 and was about as simple as your schematic and gives you a spare op amp as you don't need the gain recovery stage between BBD's.

I agree with Scruffie: if you have to over complicate the project, maybe it's better to build something else, like a DM2 or another tested and confirmed project.

The way to keep it simple would be to avoid clock noise and then avoid a lot of filtering.

I, for example, in the end built the effect without pre and de enphasis and just a 200 pf (exactly two 100 pf caps in parallel) and a 100K time pot; you don't get wining but you also are stuck with short time repeats (around 200 ms probably), wich doesn't get more than slapback.

I also discovered that if the max avaliable repeat time is short you can just use two repeats instead of just one (it is not the same of course, but it gets the job done). Also I found I really nice effect that is not easy to obtain with a standard analog echo: I like to put a lot of repeats (just on the verge of feedback) with a really short time (difficult to obtain with commercial units), the effect is a sort of unmodulated metallic flanger that I really like.

In fact, of the two I boxed, one is in a 1950A (but it use a 3208 instead of the bigger 3005), that I use only for the described effect; the actual delay can be called a doubling effect.

Another simple work around would be to use two 3205 for longer delays (longer... around 400 ms :-)), but the clock IC can drive only one 3205....

Sorry, I forgot to add that my chips were also the low voltage type. In fact I put in the power supply a protection diode and a 470 Ohm resistor in series, they both keep the B+ a hair under the maximum 9V supply for the BBD.

So I agree with Electric Druid, is should work good also at 9V.

Btw, could a voltage regulator  be a way to separate the audio power line from the clock line?

Quote from: allesz on June 03, 2019, 05:16:57 AM
Btw, could a voltage regulator  be a way to separate the audio power line from the clock line?

Yes, it would probably help separate the two power supplies. But that assumes that the shared power supply is the reason you're hearing clock whine. It probably isn't, especially if you're using low frequency clocks. At low frequencies, the losses through the delay line are worse and noise (including clock noise) goes up. And since you have the clock frequency within the audio bandwidth, you'll hear it. There's really no getting away from that. You can have "simple", or "long", or "quiet", but you can't have all three. There's a reason so many classic analog delays used pretty much every trick in the book - it was the only way to make them acceptably quiet.

If I were you, I'd pick two out of the three I mentioned and focus on getting those optimised.

Well, if the noise is coming also from the audio path, I think you are totally right. And I choose simple and quiet.

When you hook up a BBD, even an humble 3208, and you get incredibly long delays... yes terribly noisy, but incredibly long, you get mad: I couldn't admit that it was impossible to get rid (in an easy way also!  :icon_rolleyes:)  of the whining. I did my (modest) best, and gave up.

But I don't give up to hope that someone will find a compact and clever solution, possibly misusing and abusing some component, to the problem.

Quote..then go and check! It doesn't make sense that you get more signal out at a lower voltage.
My own experiments with these BBDs at 5V suggested that the maximum input was around 1Vpp at that voltage, so you might be able to get as much as 1.8 or 2Vpp at 9V supply (ooh! Woo! wow! etc ;)).

I'm testing it again, it is probably a problem with my breadboard. I have some caps that like to jump out of place, wouldn't be the first time...

QuoteSave yourself some time and hassle, find the schematic for the 'Economy Memory Man', that pulled nearly 300mS out of 3 x SAD1024 and was about as simple as your schematic and gives you a spare op amp as you don't need the gain recovery stage between BBD's.

Never had seen those! Thanks, that is exactly what I'm looking for at the moment.

QuoteI did my (modest) best, and gave up.

It is way better than you think. After making it more complicated and achieving similar results I think I'm going one step back and try a smaller version too. I really want to try the twin-T filter again. I'm having lots of fun by experimenting and annoying my wife with high pitched noises  :icon_lol:

Thanks, I don't want to dismiss the thing, by any means.
Being totally uneducated in electronics, I can build effect, tube amps, even an analog delay! Thanks to all the great diyers and DIY forum(s).


My maximum values with no whining is 200pf and 100k for the time network. How far can you go with the 1 Meg pot?

QuoteHow far can you go with the 1 Meg pot?

Haven't measured the pot, but the clock was at 5.7kHz when there was no audible whine.

At the moment it is kind of too dark, like a blanquet over the amp. But I want to remove the extra stages and start from your initial idea, just cutting a little more signal at the second stage. Your schematic sounds much clearer.

I also tested the supply of the chip again. Using a sine wave, 500hz, 0.38v at the cap in front of the BBD (I'm setting the bias with the trimmer).
I adjusted it for max. output and measured after the balance trimmer. With the max. signal coming out I changed the clock speed to have the same 0.38v at the output, which resulted in a clock rate of 4.47kHz for 9V (8.82).

When using 5V  at same clock (had to adjust speed), I had to adjust the bias trimmer again and got for max. output 0.44v at the output.

When I increase the clock rate the signal at the output also increases, so I guess I should have done the measurements at usable clock rates. But the pedal is so dark now that even at 3kHz the whine is not bothering me. The problem is that at this rate the wet signal has aliasing noises (crispy sound).

It looks to me that the chip likes 5v better, at least at this clock rate and input signal.

Now the balance trimmer at the ouput made a huge difference! You should give it a try. It will extend your delay for a couple of ms.

PS: I checked different clock frequencies and calculated the gain of the BBD for 9v and 5v supply (0.37v at the input of the BBD, 500Hz sine)

(https://i.imgur.com/Hfu25UVm.png)

The most interesting idea for me is the use of a narrow notch filter to remove the clock whine. Now, even that's going to have limitations no matter how deep the notch is, because the clock whine isn't a sine wave (and therefore is not just a single frequency) but it might well make more difference than anything else.

Bonus points if you can do it using some switched capacitor filter chip so that it tracks the clock frequency automatically.

Oh, hang on, bang goes "simple" again...

Isn't clock whine simply the (one) clock frequency? The sum of of the two out of phase clock signals?
If you set the balance trimmer exactly 'right', there should be no clock whine at the output of the bbd. Especially if you have only one bbd.
That one trimmer, when set correctly once , should solve it for any chosen clock frequency/delay time.

When I watch the bbd output without input signal I see one clock (whine) frequency on the scope and when turning the balance trim, I can pretty much remove it completely. It might still show up on the scope a little bit but at levels that are absolutely inaudible.

Aliasing distortion / heterodyning / sum and difference harmonics or whatever you want to call it: the 'distortion' that results from input signal + clock frequency can never be dialed out.
On a scope you can clearly see the 'harmonics' generated from input signal + clock frequency.
For example a clock frequency of 10kHz and a sine wave input of 1500Hz generates spikes at:

11.5KHz, 13kHz, 14.5kHz etc but those don't really matter much because a typical guitar speaker rolls off after 5kHz or so.
The ones that matter are: 8.5kHz, 7kHz, 5.5kHz, 4kHz etc. because those can be heard more clearly.
The higher order harmonics are lower in amplitude though.

The typical solution to minimise those are:

1) anti-aliasing filter: remove high frequencies from input.

2) use a clock frequency that is high enough (>  2 x max input frequency)
If your lowest clock freq is 10kHz, don't let anything over 4,9kHz in.
The first difference harmonic will be 10kHz - 4,9kHz= 5,1kHz.
You can in theory filter out anything above your highest input frequency of 4,9kHz.
This way you can remove all clock and still preserve all input. In theory, because no filter will be steep enough.
Or vice versa: if your max input signal frequency is 4,9kHz, don't let the clock come down below 10kHz.
This is just taking into account the first (and strongest) harmonic, the second and third etc will also be present and to avoid those you'll have to reduce the input range even further or increase the clock frequency.

3) LP filter at the output of the bbd, filtering out anything above the highest input frequency.

4)use more bbd's (although you'll run into problems with the maximum capacitive load for the clock)

As said earlier, the compander only solves the white noise:
1) compressor makes the input signal twice as loud
2) signal goes through bbd, constant level white noise (regardless of the level of the input signal) gets added
3) expander makes everything half as loud. The input signal and echo are back at the original level, the white noise that got added along the way is cut in half.

Well, I ended up abandoning my schematic...it was too dark.
The echopathetic is still brighter with some extra filtering. The version with the sallen-key filter after the BBD sounded better, with the additional Twin-T I could increase the pot to 150k. But I still had to try the economic memory man.

It was not as long as I expected, quite like the echopathetic. After hearing some videos of the other versions that I could find I ended up using the schematic of the memory man with the chorus switch:
(https://sites.google.com/a/davidmorrin.com/www/_/rsrc/1424189195137/home/trouble/troubleeffects/electro-harmonix-memory-man/eh-7810/EH%207810%20schem.gif)

I just used the normal clock because I'm not interested in the chorus function. Instead of the front buffer/booster I used this opamp to mix both signals and recover some of the signal lost in the filters, it was slightly quieter than when the pedal was bypassed. Even with unity gain filters it sounds quieter, could be the gain of the BBD at longer times or just the impression caused by less high end.

Well, for it to work I also changed the input to the inverting side, so that it's not grounded anymore and the non-inverting imput is connected to the  4.5v divider. The only adjustment required to run it with a single supply.

But now I'm not sure if I still can add the feedback at the same node (now the imput of the circuit).
What do you think? Can the feedback be connected to the input? The delayed signal will be filtered multiple times anyway, but what if someone uses a fuzz face in front of it?

Quote from: thomasha on June 14, 2019, 12:54:46 PM

Instead of the front buffer/booster I used this opamp to mix both signals and recover some of the signal lost in the filters, it was slightly quieter than when the pedal was bypassed. Even with unity gain filters it sounds quieter, could be the gain of the BBD at longer times or just the impression caused by less high end.

Well, for it to work I also changed the input to the inverting side, so that it's not grounded anymore and the non-inverting input is connected to the  4.5v divider. The only adjustment required to run it with a single supply.

Sorry, I don't understand this part. Why use the op-amp A1 to mix signals when op-amp A2 already does that? And why mess about with A1 to get gain when it already does that too? (Boost input)

Good point, but I thought boosting after the BBD would be better?

I though boosting at the input would only result in more distortion from the BBD, and the V3205 has a lower headroom than the MN3005, so If I hit it too hard it distorts with some clock noise over the notes (only when strumming, not a constant noise). I thought boosting afterwards would reduce that, because I boost the first repeat, which gets more audible, while the second repeat comes through the feedback loop way quieter, not presenting this distortion. Maybe it worked better with the MN3005 and the +-15V supply.

Instead of using the dry signal after A1, which I removed, I use the signal after A2, still dry, but already filtered, like in the schematic below:
(https://i.imgur.com/xazfODul.png)

So that the blend occurs at the end of the circuit. I even put a high pass filter, but I'm not sure if it really makes a big difference.

   

The way you have your FB and mix set up now, this is what happens:
A repeat goes through the mix pot and arrives at the input of IC2A (the final output opamp)
The same repeat goes through the feedback pot and trough the input of the circuit, it also arrives at the input of IC2A but inverted! because of IC1A.

So, you're mixing/summing two versions of the same repeat, 180° out of phase, at the input of IC2A.
They will pretty much cancel each other out. Because one version is louder than the other it's not a 100% cancellation, but still not a good set up imo.

Well spotted!

That explains a lot why the wet signal was not as loud! When I adapted the schematic I totally forgot to check the phase. That was dumb!


In this case I need the first stage to be non-inverting or add the feedback later.

Anyway, I tried the original circuit and it sounded better, even with the boost option. Now I'll have to correct that feedback loop.

(https://i.imgur.com/b7WtFtAl.png)

I've sometimes swapped Multiple feedback filters in a design for Sallen Key filters (or vice versa) to flip the phase. MFBs are inverting, SKs aren't, if I remember it the right way around. Anyway, it can sometimes help save an op-amp.

T.