DIYstompboxes.com

Hi,

I found myself with a lot of free time these days so I decided to go digital and try to design a PCB for the Femtoverb project based on Wavefrontsemi's chips. Below is a preliminary version of the PCB layout. Please check it out and let me know if you have any remarks so I can fix it before posting the final version.

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb.gif)

At first glance I do not think it will work unless you make it a double-sided board...  right now,  the surface mount chips look like they have to go on the bottom of the board which would put pin 1 in the wrong place.

regards, Jack

....but if you can iron that out...lordy mama...thanks!!

Thank you for the input.

Quotethe surface mount chips look like they have to go on the bottom of the board

Yes, the AL chips will be soldered on the bottom /solder/ side. The view is from solder side anyway, that's why opamps' pin 1 seems to be on the wrong side.
Any other comments are welcome.
And I'm definately not going to iron this...

Looks like a nice project. Looking forward to hearing more!!

Couple of fixes and mods were suggested by a well known forum member/moderator. I'll post the final version after everything is settled.

It's difficult to say from the picture, but the pads on the SMD devices seem like they are too short. If they don't extend a bit more than where the pins on the SMD package end, looking from above, you might get trouble when soldering. This is because you'd like to be able to heat both the pad and the pin simultaneously. The only place where you can put the tip of the soldering iron is on the pin itself if the pad is too small. The heat from the tip must then be conducted through the pin to the pad in order to heat it sufficently for the solder to adhere. Maybe not the correct technical english terms, but do you get my point?

This shouldn't be the case if the footprint is from a library in your PCB design software. If you draw it yourself, this should be considered in order to make the soldering easier.

The SMD footprint is from my software library. I checked the data sheets for dimensions and  I took some mesurements with mm ruler. It looks like there's enough space for soldering but I edited the library package and extended the pads a little bit more. You can print the picture at 300dpi to get it down to real size. There's still some space for extension if necessary.


(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb.gif)

I scaled and printed out the original last night and placed hips over top of it.  Spacing looks fine.

If you have an acompanying traces-only image, that'd be nice to have.  Otherwise, I spent some time digitally faking one last night that I'd be happy to post.  Your call.

I have traces only picture and I'm not posting it because I wanted with the help of those of you who are more into these things to agree on a layout. If you don't have any other remarks/considerations I will post the layout /traces only/ in the next couple of hours.
I just noticed that for some reason the newer fixed layout doesn't seem to show up and both pictures are in fact the same picture.
I hope everything is OK with this one:

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_fixed.gif)

That looks GREAT! 8) 8) 8) 8) 8)

I'm always impressed by layouts like this because I just default to two layers.

One tiny nit-pick and I do stress tiny, would be to move those extra pads between the ADC input coupling cap + leads and the shunt cap, to the traces that connect the - side of the coupling caps and resistors. I suppose there isn'y any reason why they wouldn't work 99.99% as well where you have them though. Count this change as sprinkled candies on top of the icing on the cake.

Again, FANTASTIC job! 8)

OK, below is the final version /for now.../ of my Femtoverb layout. There are still couple of more things that can be done but it's enough for a start.
The PCB is mirrored and ready to go now. The AL chips are soldered directly to the traces/pads  /bottom/. To go down to real size print at 500dpi.
Thanks for all your input  and  help. Peter thank you for the good words, the encouragement and all your suggestions !
If you feel like something can be improved or moved around just let me know.




(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_parts_300dpi.gif)


(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_solder_500dpi.gif)

Whoops... I see one more thing....

You didn't sign your artwork. ;)

Now how difficult is it to etch this board with PnP? If it's pretty easy (I don't see why it shouldn't be), this will probably be the first DSP project that a LOT of people build. 8)

Anyone building this, please post a build report. :D

I'm not ironing this and that's for sure. At least not this time. A friend of mine will make it for me at work.
As for signing its well documented where it came from...
I don't know what's going on at Wavefrontsemi but more than a week ago  I asked them how to order 2 pcs from each chip  and all I got so far is only one email  asking me whether I'm a company.
Peter, any ideas?

Try posting on the Wavefront Semi forum and you should get a sales person fairly quickly. I know I've seen them post contact info there in the past.

I hope Small Bear will consider carrying these chips.

Quote from: Mark Hammer
I scaled and printed out the original last night and placed hips over top of it.  Spacing looks fine.

So it's settled then. :icon_wink:

Quote from: bassmeister on February 23, 2006, 09:18:15 AM

Quote from: Mark Hammer
I scaled and printed out the original last night and placed hips over top of it.  Spacing looks fine.

So it's settled then. :icon_wink:

Jeez I hate those small fonts where you think you see all of the word but you really don't. :icon_rolleyes:

I'm gonna print me up a PnP sheet tonight with the mask (6 copies, actually) and see how it goes.  I had done some initial exploration preparing an SMD-to-DIP adaptor, and that seemed to work out nicely, so I suspect this will work out just as well.

Impatiently awaiting the first build report!

Impatiently waiting to build! (I got my chips about a year and a half ago!)

Well, the board etched beautifully with PnP.  It's fluxed and tinned, and awaiting drilling.

The thing with a composite (SMD plus through-hole) board like this is that one doesn't want to risk the chips suffering static damage by installing them first and then lumbering about with passive components.  At the same time, even with great end-nippers it gets hard to get in there with the iron tip and reflow the solder on the pads quickly and efficiently after you've installed all the passive stuff.

Quote from: Mark HammerWell, the board etched beautifully with PnP.  It's fluxed and tinned, and awaiting drilling.

8) 8) 8)

One of the last issues to deal with in using these DRE chips is the effects switching. RJ (another user here) and I came up with a neat way to switch effects as long as you don't need more than 12 of the 16 built-in algorithms.

The DRE has four lines for selecting the program in use. These lines are all pulled high internally by the DRE so without anything connected you get program 16 (or 1111 in binary). Selecting other programs involves pulling from one to four lines low to create a different binary number. The internal programs are arranged in such a way that an encoder that uses "Gray Code" will nicely switch between the options in a sensible sequence. Gray Code was an idea that made it easier to build multi-bit encoders, but the output  order is a long way from 1,2,3,4.  You can use a 4 bit binary encoder, but the effects order will seem like a jumble.

The really good news here is that a simple 12PST rotary switch can accomplish the task with the help of a bunch of diodes. First ground the common connection of the switch. Now from each of the 12 poles, you can connect a couple of diodes to represent the zeros for that selection.

Example:  Say you wanted to select program 6. In binary you represent the decimal number 6 as 0110. To generate this number for switch position 1, you would connect the 1st pole to the cathodes of two 1N914 diodes. Now connect the anode of one diode to the binary 8 input and connect the anode of the other to the binary 1 input of the DRE. The diodes allow the lines you want to be pulled to ground when the switch is in that position, but they don't interfere when the switch is in another position. This is nothing more than an emulation of a 12x4 bit ROM with mechanical addressing.  :icon_biggrin: You can control the order of the effects, it's cheap, it uses easily available parts, and it's not too hard to select three programs to eliminate along with the "vocal cancel" karaoke program. The idea of a rotary speaker simulator with no LFO control is of limited utility too.

The encoders listed on Femtoverb's schematic are back in stock at www.alliedelec.com !

http://www.alliedelec.com/Search/SearchResults.asp?SearchQuery=753-0115&Submit=find+it%21

Here's a layout with component values:

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_layout.gif)

Thanks!!  That's going to speed things up a lot.

i feel like a dolt for asking--everyone here seems to be in the loop--but is there a working link to a schem?  the original seems to be nonfunctional.  thanks, gm

edit:  faster than a speeding bullet--there's some of the info i wanted, two posts up. :-)  still a schem would be nice.

If you look at the original Femto-Verb post (on this sub-forum) by DaveTV, you get a link to the schematic and more discussion.

http://dstockett.50megs.com/femtoverb.gif

:icon_biggrin:

Quote from: Peter Snowberg on February 28, 2006, 08:37:39 PM
If you look at the original Femto-Verb post (on this sub-forum) by DaveTV, you get a link to the schematic and more discussion.

http://dstockett.50megs.com/femtoverb.gif

:icon_biggrin:

yes, i tried that!   :icon_lol: i don't get the image.  i get a "remote linking forbidden" image.  :(

Hmmmm...

It works fine from here. I don't know.

I put a copy here:

http://img.photobucket.com/albums/v280/Snowberg/diy/femtoverb.gif

Quote from: Peter Snowberg on February 28, 2006, 09:09:46 PM
Hmmmm...

It works fine from here. I don't know.

I put a copy here:

http://img.photobucket.com/albums/v280/Snowberg/diy/femtoverb.gif

thanks!  it works at home, too.  must be something about my employer's setup.  >:(

Pretty much all populated now, though no functionality yet.  That blasted 12mhz crystal is around here somewhere!!

A word of advice to first time SMD solderers.  Solder the Wavefront chips on before you start crowding them with the through hole components.  There are a number of passive components whose leads, no matter how neatly trimmed, make it diffcult to get your soldering tip in a position to flow the solder on the pad without providing dangerous heat levels to the chips.

And get yourself some x3.5 reading glasses from the bargain store.  Those pins and pads are small!!

has anyone scored the chips for this lately?  i wrote to wavefront sales but got no response.  i see in other posts that this has happened before.

Quote from: gaussmarkov on March 16, 2006, 05:20:57 PM
has anyone scored the chips for this lately?  i wrote to wavefront sales but got no response.  i see in other posts that this has happened before.

yeah, they don't respond to e-mail that well.

I ended up calling, and got Randy Yorsten on the phone(really nice guy BTW). the next day I had my e-mail invoice.

I decided to get 2 full sets of chips since I'm going to try building the Axoris Miss Parker project, and they charged me for less than half of what I ordered. 8)

nice folks!

Quote from: idiot savant on March 17, 2006, 12:09:35 AM

Quote from: gaussmarkov on March 16, 2006, 05:20:57 PM
has anyone scored the chips for this lately?  i wrote to wavefront sales but got no response.  i see in other posts that this has happened before.

yeah, they don't respond to e-mail that well.

I ended up calling, and got Randy Yorsten on the phone(really nice guy BTW). the next day I had my e-mail invoice.

I decided to get 2 full sets of chips since I'm going to try building the Axoris Miss Parker project, and they charged me for less than half of what I ordered. 8)

nice folks!

excellent!  thanks for the feedback.  i will give them a ring tomorrow.  and thanks also for the axoris miss parker reference!

It took me almost 3 weeks and couple of angry emails to get them finalize my order. Don't rely on emails - grab the phone and talk to them!

Well, I finally found the 12mhz crystal I had been searching for, stuck in the last few passive components, and finished up my board last night.  Wired it up, fired it up and.........nothing.  The clean analog side works just fine, but I'm not getting anything out of the digital side except for some digital noise.  I hope to goodness that isn't a sign of a blown chip, because this is not somethng I'm going to know how to troubleshoot.
I'm going to give the solder joints the once over today with good lighting and a magnifying glass.  The pads are so close together that the flux tends to bridge the gapsbetween solder pads, and the glint of the flux bridges makes it hard to see if there is solder in there too.

I waswishing I'd have sweet reverb to enjoy and good news to report, but not yet.

Incidentally, there is an extra op-amp stage in the layout that I didn't recall seeing any comment about in this thread.  As near as I can tell from the layout shown, it is simply a hiss-reducing LPF section to clean up the digital path a bit, right?  The other thing not made abundantly clear is that the two pads for "Vol" are the pads for the gain-setting feedback resistor in the output stage, whih can be a pot OR a fixed resistor.

I see a missing and rather vital cap... There needs to be a small electrolytic (anything from 1uF up to 22uF, shoot for 10uF) connecting pin 12 of the DRE to ground. That cap provides the reset pulse to get things running so it's not surprising you still don't have sound. The good news there is that the pads required already exist, it's just a matter of adding the cap between the diode going to pin 12 and the ground jumper going to the 10uF next to it.

Once that one is installed if you still don't have sound, verify (with a scope or high impedance logic probe) that the wordclock is present on pin 9 of the DRE. That will tell you that the clocks are running. If you have clocks but no sound, next check for digital data on pin 16 of the DRE and then pin 1.

To check the analog input sections, you can audio probe right up to pins 1 and 2 of the AL1101 ADC. If you have signal and clocks so far, audio probe pins 1 and 2 of the AL1201 DAC. You should be able to get the wet signal from pin 1 of the NE5532 closest to the AL1201.

You're almost there!  :icon_biggrin:

thanks for the progress report, Mark!  i've been listening for your reverb.  :icon_biggrin:

Quote from: Peter Snowberg on March 20, 2006, 11:47:40 AM
Once that one is installed if you still don't have sound, verify (with a scope or high impedance logic probe) that the wordclock is present on pin 9 of the DRE. That will tell you that the clocks are running. If you have clocks but no sound, next check for digital data on pin 16 of the DRE and then pin 1.

argh!  my new dmm doesn't have a "high impedance logic probe."  are there any other dmm features that i need for digital stuff?

Thanks for that tip, Peter.  I'm hoping that's all it is.

Pins 15 and 16 of the AL1201 also seem to me to be just hanging there, and not tied to ground as they are in the Femtoverb schematic.  I linked them and ran a line to ground but perhaps there is some other deeper reason why those connections were not included in Metal Guy's layout?

Hi,

Sorry to hear it didn't work from the first time. The preliminary layout was out long enough to be checked thoroughly and corrected. If anyone finds any other flaws please let me know so I can fix it.

QuotePins 15 and 16 of the AL1201 also seem to me to be just hanging there

My schematic /version 1.5/ shows they are not connected to anything. In a PM Peter suggested to put pads there - he can explain better why.

QuoteI see a missing and rather vital cap...

Actually the cap is there but is connected to  ground and +5V. Should be pin 12 and ground.
Below is a layout with the pin 12 cap fixed:

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_layout_2_300dpi.gif)

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_solder_2_500dpi.gif)

High speed digital logic is beyond what a DMM wants to deal with but you can construct a very simple logic probe very easily.

Here is the recipe for a simple logic probe:

- Grab a 4049 or 4069 inverter. Power that from the same V+ as the circuit under test
- attach a 10M resistor from V+ to the input of one of the inverters. This is your input inverter.
- attach that inverter's output to the input of the next inverter and also to a RED LED with a 1K resistor to ground.
- attach the 2nd inverter's output to a similar red LED+1K to ground.

When there is no input, the 10M will pull that input high causing a low output on the 1st inverter and a high out on the 2nd one. The second LED will light and that shows the input is either open or connected to a high level input. If you touch the input to ground the 1st inverter will now output high and the 2nd one low, causing the 1st LED to light and the second one to go out.

If you touch the 1st inverter output to a clock, both LEDs will light up equally and both be dim.


Mark, do you mean pins 15 and 16 of the AL1101 or the AL1201? Neither chip should have anything connected directly to ground as that may let the blue smoke out. 15 & 16 of the AL1201 should float; they're outputs and as the built-in algorithms are stereo, they also output signal from the other input channel. The pads are there to allow for experiments in mixing the channels. The AL1101 has the 2nd input channel coupled to ground via a pair of 10uF caps to silence them. They're internally biased to 1/2Vdd and should not get grounded directly. There are a couple of extra pads on the utilized input channel to allow for the effective bridging of the used and unused inputs.

Quotedo you mean pins 14 and 15 of the AL1101 or the AL1201?

Peter, do you mean pins 15 and 16?

QuotePeter, do you mean pins 15 and 16?

:icon_redface: Yep, sorry. Make that 15 and 16.

Hey Mark, did you get the thing running?

This is my first time back at the bench in a while, so no, not yet.

would someone please check me on this?  i think there is a problem with the version 2 layout.  to find it, note that on the schematic (http://dstockett.50megs.com/femtoverb.gif (http://dstockett.50megs.com/femtoverb.gif)) there are a pair of 10uF cap/200 ohm resistor combinations between IC2 and pins 1 and 2 of the AL1101.  and there's a 4700p cap across those pins.  on the layout the 10uF caps hook up directly to the 4700p cap instead of having the 220 ohm resistors in between.  it's like the caps and resistors are switched.

edit:  or does that even matter?  i guess order is not important there.  :icon_redface:

Quoteit's like the caps and resistors are switched

Yes, they are.
This was suggested by Peter Snowberg with some mods in mind. Maybe he can give you some more details.

I'm almost done populating my PCB and next week I hope to be able to power it up.

Quote from: MetalGuy on May 19, 2006, 04:23:08 PM

Quoteit's like the caps and resistors are switched

Yes, they are.
This was suggested by Peter Snowberg with some mods in mind. Maybe he can give you some more details.

I'm almost done populating my PCB and next week I hope to be able to power it up.

cool and thanks for the reply.  fwiw, i am going over the layout carefully and i hope it's helpful.  i put the schematic into eagle and now i am laying out your layout again, using the rubberbanding and the design rule checker in eagle to confirm everything.  i haven't started to build it yet because i still have a few components (crystal and regulator) and some pnp blue to get.

i hope yours works first time!  :icon_biggrin:

metalguy,

i just finished tracing your layout and here are some additional questions.

• i wonder whether the 78L05 is backwards, with IN and OUT switched?  if i read the layout right, OUT is currently connected to +9v and IN is connected to +5v.  my (inexperienced) reading of the datasheet says they should be the other way around.
• the 2M2 pulldown resistor on the OUT looks to me like it is connected to Vr when really it goes to GND, one hole over
• two 1uF caps in the schem are 10uF in the layout:  the dc blocking cap on the OUT and the supply filter cap on the Vr supply
• is the +9v supply missing the 100uF filter cap and the 1N4002 polarity protection diode?

everything else matches the schematic as far as i can tell.  i put the schematic into eagle and then built up your layout from there--so there's a pretty good chance there's nothing else different.  i guess the cap values aren't critical, nor the polarity protection and the 100uF cap.  maybe peter changed those, too?  and i admit that i do not know how regulators are supposed to be hooked up.  so maybe it's all good.  :icon_biggrin:

--gm

Quotei wonder whether the 78L05 is backwards, with IN and OUT switched?

You're right -  it's backwards.

Quotethe 2M2 pulldown resistor on the OUT looks to me like it is connected to Vr when really it goes to GND, one hole over

You got me one more time!

Quotetwo 1uF caps in the schem are 10uF in the layout:  the dc blocking cap on the OUT and the supply filter cap on the Vr supply

I don't think the values are critical but I'll correct them.

Quoteis the +9v supply missing the 100uF filter cap and the 1N4002 polarity protection diode?

They are omitted intentionally - to save space and because I usually mount them directly on the DC jack.

Thanks for noticing and informing me for these details on time.
Below is the corrected layout:

(http://i7.photobucket.com/albums/y263/GreggPics/Femtoverb_parts_300dpi_corrected.gif)

metalguy, thanks for the new layout and the additional comments.  i like to move stuff like the power supply filter off the board sometimes, too. :icon_biggrin:

everyone, i have a different sort of question about the circuit design itself.   i hope someone will answer.  are some of the caps on the schematic by davetv supposed to be physically close to the pins of the ICs?  i remember puretube making a comment about this sort of thing in another thread about R.G.'s variable stutter circuit. 

if not, then what's the deal with all the .1uF caps across the +5V and GND rails?  i am thinking of the ones across pins 12 & 13 of the AL1201, pins 14 & 15 of the AL3201B, pins 12 & 13 and 6 & 7 of the AL1101 (because digital ground and GND seem to be the same thing).  you could also throw into this list the cap on pin 6 of the AL1201.  that could just as well be across pins 5 & 6.

if these caps should be physically close to these pins, then how would anyone know this just from looking at the schematic?  and how close should the pcb designer put them?  how should other neighboring components be treated?

thanks in advance, gm

you can find IC-sockets with built-in decoupling caps - that`s physically close as it can be!  :icon_wink:

Quote from: TELEFUNKON on May 21, 2006, 05:26:30 PM
you can find IC-sockets with built-in decoupling caps - that`s physically close as it can be!  :icon_wink:

ah, a clue:  "decoupling caps."   :icon_wink:  thanks TELEFUNKON!  for those who were unsure (like me):

• Re: Amp-op decoupling and unused XOR gates by R.G. (http://www.diystompboxes.com/smfforum/index.php?topic=40639.msg292297#msg292297)
• Coupling vs. Decoupling Capacitors by R.G. (http://www.diystompboxes.com/smfforum/index.php?topic=20425.msg125783#msg125783)
• Understanding 'decoupling' by R.G. (http://www.diystompboxes.com/smfforum/index.php?topic=34781.msg244806#msg244806)

there are probably many other threads.  these come up first in a search and seem to cover the ground.  check out the related comments by niftydog and puretube. :icon_biggrin:

I got mine working from the first time! Populated the board, checked everything several times, connected it to the power supply and heard the slapback echo.
I didin't have time to experiment too much but all programs are working it sounds good so far. I'll post some pictures later today.

CONGRADZ!! :icon_biggrin:

this is very good news!   :icon_biggrin:  excellent.   :icon_cool:  i still have to order that crystal and the regulator ...

how are you going to box it?

About boxing I'm thinking a standard 1590 box but I'm also thinking about  adding a PIC control unit which will be able to memorize 3 settings for my 3 channel amp. Because this will complicate things maybe I'll leave this one as standalone FX unit and make another one for my amp. Unfortunatly I didn't have time to play with it too much but this weekend I'll take some more time to experiment. Below are some picures of the unit:

(http://i7.photobucket.com/albums/y263/GreggPics/FemtoverbParts.jpg)

(http://i7.photobucket.com/albums/y263/GreggPics/FemtoverbSolder.jpg)

(http://i7.photobucket.com/albums/y263/GreggPics/FemtoverbXmasTree.jpg)

beautiful!  thanks for posting the pics! :icon_cool: :icon_cool:

 :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool:
:icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool:
:icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool:
:icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool:
:icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool: :icon_cool:

Today I spent couple of hours playing with the unit and noticed that the effects switcing order is somewhat different from what it should be. Looking at the schematic I'm assuming that I'm facing the encoder /the shaft side/ - that's how I connected it. The effects order goes like this /backwards/: Rotary - Vocal - Chorus Room /Wah/ 2 - Chorus Room 1 - Delay 2 - Delay 1. So far so good. The following should be Flanger  - Chorus but in fact three rooms/halls/plates follow then Flanger - Chorus and again four rooms/halls/plates. I'm using the same encoder as from the schematic.

Any ideas what could be wrong?

encoder "pinout"?

The encoder is connected as per schematic.
I spent couple of hours staring at the encoder's output table:

http://catalog.sensing.honeywell.com/printfriendly.asp?FAM=encoders&PN=510E1A48F416PC

Also I found this:

0110 Hall 1 Bright hall reverb for drums, guitars, and vocals.
0010 Hall 2 Warm hall for acoustic guitars, pianos, and vocals.
1010 Room 1 Hardwood studio for acoustic instruments.
1110 Room 2 Ambience for acoustic mixes and synth sounds.
1111 Room 3 Warm room for guitars and rhythm instruments.
1011 Plate 1 Classic plate reverb for lead vocals and instruments.
1001 Plate 2 Sizzling bright plate reverb for vocals and drums.
1101 Plate 3 Short vintage plate reverb for snares and guitars.
1100 Chorus Stereo chorus for guitars and pianos.
1000 Flange Stereo flanger for jet wash effects.
0000 Delay 1 125ms slapback delay for vocals and guitars.
0100 Delay 2 190ms delay for percussive arpeggios.
0101 Chorus/Room 1 Chorus with reverb for guitars, synths, and pianos.
0001 Chorus/Room 2 Auto-wah guitar effect with reverb for lead instruments.
0011 Vocal Cancel Removes lead vocals from many stereo recordings.
0111 Rotary Speaker Rotary speaker emulation for organs and guitars.

These codes are valid only if you count encoder's pins /from the output table/ as follows : 8 - 1 - 4 - 2  /physically 5 - 1 - 2 - 3/. This exactly according to the schematic.
On the other side AL3201 pins 5, 6, 7 and 8 are marked as  Prog0, Prog1, Prog2 and Prog 3. Why the first bit is assigned to pin 8 /Prog3/ then? Shouldn't bit 1 go to Prog0, bit 2 to prog1 and so on?
Otherwise everything is working only the effects order is somewhat messed up.

That's a Grey Code Encoder.
Binary would be easier to relate.

Quote from: toneman on May 28, 2006, 12:47:28 AM
That's a Grey Code Encoder.

The idea behind Gray code is that you only toggle one bit at a time. In ordinary binary code the worst case scenario is when you switch all bits at one time, for example 1111 -> 0000 (of course one, two or three bits can switch simultaneousy as well). In between states there can be short periods of undefined "middle" states (don't know the technical term) when the physical contact surfaces for each bit don't line up exactly, a problem which might get worse with time when the encoder wears out. What happens is that don't jump from for example 0111 -> 1000 (four bits toggled, binary code) but 0111 -> nnnn -> 1000.

Quote from first search hit at Google using "Gray code" as search words:
"Gray codes are particularly useful in mechanical encoders since a slight change in position only affects one bit. Using a typical binary code, up to n bits could change, and slight misalignments between reading elements could cause wildly incorrect readings."

So, what's it sound like?  Any good?

I would say very good and this is expected at 24-bit/ 48kHz. The sound is nice and fat. The biggest "drawback" is everything is preset but it provides several useful presets that I needed. And let's face it - there's no way you can get several professional quality types of reverbs /and other effects/ in one DIY project for this money. This is maybe the first  DIY real reverb/s project. And let's not forget how big is a single reverb tank...
So is it perfect? Hardly. Does it worth the money and the effort? You bet!

How did you guys go about getting the chips fram Wavefront? Just call and order them? It seems like you guys had a heck of a time getting them. Nice job though! I'm making this my next project...and first attempt at digital stuff..

I got the chips directly from them - it took me almost 3 weeks and couple of angry emails.
AFAIK currently these chips  can't be found anywhereelse. Steve, where are you?

I got my chips from them directly too... but that was when they were called Alesis-Semiconductor. I spent about 15 minutes on the phone with the CEO and he was very helpful. Their sales guy was very helpful too and while he kept the engineers fairly insulated from the outside, they get all my questions answered very quickly.

Quote from: MetalGuy on July 09, 2006, 05:58:05 PM
I got the chips directly from them - it took me almost 3 weeks and couple of angry emails.
AFAIK currently these chips  can't be found anywhereelse. Steve, where are you?

i got mine direct and it took much more than 3 weeks.  persistence works and, though they
were slow, they were also very nice.

still haven't built this yet.   :icon_confused:  it will be my first pnp blue project and i tend
to approach the "big moment" slowly.  :icon_biggrin:  and i have had almost no bench time
for the last several months.  :P  i was really pleased to see that you got it going, metalguy.
nice work!

Everything works fine except for the problem mentioned above  - the efffects order is somewhat different.
I would really appreciate if any one could solve this "mystery"?

If your encoder outputs Grey code, then the problem is going to be in mixing up the data bits.

I'm using the same encoder as suggested on the schematic, ordered from Allied. It's connected as per schematic.
I'm just curious if anyone is using the same encoder and has similar problem?

I spent several hours trying to figure out why the effects order is messed up because there's one thing for sure - if you connect the encoder as per schematic you can't get that effect order. Below are some observations concerning the matter.

When  rotating the encoder clockwise depending on where you start you're going to get the following output /pin 1,2,3 and 5 - pin 4 is common or gnd/:

1   1   1   1
0   1   1   1
0   1   0   1
1   1   0   1
1   0   0   1
0   0   0   1
0   0   1   1
1   0   1   1
1   0   1   0
0   0   1   0
0   0   0   0
1   0   0   0
1   1   0   0
0   1   0   0
0   1   1   0
1   1   1   0

where 1 is +5V and 0 is 0V or ground.

When comparing the original effects order table from my previous post and this one you'll notice that that effects order using this encoder is not possible. Why? Let's take a look at the position of 1111 and 0000 in both tables. In the first table 0000 is 6 positions away from 1111, while in the encoder's table it's 10 positions away /in both cases clockwise/ which means that no matter how you combine those bits you can't get that effect order. The only way you can get that is to substitute 1 for 0 and 0 for 1 and connect as 5123, not 5312 as suggested. I don't know if that is possible - maybe some of the experts should chime in.
Next was to check if any effects order is posssible and how. After spending couple of more hours combining bits the only combination  that promises a plausible effect order turned out to be 2135. This way you should get the effects in the following order:

2   1   3   5

1   1   0   0   Chorus
1   0   0   0   Flanger
1   0   1   0   Room 1
1   1   1   0   Room 2
1   1   1   1   Room 3
1   0   1   1   Plate 1
1   0   0   1   Plate 2
1   1   0   1   Plate 3
0   1   0   1   Chorus/Room 1
0   0   0   1   Chorus/Room 2
0   0   1   1   Vocal
0   1   1   1   Rotary
0   1   1   0   Hall 1
0   0   1   0   Hall 2
0   0   0   0   Delay 1
0   1   0   0   Delay 2

which means that encoder's pin 2 should connect to AL3201's pin 8, 1 to 7, 3 to 6, 5 to 5. I haven't tried this yet but the weekend is near and it will be done. I hope it works.
If you see any flaws in my observations please let me know so we can finally put the cherry on top of this project. Also if anyone has any suggestions concerning this problem I would appreciate if you share them with us.

It's now confirmed that if you connect the encoder in the way described above /2135/  you're getting that effects order.

Quote from: MetalGuy on July 21, 2006, 04:03:05 PM
It's now confirmed that if you connect the encoder in the way described above /2135/  you're getting that effects order.

thanks for working this out!

Quote from: A.S.P. on May 27, 2006, 06:55:40 PM
encoder "pinout"?

2135 is backwards/mirrored 5312 ...

Yes, it is. Any other combination I've tried works only partially - at some point the effects get mixed.

oops - my last post missed a wink/hint-smiley...  :icon_wink:

One question about these guys, is there any layout with components values, or any parts list.. any sound sample.. difficulty something to have an idea, how it sounds etc.


thansk a lot


bye bbye

DAC

I don't know of any sound samples on-line, but the capabilities are almost exactly the same as the Alesis Picoverb, but the FemtoVerb is more suited to guitar signals.

See a review of the picoverb here: http://www.soundonsound.com/sos/Mar03/articles/alesispicoverb.asp

For a layout with component values, read this whole thread. :)

Ok thanks a lot. I've got a problem, i can't find the components. I'll have to wait until my dad goes to USA or a friend, he's a pilot  that's why :P


Do you know the difficulty grade of this project, or anyone that finished it to ask for it, or for pics?


Thanks a lot


Bye bye

Dac

As Peter said if you read the whole thread you'll find answers to most of your questions including pictures.

Hi, wow - that's a great looking project, nice one metalguy!

How much did it cost you to make in total?

got a few sets of the 3 needed ICs... PM me.

or check here: For Sale (http://www.diystompboxes.com/smfforum/index.php?topic=49130.new#new)

Would you be able to post a link to the PCB layout in PDF form (hopefully to scale)? Thanks

Allied is out of the suggested encoder.

Do you think that this is a suitable alternative?
Digikey part # - CT3003-ND
Its described as a "ROTARY ENCODER 4 BIT GREY CODE"

QuoteWould you be able to post a link to the PCB layout in PDF form (hopefully to scale)? Thanks

I don't have it in PDF form but you can scale it down in Photoshop, Paint Shop Pro or similar by setting the resolution to 600dpi.
Hope this helps.

Here's the pdf I made.

* Make sure that scaling is off when you print

http://www.cs.utexas.edu/~dtb/Femtoverb_solder_500dpi.pdf

enjoy!

Thank you. It worked perfectly.

Glad I could help!

Does anyone know if Digikey part # - CT3003-ND will work for the encoder?

Allied isnt getting another shipment of the honeywells until Feb 7th.

Thanks!

hello i'm fabricio from france
thank to pure tube who send me some chipset
thank to metalguy who designed the pcb
femtoverb is working. nice job. I just want the reverb preset so i use the rotary switch method (good idee work perfectly)
just one probleme : there is a "plop" when i switch on the pedal. What wrong ?

excuse for my english

nobody ?

from a look at the layout in reply#9
it seems that the input cap is floating.
(can`t find the schemo)

a so called pulldown resistor (1M) from input to ground will probably work.

The final correct layout is Reply #48.

that one also got input cap floating.
solder 1M from input pad to ground.

According to the schematic the 1M resistor after the input cap connects to Vr, not to ground and it's exactly where it connects now.

but before the input cap?
(where popping originates).

There's no such on the schematic but if you want to add one it's OK.

ok - after waiting half an hour for the board to get back to life
and going through 50 replies
and the search function
I finally found the schematic drawing:
http://dstockett.50megs.com/femtoverb.gif

It`s not just "OK" to add a pulldown resistor at the input,
but "a must" IMHO.

you`re not to blame for that, MetalGuy, BTW.

Hi,

I am new to the forum regarding posting. I have been reading quite a bit and I am back to pedal building since a year. Have done quite some builds since and have learned a lot reading here. Thanks for that. This is a great learning place.

I just finished the build of the Femtoverb. Thanks to Metalguy.
I used a chipset that I got from Puretube. (thanks for that!!)

I made the PCB myself from the layout in this post. I read somewhere that this was not one that would be easy to "iron'.
My experiences was that it was easy to iron.

I live in Europe, and PNP is hard to get here. I use the photopaper method, using epson SO41126 paper. Works like a charm.
Soldering the femtoverb chips was the biggest challenge. I just realized that next time all I have to do is glue them to the pcb first then solder....

The pedal worked right away.
To my taste there was not enough wet signal.
To increase that I changed the 10k resistor between 1/2 op-amp 1 and the first 1/2 op-amp 2. This give a lot More signal going into the a/d converter later on. In my situation it can handle it easily. No distortion or whatsoever. (I am using guitars with humbuckers mostly, so it would apply to most other situations I would guess)
This helped a lot in getting more wet signal.

I read in a post somenone changed the mix pot to 1k, but to my experience that only changes how much you can adjust between wet and dry. 10 k makes a nice curve of adjustment.

Nevertheless I did not like the way the mix know works. You loose to much sound when going from dry to wet, as the drylevel goes down when you turn the mix to the wet side.
Instead of the mix pot I used two 10k resistors. one coming from the dry and the other from the wetsignal, as a 'set"mix. Then I used the mix pot as a wet level pot. In fact connected as it was: slider now going to the 10 k resistor. and the side where first the dry signal was connected now connected to Vr. This gives you a wet signal level that you mix with the level of the dry signal. For me a much better sound.

Some of the effects need the wet signal to be louder then the dry signal. To achieve that I changed the 10 k resistor of the "set" mix created with the two 10k resistors to a 4k7. In that way you can rise the wet level above the dry level.

This works really well now with a much better sound.

Overall it is a nice pedal and a very nice build. A lot of fun. :icon_twisted:

For me the reverbs are a bit modest. I do not know if it would be possible to change that?? I am open to suggestions. :icon_confused:

Thanks again for the info.

Marc

PS as soon as I find out how I will post some pictures

Bump...

Anyone have any suggestions??

Thanks, Marc

Hello, Marc. I'm glad you're enjoying the Femtoverb. I also played around with different ways to improve how the dry and wet signals blend, but in the end I settled on simply changing the mix pot from 10k to 1k, which is how it is in the original circuit that the Femtoverb is based on. To my ears, I can get a nice blend between the wet and dry, without any noticeable drop in volume when turning knob from one end to the other.

One idea you might consider is using an op amp mixer to control the blend; maybe something similar to the output section of Scott Swartz' PT-80 delay. However, I don't think you can get 100% wet and 100% dry with this type of mix control. I'm not entirely sure, but I believe only one of the signals can be 100%, while the other just blends into it.  Still, for the Femtoverb you could design the mixer so that the reverb is the signal that can go to 100%. If you think about, is it really that important to be able to dial in a 100% dry signal on your reverb pedal? Isn't that what the foot switch is for?

Hi DaveTV,

Thanks for responding.

Yes it was your post I read on changing the mix pot to 1k, thanks for sharing the info.
I choose the solution I described which works great for me. If I get a chance I will draw up a little schematic of that for others in the future.

Did you manage to find a way to get a bit deeper reverb as, to me , the reverb could be a bit more intense. (this is not about the mix of dry and wet but about the reverb intensity)??

Thanks, Marc

feedback?

Marc, I agree. The reverb sounds are good, but it would be nice to be able to tweak them further. As I'm sure you know, the AL3201 chip was designed to be programmable so that it can be used in a wide variety of audio devices. All of the effects in the AL3201 chip can be accessed at the programming level and tweaked, but unfortunately this is beyond the scope of the Femtoverb (and my own knowledge and ability). The Femtoverb relies entirely on the default settings for each effect, so there really isn't any way to change the reverb sounds unless you program the AL3201 yourself. So the answer is yes, there is a way to make the reverb sound deeper, but it requires going beyond the Femtoverb's capabilities. Luckily, you'll find a lot of enthusiastic support here in this forum if you're ambitious and want to get deeper into DSP.

Quote from: puretube on February 14, 2007, 01:30:29 PM
feedback?

Hi Ton,

Any suggestions on using feedback in this design??

Dave,
Yes it would be interesting to see if someone comes up with a solution.

Did you ever have a look at this:
http://www.profusionplc.com/products/RA-FX1V.html
It has been mentioned before. It is a complete board, afaik based on the same chipset, but it has the extra parameter control.
Would it be possible to add that to the femtoverb?

Has anyone tried a dual gang pot to control the blend? Wouldn't this allow a better mix control? I read somewhere that you should use soldering paste for the chips. Where do you get soldering paste? Couldn't you grind some solder and mix it with some flux? I'm not clear on how to switch between the different reverb settings. I think I might start on this project this weekend if I have everything I need.

a fine point on your iron, and thin tin wire will work well, especially with magnifying glasses.
some solder-braid ("wick") will help for accidental bridges.

don`t be afraid of SMD - as long as you can clearly see what`s happening, it will work.

(i had bad experiences with (bought) solderpaste remnants underneath chips / beween pads,
that caused intermittent (high-ohmage) shorts in CMOS circuits...)

Thanks Puretube. How do you control the switching of the different reverb settings?

O.K. I see the Honeywell/Clarostat on Allied. How is it wired? Does anybody have a wiring diagram for the jacks and pots?

C'mon guys I need some help with the offboard wiring of this. Does the 2M2 resistor go to ground? I see it does on the last schematic posted. but not on the previous ones. Would it be possible to replace the 4.7uF cap with an electrolytic? Which way would it be oriented? How difficult would it be to install a LED display to show the type of reverb being used? It could display Plate, Hall, Deep, etc. I'm really excited to get this project finished.

Quote from: ranchak on March 09, 2007, 12:48:37 AM
C'mon guys I need some help with the offboard wiring of this. Does the 2M2 resistor go to ground? I see it does on the last schematic posted. but not on the previous ones. Would it be possible to replace the 4.7uF cap with an electrolytic? Which way would it be oriented? How difficult would it be to install a LED display to show the type of reverb being used? It could display Plate, Hall, Deep, etc. I'm really excited to get this project finished.

Hi ranchak,

This is not a fast progressing post as not many have build this pedal apprently. That is probably why you are not getting a fast response.

Here is what I can contribute.
But there is a lot of information already in this thread. Go back a few pages and the answer on the wiring of the  encoder is described already. The original wiring in the schematics is not how it should be, as it will give another order of the programs. Read here to see what Metalguy wrote: http://www.diystompboxes.com/smfforum/index.php?topic=42305.60. It worked out good for me with a encoder I got from vishay (in europe)

yes the 2.2M resistor goes to ground. Are you working with the PCB of this thread, because that resitor is already in there.

I see no 4,7 uf cap? Which one are you referring to??

I can not give you an answer to your question about a program led display. Would be interesting, but not necessary for me with the rotary encoder being clear as it is.

Hope this helped?

regards, Marc :-\

Hi Mark, there is a 4.7uF cap in the upper left corner of the schematic on page 3 of this thread, it is listed as 4n7. I did read the response from Metalguy, I just wasn't sure of where to start (pin 1, etc). I need to do some studying on this topic. I'm pretty good at looking at schematics or a PCB and seeing what is happening and how to hook up the offboard wiring, but I don't have the experience/knowledge to do it on this project. I don't know how to hook up Mix A, MixW, MixB, Vol, etc. Thanks for your help, Ron

Quote from: ranchak on March 09, 2007, 11:01:24 AM
Hi Mark, there is a 4.7uF cap in the upper left corner of the schematic on page 3 of this thread, it is listed as 4n7. I did read the response from Metalguy, I just wasn't sure of where to start (pin 1, etc). I need to do some studying on this topic. I'm pretty good at looking at schematics or a PCB and seeing what is happening and how to hook up the offboard wiring, but I don't have the experience/knowledge to do it on this project. I don't know how to hook up Mix A, MixW, MixB, Vol, etc. Thanks for your help, Ron

Hi Ron,

First of all: are you actually working on the project?? Did you solder all the components already? You are saying that you do not have the experience and the knowledge but at the same time I get the impression that you are working on the project?? It is surely not an easy project.

Ok Let's see if I can help you out:
The cap is a 4.7 nf (or 4n7) and not a 4.7 uf. I would stick to that value and not use an electrolite. Maybe someone else knows exactly why?

As far as connecting the rotary encoder goes: if you look at the encoder from the top down (shaft towards you)
and if the pins are : 1-2-3-G-5
Then the encoder's pin 2 should connect to AL3201's pin 8, 1 to 7, 3 to 6, 5 to 5. G shoiuld go to ground.
(if you look at the pcb layout the reading of the pins of the AL3201 start from the dot  1-8 down)

Is that clear??

Then Mix A, MixW, MixB they go to the mix pot. A and B being the outer lugs and W being the centerlug.. Same for VolA, VolB and VolW.

Marc

Mark, what I meant is I don't know shit about digital circuits. I've built many of the projects from GGG and the Tonepad site. I'm in the process of making a two stage tube preamp from a design from Hoffman amps. I've just never seen the pot wiring referred to as MixW, etc. The only thing I have left to do on this project is the 4.7nf cap, the crystal and the offboard wiring. I'm hoping everything works as I am not looking forward to troubleshooting. I'm a visual person, I need to see the whole project so I can see the big picture (a problem related to A.D.D.), that's why I am having some trouble with this project. I really appreciate your hope in this project. Thanks, Ron

Quote from: ranchak on March 09, 2007, 09:12:54 PM
Mark, what I meant is I don't know shit about digital circuits. I've built many of the projects from GGG and the Tonepad site. I'm in the process of making a two stage tube preamp from a design from Hoffman amps. I've just never seen the pot wiring referred to as MixW, etc. The only thing I have left to do on this project is the 4.7nf cap, the crystal and the offboard wiring. I'm hoping everything works as I am not looking forward to troubleshooting. I'm a visual person, I need to see the whole project so I can see the big picture (a problem related to A.D.D.), that's why I am having some trouble with this project. I really appreciate your hope in this project. Thanks, Ron

Sounds like you are almost there. Hope my information worked.
It is no problem that you do not know a lot about digital circuits with this one. If you managed to solder the smd parts then most of the project is like an analog project. The right wiring of the encoder is important but not crucial: if wired wrong you just get the wrong effect order. This is an easy fix if that would happen.

Is the off board wiring  clear now??

Have fun.

Marc :icon_twisted:

I think I got it under control. What values did you use for the Mix and Volume pots? ( I see a 10KB).

Quote from: ranchak on March 10, 2007, 02:53:04 PM
I think I got it under control. What values did you use for the Mix and Volume pots? ( I see a 10KB).

Hi

50k log for volume
10k lin for the mix but I wired it differntly (see my previous post; my first in this thread) to get a better sound and a different balance.

I can't believe I went this long without asking or even thinking about this.

Flanger and chorus effects are created by mixing a delayed version of the signal with a real-time version of the signal.  When the Wavefront chipset is set to flanger/chorus, do you need the mix control to produce the effects or is the mix/blend pot used to combine a flanger effect in conjunction with a clean tone?  I.E., do you need to pan fully wet to hear the most intense flanging effect?

Hi, Mark. Yes, the mix pot is used to combine the flanger/chorus effect with the dry signal. When the mix pot is turned all the way up, you hear only the flange/chorus/reverb signal from the Wavefront chip. The dry signal splits off into two paths after the first opamp. One path sends the signal through the Wavefront chips while the other path sends the dry signal directly to the mix pot.

Gah!! Still haven't reduced the ambiguity. :icon_biggrin:

When the mix pot is set fully wet in the flanger and chorus modes, is this a) ONLY wet signal or is it b) max intensity flanging consisting of wet+dry in 50/50 proportions?

I know what you're asking, Mark. When the mix pot is set fully wet, the flanger and chorus modes are 100%. To my understanding (which I admit is fairly limited), the circuit treats all of the effects coming from the Wavefront chips the same (reverb, flanger, chorus, delay, etc). The Wavefront chips simply take the dry guitar signal, process it, and then they output what is the effect's 100% wet signal (regardless of whether the effect is reverb or flanger or chorus).

This 100% wet signal is then simply blended together with the 100% dry signal at the mix pot. Theoretically, the mix is 50/50 when the mix pot is at its midpoint, and to my ears (with a 1k pot), this seems to be the case for all the effects (although some of the effects may "stand out" more at the 50/50 point). And yes, when the mix pot is at its min and max points, you get 100% dry signal on one end and 100% wet at the other.

This blend pot seemed a little simplistic to me for a "sophisticated" digital circuit, but it's how it's done in the original schematic I took this from (and probably in the Picorverb). I had thought to try blending the signals using an opamp mixer, but I recall Peter had said this might not work well since it wouldn't allow for a 100% dry/100% wet mix (and some of the effects sound best when mixed in more than 50/50).

Just stumbled onto this: http://www.analog-monster.de/SingleChipReverb.jpg

Unfortunately there is no accompanying commentary, but the poster's webpage is here: http://www.analog-monster.de/schematics_en.html

From the looks of it, and the number of LEDs, I'm guessing the single momentary switch cycles the chip through the 16 programs.

Interesting, but too many additional parts. I'd skip the LEDs and use a PIC.

FYI:
On monday, Wavefront Semiconductor (http://www.wavefrontsemi.com/) has announced the EOL for the 1101 & 1201 for early next year...

Last Order: before December 1st, 2008.

So I was in the process of building another pedal and I was waiting for the paint to dry.  I figured, since I have the jacks and power supply all out and ready to go, I should hook up the Femto-Verb and see what it sounds like.  I populated the board a while back and have been waiting to hear it since.
Soldering the SMD components was pretty easy.  I had to make sure I took my time and stayed patient.   I decided to solder the passive components around the ICs in first because I felt like the less heat I had around them, the better.  Next, I took a tiny dab of glue and glued the ICs to the board.  I waited for the glue to dry and then lightly coated the legs of the ICs where they met the solder traces with paste flux.  I use the Xytronics soldering station that others here have recommended, along with 63/37 solder that has also been suggested here.  I got both from Circuit Specialists.  Setting the iron to a temperature that wouldn't scare me for use on an IC was important too.  I don't remember what it was now, but if you have an adjustable iron you start to get a feel for what works for different situations. I can honestly say that once I started soldering it was as easy as soldering anything else.  It was pretty fun too!  I used a magnifying glass to make sure there were no solder bridges. 
I jumpered the volume control and the mix pot so that the signal was entirely wet.  I wanted to hear the effect full on.  I used a Honeywell/Clarostat 510E1A48F416PB rotary encoder from AA Electric in Wisconsin.
A few things before I get to the sound.  I noticed the layout doesn't have a place for D2 or the 100uf cap.  A little searching earlier in this thread has led me to find that MetalGuy intentionally omitted them since he usually wires them directly to the power supply jack.  I have omitted them for the time being.  I did notice that the 5 volt regulator gets a little hot.  I realized I wasn't grounding it.  I added a jumper from the pad just below the 1uf cap and 100n cap that connects to the middle pin of the regulator.  I measured my power supply at about 9.2 which is what my Power-All always gives off.  On the other side of the regulator I got 5.02.  Even after grounding it, it still gets warm.
Scrolling through the settings with the encoder, I noticed, as others have, that they aren't going in order.  Right now that's not so important, but when it comes time to label an enclosure with the settings I'd like to know which one is which!  It would be more crucial for the more subtle changes in Reverb and not for the obvious settings like Flange or Rotary Speaker.
I'm enjoying all of the sounds.  The main downside is you can't control the settings - other than volume and mix of course.  You can, as others have mentioned, do much worse.  It is a nice and fairly compact layout.  The Flange is pretty standard, not too deep or fast.  The Rotary Speaker is a fun setting.  One of the Chorus settings is a little seasick sounding. Of course, since I currently have the mix totally wet, I'm only getting the echoes from the Delay settings and not my original notes.  The Vocal Cancel setting just gives me clean guitar without any effect.  The Reverbs are all decent at the very least.  I say decent because some settings might work better for other instruments/situations.  I read the review of the Picoverb someone linked to and it's an accurate description of the Femto as well.
Overall, this is a very fun effect.  Most people will probably want to buy a reverb, but for those that insist on making one, this is a great way to go.  Sadly, as noted in the previous post, these chips are likely hard to come by now.  When I ordered mine from Wavefront last spring they sent them out in less than a week.  For those that already have the chips or know where they can still get them, don't skip this project!
Many thanks to everyone who worked on this project.

QuoteScrolling through the settings with the encoder, I noticed, as others have, that they aren't going in order.

This has been discussed already. Check out the thread for the new way of wiring and effects ordfer.

Since the AL1101 and AL1201 are just AD and DA converters, they can be replaced with other ADA chips right?

I also have a question.
This has been mentioned before but no definate answer was given yet.
Has anybody tried these pre-installed boards?
They look good and are quite affordable.
I think they also use the Wavefront AL3201 chip used in the Femtoverb but I'm not sure. 

RA-FX1V
http://www.profusionplc.com/pro/gex/pcatdtl0?ipartno=RA-FX1V
http://www.profusionplc.com/images/data%20sheets/ra-fx1v.pdf

QuoteThe RA-FX1V module is a tiny daughter card assembly that can be installed on main boards in products such as guitar amplifiers, mixers and portable pa systems.
Connection to the RA-FX1V is simple. It has two analog input pins and two analog output pins for easy stereo in and stereo out connection. Four digital input pins are used to select one of 16 DSP programs. An additional analogue input is used for an adjustable parameter input to add a variable parameter per program.

$ 34.80

RA-FX3C
http://www.profusionplc.com/pro/gex/pcatdtl0?ipartno=RA-FX3C
http://www.profusionplc.com/images/data%20sheets/ra-fx3c.pdf

QuoteThe Module is a tiny daughter card assembly that can be installed on main boards in products such as guitar amplifiers and mixers.
Connection to the RA-FX3C is simple. It has two analog input pins and two analog output pins for easy stereo in and stereo out connection.
Four digital input pins are used to select one of 16 DSP programs.

$ 33.12

QuoteHas anybody tried these pre-installed boards?

I haven't tried any of them but I tried this one /~30 Euro/:

It uses an ordinary pot to change program. There's an option for effect parameter control as well /haven't tried that one yet/.

http://www.fk-industrie.de/en/produktkatalog/BG/EM/381.html

The whole line of DSP modules:

http://www.fk-industrie.de/en/produktkatalog/BG/EM/0.html

The ACE99  features ready to install  LED or LCD display.

For full technical details check out data sheets.

Original manufacturer of ACE DSP modules:

http://www.thk-japan.com/

Hi,

It has been a while (2007) since I build this pedal.

I never was satisfied with the amount of effect that I got from the unit.

I changed the way the mixer was designed so I could get more "effect".

Nevertheless my build does not sound as deep as I would expect.

Today I compared it to the picoverb sounds and discovered that that one sounds way better then my femtoverb.

As far as I know it is based on the same chip set.

Could it be that there is something in the design that is not giving me the full effect.

I have checked, rechecked,checked, rechecked my build. Reflowed the solder. Components are as indicated (unless they are not good).

If I read the remarks of other build I think there are more with the same problem.

Any ideas?

Best, Marc

QuoteI never was satisfied with the amount of effect that I got from the unit.

I'm using the board I posted earlier and with the stock 1k mix pot it goves me way more effect that I need so it's a matter of taste as well.
To me all effects are deep and juicy enough but on the same Mix setting they are not equally deep so you'll need to adjust the Mix for each effect.

Sometimes you check and recheck for a problem 10 times and you can't find out what it is but on the 11-th time you get lucky :)

geh, those wavefrontsemo dacs/adcs arent very easy to find anymore

Quote from: MetalGuy on February 01, 2010, 04:21:30 PM

QuoteI never was satisfied with the amount of effect that I got from the unit.

I'm using the board I posted earlier and with the stock 1k mix pot it goves me way more effect that I need so it's a matter of taste as well.
To me all effects are deep and juicy enough but on the same Mix setting they are not equally deep so you'll need to adjust the Mix for each effect.

Sometimes you check and recheck for a problem 10 times and you can't find out what it is but on the 11-th time you get lucky :)

I guess I am up for the 12th time. :icon_wink:



I have signal. I have effect. It is just to shallow and certainly if you compare it to the picoverb.........

Edit: I changed the mix back to 1k and dropped the mix setup construction I had before.
Works Ok still would like more effect depth.

Would it be possible to have the signal going in the AL1101 louder??

I alreday soldered a 10K across the 10k going from opamp 1 to opamp to. That changed it a bit. What would be the best way to do this?


Thanks, Marc

Bump??

QuoteWould it be possible to have the signal going in the AL1101 louder??

I alreday soldered a 10K across the 10k going from opamp 1 to opamp to. That changed it a bit. What would be the best way to do this?

This is the easiest way to do it. Other thing you can do is to recheck everythnig again.

Quote from: MetalGuy on February 06, 2010, 05:04:21 PM


This is the easiest way to do it. Other thing you can do is to recheck everythnig again.

Well this was intersting.
I had checked my built several times, reflowed the connections and checked components.

Nevertheless: challenged by Metalguys remark to check again I reflowed my built again.
Now the sound is much deeper! Including the extra 10k that I had already installed I now have a better sound.

It still is puzzeling me what has happened, but that probably stays a mistery.

Thanks Metalguy for the extra encouragement.

Marc

QuoteThanks Metalguy for the extra encouragement.

I'm glad it worked for you.
I was speaking from experience. Sometimes you check 10 times and everything
seems to be OK. Then you decide to check one more time and you find the problem.
Sometimes you may need another 10 times...

Quote from: MetalGuy on February 09, 2010, 04:52:55 PM

QuoteThanks Metalguy for the extra encouragement.

I'm glad it worked for you.
I was speaking from experience. Sometimes you check 10 times and everything
seems to be OK. Then you decide to check one more time and you find the problem.
Sometimes you may need another 10 times...

Or even then you need someone to put you back at work again.. ;)

Thanks

I have finally finished my Femtoverb which I started to build about 2 years ago.

I also build a PIC controller with LCD to control it and show the chosen preset.

I made a little demo to show the different presets
http://www.youtube.com/watch?v=35whSVXIhn0
The mix pot is in center position.
It is rather noisy and the mix pot acts a little strange as if the wet signal is inverted.

QuoteI have finally finished my Femtoverb which I started to build about 2 years ago.

Looks nice. I didn't get what was noisy. Mine isn't.
If you use it after some high gain pedal however you'll hear some high frequency artifacts (maybe due to the high sampling rate and resolution) but they can be removed with the help of a simple filter.

Quote from: MetalGuy on February 16, 2010, 05:05:05 PM

Looks nice. I didn't get what was noisy. Mine isn't.
If you use it after some high gain pedal however you'll hear some high frequency artifacts (maybe due to the high sampling rate and resolution) but they can be removed with the help of a simple filter.

Maybe not in the video I made, but I can hear the noise especially when notes are decaying and it stops when I cut the note.
I will try some extra filtering.
I think I will also change the circuit around the mixpot a little bit.
What I want to try is to let the dry signal pass to the output and replace the mixpot with an effect level pot.
This may not be the best solution for the chorus and flanger mode, but I think it will give more control over the reverb depth without changing the dry note's high
frequencies.
To me it looks like there's a treble cut when I mix in more reverb.

At first I was not too happy about the reverb sounds, but now I've played it with it a little bit more, I'm starting to like them.
I think my final version will be without display but with a six step rotary switch with only the 2 hall reverbs, 1 room reverb and all 3 plate reverbs.
Maybe I will still use a PIC processor to do the Gray codes otherwise I would need a lot of diodes to do so with the switch only.

Hey Andre,

Nice build and nice to see this with the LCD!!

Could it be that you have added some reverb to the sound of the Femtoverb build on the demo? On the rotary preset I hear a reverb wich is not in the preset of the AL3201 as far as I know.

Marc

Quote from: gitaar0 on February 17, 2010, 12:49:33 PM
Hey Andre,

Nice build and nice to see this with the LCD!!

Could it be that you have added some reverb to the sound of the Femtoverb build on the demo? On the rotary preset I hear a reverb wich is not in the preset of the AL3201 as far as I know.

Marc

Now that you mention it.... yes.
I first used another amp with reverb off, then for the demo I switched to my Peavey Deltablues and forgot to turn off the reverb.
So it's reverb with a bit of reverb.  :D
Maybe I will do another demo with no reverb on the amp.

Quote from: Andre on February 17, 2010, 01:01:44 PM

Quote from: gitaar0 on February 17, 2010, 12:49:33 PM
Hey Andre,

Nice build and nice to see this with the LCD!!

Could it be that you have added some reverb to the sound of the Femtoverb build on the demo? On the rotary preset I hear a reverb wich is not in the preset of the AL3201 as far as I know.

Marc

Now that you mention it.... yes.
I first used another amp with reverb off, then for the demo I switched to my Peavey Deltablues and forgot to turn off the reverb.
So it's reverb with a bit of reverb.  :D
Maybe I will do another demo with no reverb on the amp.

Andre, Very much impressed by your Femtoverb and the youtube "show-off"" :icon_mrgreen:

Peter ;)

OK, I know I am resurrecting a very old thread, but I am probably the last person in the world to have a few sets of the Alesis chips available, and I am about to try and build a Femtoverb.  I am intending to change the PCB layout a bit, so if anybody happens to have an Eagle file for this I would love it so I don't have to redo the whole thing.  Anyone.....?  Please....?

I have done a board for this, and it has been over a week since I last made any changes.  If there is anyone out there who can check it for me in case I have missed anything I would be grateful.  Going to send it out for manufacture soon-ish.

http://www.filehosting.org/file/details/724315/Femtoverb%20sch%20and%20brd.zip

Since this old thread has been resurrected, it is worth noting that pin-for-pin compatible chips are now made by Coolaudio that should work in this project. Note however that Coolaudio has different default programs for the DSP and the max supply voltage is 3.3 instead of 5V. See this page for a complete comparison of the 16 program defaults and the other known differences:

https://cabintechglobal.com/?prod=V1000

--Mark

Coolaudio also don't make the separate A/D and D/a converters - they make a combined CODEC instead.  It would be possible to modify the schematic and board to use that part instead.  As for my project, the boards are back from the manufacturer, and I have started soldering before getting distracted by a different project.  I'll get back to it soon hopefully.