Crossfader Thread (will eventually be the build thread)

J3RK · Post by **J3RK** » Fri Mar 30, 2012 1:26 pm

Moving the original message to the bottom of this post, to put the BOM, notes, and other materials in plain site:

QTY  PART-REFS           VALUE               CODE                          
---  ---------           -----               ----                          
Resistors
---------
8    R1,R5-R7,R13,R16,   30.1k                                             
     R25,R27
2    R2,R10              60.4k                                             
2    R3,R11              1M                  ***Possible Mod***                                               
8    R4,R8,R12,R14,R31,  100R                                              
     R35-R37
15   R9,R15,R17-R22,R26, 10k                                               
     R28-R30,R32-R34
1    R23                 14k                                               
1    R24                 11.5k                                             

Capacitors
----------
4    C1,C3,C5,C8         330pF                                             
2    C2,C6               22uF                *AC Coupling caps &#40;jumper for DC&#41;
6    C4,C7,C9-C12        100pF                                             

Integrated Circuits
-------------------
2    U1,U3               LM13700                                           
3    U2,U4,U6            TL072                                             
1    U5                  LT1013                                            

Transistors
-----------
2    Q1,Q2               2N3906                                            

Diodes
------
3    D1-D3               1N4148                                            

Miscellaneous
-------------
3    BIAS,VB1,VB2        100k     

***R3 and R11 can be changed to lower values to impact the overall bias of the VCAs.  1M was chosen as a trimming value.  VB1 and VB2 will only impact the bias slightly.  Low values &#40;like 10K, will shift the signal all the way to one side of DC, but may introduce some changes/distortion to the waveform.&#41;  More extreme &#40;lower&#41; values may be more interesting when these controls are brought to the panel.  1M is perfect for trimming in cases where VB1 and VB2 will be trimpots internal to the module.***

Code: Select all

12V Modifications&#58;

For 12V operation, please change the following resistors&#58;

R23 from 14K to 11K
R25 and R27 from 30.1K to 24.3K
R24 from 11.5K to 12.1K

The new range of the CV section should be 0V to -11V, with a crossover point of around -5.5V.

Code: Select all

Notes&#58;

One thing I would like to call attention to after thinking about it a bit.  VB1 and VB2 are bias pots.  Not input attenuators.  I'm not sure that this has been clearly called out up until now.

Clarke's panels leave these controls open to however people would like to use them, which is cool.  One could easily wire the pots up as input attenuators for the VCAs, and use trim-pots internally for VB1 and VB2.  

Or, VB1 and VB2 can be brought to the panel, and R3 and R11 can be tweaked to taste for biasing the signal around DC.  I'm going to set these resistors to an extreme value, and post some results in this thread.  &#40;just to give an idea of what these will do.  I've seen some fairly interesting results in the simulations that I ran, and hope they translate to interesting audio.  I'll post notes on this shortly.

One more note.  The output resistors can be bumped up to whatever your preference is for output protection.  The op amps only source around 20ma of current, so 100R should be ok in most cases &#40;unless you connect it to a power rail or something.   &#58;deadbanana&#58;  There's no reason not to stick 1K resistors there to be on the safe side.

***Original Message***

Hey!

So here is the requested crossfader thread.

Since it will have different uses, (as a part of the dual 291 configuration, stand-alone, etc.) I made the thread title very generic.

I'm calling it the FadeX for my own purposes, but also like the name that Clarke chose for his panels. It kind of falls in line with some of the other panels out there. So call it whatever you like, but here's where we can discuss it.

The test boards will be here in a few days. Once I get one tested, I'll toss up the BOM, start making some build notes, and put up other materials as needed.

One issue that has come up, is that I'm trying to standardize my board sizes and mounting hole spacing. I'm trying to make all of my boards 3" wide, by 1, 1.5, 2, 3, and 3.5 tall. I want them all to use 2700 thousandths hole spacing (2.7") The 291 boards I believe are either 2750th" or 2800th" so this creates a bit of a mounting issue. One approach that I've been discussing with Clarke, is to do one side in 2700, and the other in 2800 or whatever the 291 turns out to be. (I'll be checking this later.) I'm not sure that this will work as it might go beyond the board edge though, so that's one more item I have to figure out before the final production run gets ordered.

If anyone has any novel mounting ideas that would allow us to match these up in a usable way with the 291s, but allow it to keep the new, normalized spacing, I'd love to hear them.

clarke68 · Post by **clarke68** » Fri Mar 30, 2012 1:43 pm

Thanks Dustin!

Here's the first draft of my dedicated Crossfader module panels. The idea was to do a full-featured Crossfader in 8HP, and then a more svelte version in 4HP:

Discuss!

frozenkore · Post by **frozenkore** » Fri Mar 30, 2012 1:46 pm

Let's see, how many... I think I need to just start rolling dice. I'm up for two 4hp and one 8hp and corresponding boards.

JP · Post by JP » Fri Mar 30, 2012 1:55 pm

4x4hp please !

microfauna · Post by **microfauna** » Fri Mar 30, 2012 2:29 pm

FADE CV IN attenuator

on the 8HP

edit
(actually on both, instead of MIX INV OUT)

theabsent · Post by **theabsent** » Fri Mar 30, 2012 2:34 pm

microfauna wrote:FADE CV IN attenuator on the 8HP

Yes, that would be nice. I'm very likely interested.

CLee · Post by **CLee** » Fri Mar 30, 2012 2:41 pm

Crossfader boards... yes very interested.

Can the xfader be wired up to be a panner easily? Should be, yes?

Thonk Support · Post by **Thonk Support** » Fri Mar 30, 2012 2:53 pm

are they equal power/voltage?

J3RK · Post by **J3RK** » Fri Mar 30, 2012 3:12 pm

The fader is buffered in a way that there is no dip in amplitude when panning from one source to the other. This is what allows it to work when fading one waveform to another in a VCO application. I can post the plots I have soon, showing how it works.

If you look at just the plots of the two VCAs, then you'll see a dip where they intersect. When looking at the mixed output though the amplitude is the same across the entire fade.

I assume this is what you mean.

I hadn't given any thought to panning. There may be a way to do this, but I'll have to think about the signal path a little bit. Each input is going to a separate VCA, which is then mixed into a single output.

You might be able to using the individual VCA outputs, but you'd likely have to feed both inputs the same signal to make it work. This would be pre-mixing stage though, so there would be a dip in amplitude toward center if I'm not mistaken. Since I don't have the test boards yet, I'll run some simulations for some preliminary results.

clarke68 · Post by **clarke68** » Fri Mar 30, 2012 3:13 pm

microfauna wrote:FADE CV IN attenuator

theabsent wrote:Yes, that would be nice.

Okay...that's two votes. I'll mock it up this weekend.

CLee · Post by **CLee** » Fri Mar 30, 2012 3:21 pm

J3RK wrote: I hadn't given any thought to panning. There may be a way to do this, but I'll have to think about the signal path a little bit. Each input is going to a separate VCA, which is then mixed into a single output.

You might be able to using the individual VCA outputs, but you'd likely have to feed both inputs the same signal to make it work. This would be pre-mixing stage though, so there would be a dip in amplitude toward center if I'm not mistaken. Since I don't have the test boards yet, I'll run some simulations for some preliminary results.

It's something I would probably do with a set of boards. I think panners and xfaders are the same circuits, just a xfader sends 2 signals to 1 place and a panner sends 1 signal to 2 places. So I'd just need to tap into the VCAs before they're summed, buffer them and bring them to outputs.

Thonk Support · Post by **Thonk Support** » Fri Mar 30, 2012 3:22 pm

J3RK wrote:The fader is buffered in a way that there is no dip in amplitude when panning from one source to the other. This is what allows it to work when fading one waveform to another in a VCO application. I can post the plots I have soon, showing how it works.

perfect, I'll take 4-8

JJ · Post by JJ » Fri Mar 30, 2012 4:27 pm

Could those attenuators be bipolar so that one could invert the input? This would be handy when fading between multimode filter outputs, for example, or when creating new waveforms at audiorate.

With input1 normalled to input2 inverted we could even fade between the original signal and its inversion. Or just use the module as a 2:1 bipolar mixer.

I'm really tired, I hope I came across sane.

J3RK · Post by **J3RK** » Fri Mar 30, 2012 4:34 pm

JJ wrote:Could those attenuators be bipolar so that one could invert the input? This would be handy when fading between multimode filter outputs, for example, or when creating new waveforms at audiorate.

With input1 normalled to input2 inverted we could even fade between the original signal and its inversion. Or just use the module as a 2:1 bipolar mixer.

I actually have a 1"x3" precision CV polarizing mixer board designed that could do this, and adds a 5V reference voltage for shifting things around DC.

I think it would require hacking things up a bit on the current fader layout though.

This is a good time to bring up (again) the fact that on the 291 boards, the LP tap is inverted. So, one can either set up an inverting amp on the 291 board in the mod space for this tap, or a little polarizing processor might be nice.

obscurerobot · Post by **obscurerobot** » Fri Mar 30, 2012 4:36 pm

I'm going to need a PCB for the 291, might want a few extras because why not?

Clarke's panels are great as usual. I think I prefer the 8hp, but may not be ordering any panels this time.

ringstone · Post by **ringstone** » Fri Mar 30, 2012 4:59 pm

Very nice! I certainly could use 2 or more of these... Are there any special parts required for the VCA, such as SSM2164?

Cheers
Blair

J3RK · Post by **J3RK** » Fri Mar 30, 2012 5:33 pm

ringstone wrote:Very nice! I certainly could use 2 or more of these... Are there any special parts required for the VCA, such as SSM2164?

Cheers
Blair

The only part that could really be classified this way would be two LM13700 dual OTA chips. They're still made, and are sold at Mouser, Digikey, and a lot of other places too. (I imagine SmallBear would have them too.) I just bought a ton of them, because I use them in quite a few things. I think they were under $2 each.

The next special part down from that would be the recommended LT1013 or LT1112 dual precision op amp. These are also easy to find. You only need one per board just for the fader section. You can actually use other dual op amps (TL072, etc.) but I recommend the LT here.

The mix output stage can be a simple TL072, but you could go with an OPA or something like that that's a little fancier in the audio department if you like.

Otherwise typical standard parts. I'll get a parts list/BOM type thing up as soon as I test the proto board in a few days. I just want to make sure all the parts values work.

emdot_ambient · Post by **emdot_ambient** » Fri Mar 30, 2012 6:36 pm

Oh...nice project. Been waiting for these to show up. Most likely I'll be in for 4. Very useful.

bsmith · Post by **bsmith** » Fri Mar 30, 2012 6:47 pm

Def tote in for 3 thankya thankya.

ringstone · Post by **ringstone** » Sat Mar 31, 2012 1:23 am

J3RK wrote:
ringstone wrote:Very nice! I certainly could use 2 or more of these... Are there any special parts required for the VCA, such as SSM2164?

Cheers
Blair
The only part that could really be classified this way would be two LM13700 dual OTA chips. They're still made, and are sold at Mouser, Digikey, and a lot of other places too. (I imagine SmallBear would have them too.) I just bought a ton of them, because I use them in quite a few things. I think they were under $2 each.

The next special part down from that would be the recommended LT1013 or LT1112 dual precision op amp. These are also easy to find. You only need one per board just for the fader section. You can actually use other dual op amps (TL072, etc.) but I recommend the LT here.

The mix output stage can be a simple TL072, but you could go with an OPA or something like that that's a little fancier in the audio department if you like.

Otherwise typical standard parts. I'll get a parts list/BOM type thing up as soon as I test the proto board in a few days. I just want to make sure all the parts values work.

Thanks mate! Always try and keep a stock of LM13700 on hand if I can...

Cheers
Blair

asterisk · Post by **asterisk** » Sat Mar 31, 2012 9:26 am

id be in for 3-4 of these boards.

id also vote for normalling VCA1->VCA2
then you could do panning easily that way as well.

are these faders DC coupled?

janvanvolt · Post by **janvanvolt** » Sat Mar 31, 2012 2:46 pm

ONe (large) Panel and the PCB please guys!

PF · Post by PF » Sat Mar 31, 2012 3:59 pm

Ill take 4 - 6 pcb's, i get back with a final amount.

ATOM · Post by **ATOM** » Sat Mar 31, 2012 4:45 pm

Another J3RK project

I take 4 pcbs.

ATOM

JJ · Post by JJ » Sun Apr 01, 2012 1:39 pm

I'm pretty sure I want 4x 4HP panels and 4 PCBs.

MOD WIGGLER

MOD WIGGLER

Crossfader Thread (will eventually be the build thread)

Crossfader Thread (will eventually be the build thread)