Spin semiconductor fv-1

[PharaohAmps]

Well, I get all kinds of horrible noise from mine when it's connected to my PC over USB.   But when I run it from a linear power supply (xfrmr -> bridge -> 9v regulator) I get no noise at all.  Well, a bit of hiss from my mixer but nothing terrible.  My input connection is super-crappy too, I just pulled up the R in and out RCA jacks and routed them to a TL072 as buffer and mixer.  Works okay but I've got to get some 3.3v opamps to put it together in the end.

Nothing weird on the fading signal for me, could it be that as the s/n ratio goes down you're hearing it more?

Matt Farrow

[puretube]

ouff - I remembered that I hadn`t heard it in the beginning of experimenting...

guess what?

in the meantime I had my (relatively new) soldering station switched on
to solder some mono/stereo-adapter...


must be µP controlled, and adding to the wave-chaos...   

[puretube]

after a night`s sleep, the intermodulation is still there without the solder-station,
but the scope verified, that the guitar-signal is already dirtied up by the sm-powersupplies, before it goes into the Spin...

gotta clean up the sloppy wiring!

what the scope discovered, too:

the Spin`s "dry" output signal is being phase-shifted from ~0° at very low frequencies
up to several thousands of degrees at high audio-frequencies
compared to the "original" (=input) signal.

when sweeping a signal-generator from low to high,
the 2 traces on the scope look like those of a >24 stage phaseshifter... 

(I noticed this, when trying to watch the modulation-depth of the tremolo-preset).

[RaceDriver205]

Out of interest, why would one want to do anything with DSP?
The advantage of making your own pedals is that they are analog and thus much better sound than cheap multifx digital units? If you want the advantages of DSP, why not just by a DSP multifx machine?
Not being provocative, I just want to know.

[puretube]

a.o.: coz of the nonavailability of some BBD chips,
and to do things, you can`t easily do analog/non-mechanical:

Delay - Reverb - Pitching

and:
http://www.diystompboxes.com/smfforum/index.php?topic=42359.0
http://www.diystompboxes.com/smfforum/index.php?topic=23055.0

[Mark Hammer]

Out of interest, why would one want to do anything with DSP?
The advantage of making your own pedals is that they are analog and thus much better sound than cheap multifx digital units? If you want the advantages of DSP, why not just by a DSP multifx machine?
Not being provocative, I just want to know.

Multifx units that attempt to accomplish too many concurrent tasks for the processing power they pack get a deservedly bad reputation.  DSP-based effects that do not attempt to reach farther than their capabilities can do some pretty special things.  More and more single-function DSP pedals are turning up in the market and earning kudos from players who find them more than satisfactory.  I still think that distortion is generally accomplished more effectively n the analog domain, but apart from that many digital pedals do a magnificant job.  I am the proud owner of most of the Line 6 Tone Core pedals and I enjoy playing them as much as my 100% analog ones.

Certainly one of the things that digital has done extremely well over the years is reverb.  Good reverb CAN be done entirely in the analog mode, but is often limited by the space required (got a spare 500-gallon oil drum, buddy? or would you mind hoisting this 5ft x 5ft plate reverb into the minivan for the gig?) and is subject to mechanical noise intrusion and very restrictive control over reverb parameters.  A simple, cheap circuit that gives some control over a pleasing complex reverb is a wonderful thing to have available at one's disposal.

As for relatively trivial things like flanging, chorus, and simple delay, I would agree with you that sometimes these tasks can be accomplished quite well (and occasionally even better) in the analog domain for very modest cost and time/effort.  When you get them thrown in for essentially free along with an extremely cost-effective digital reverb, though, it is too appealing to resist.

The other area where DSP often outshines analog is in the area of modulation.  The sorts of circuits that could provide the modulation options available in many DSP-based pedals would require cost-prohibitive or space-prohibitive ciruitry to accomplish in the analog domain.  I suppose if you're John Frusciante, and can have a wall of patched Doepfer, Modcan, Catgirl, or MOTM modules sitting offstage, operated by a tech, then you can go the all-analog route, but most musicians are not in that league.  The modulation capabilities I have in my Tone Core Liqui-Flange (just a bit bigger and chunkier than a typical DOD pedal) would take something the size of that largest E-H pedals to accomplish in analog (maybe bigger) and even then it would be a tight fit, likely to be noisy, and bloody expensive.  Small digital circuits may not yield ALL the flexibility of 100% analog, but they will easily take you 80% of the way there at a fraction of the cost and with 98% of the audio fidelity (well, as much as ears ruined by neodymium headphones and full-tilt iPods can detect).

Of course, here we reach a different sort of "digital divide".  You and I can likely build analog circuits, and even improvise within that context, but when it comes to digital stuff we are hampered by the required knowledge.  Heck, I understand a bit about assembler code, but I couldn't burn an EPROM if my family's life depended on it.  Many folks look back with fondness on the golden era of electronics hobbyist magazines when the projects did not require $400 scopes and wads of PIC knowledge to build things.  I know how they feel.  Many beginners here stay completely clear of this separate forum, like it's some sort of dark alleyway where danger lurks.  Happily, things like the SPIN chip have gone the distance to make many of the benefits of digital available to "regular folks" like us.

[David]

This is your roving cyber-reporter David, coming to you from the digital "Dark Side".  Now that I've been assimilated by the digital Borg, I have the barest inkling of the beginnings of understanding of this shadowy realm.  I believe the esteemed Mr. Hammer is right.  Analog is preferable whenever possible (which is why my lonely, unlamented GT-3 is gathering dust in my gigbag -- which couldn't make me happier!) -- however, to mangle a quote from R.G., there are some things that are a lot easier to do digitally -- some of those being MIDI controllers, stompbox routers, loopers...
...  and now this.  I can't wait to see how this "Spin"s out!

I think I've had too much coffee...   

[PharaohAmps]

Certainly one of the things that digital has done extremely well over the years is reverb.  Good reverb CAN be done entirely in the analog mode, but is often limited by the space required (got a spare 500-gallon oil drum, buddy? or would you mind hoisting this 5ft x 5ft plate reverb into the minivan for the gig?) and is subject to mechanical noise intrusion and very restrictive control over reverb parameters.  A simple, cheap circuit that gives some control over a pleasing complex reverb is a wonderful thing to have available at one's disposal.

+1 on reverb.  The FV-1 comes with some very decent reverb programs in its ROM.  If all you want is a nice reverb, you can get away with an FV-1, a 3.3V regulator, and a dual opamp.  Seriously.

If you want to get jiggy with it, then you can pick up an inexpensive serial EEPROM programmer (about $20 on ebay) that will work on 24LC32's, and the free software from Spin Semiconductor will make .HEX files.  You don't really NEED the dev board, but its nice and quick.  The FV-1 will do certain things MUCH better or at least much more easily than any analog circuit - reverb and pitch-shifting are good examples.  It's easy to program and has a veteran effects designer behind it (Keith Barr.)  It's a great starting point for DSP newbies.

Right now I'm working on interfacing the program pins to a PIC, so that I can change the clock speed and mute / unmute the straight signal when I change programs.  It's pretty easy to do, actually, but I'm a PIC newb too   One more thing to learn!

Matt Farrow

[puretube]

yes, a BIG THANX to Keith Barr!!!



[next step he takes, probably will be the 9V µ-power version:
<1µA in standby/bypass mode,
2mA (running thru an LED...), when ON;
with built-in electronic standby-detector (aka "mill-tbp",
so that a normal 4PDT stompswitch can be used for true bypass] 

[SeanCostello]

+10000 on the thanks to Keith Barr. I have been designing reverbs for about 8 years now, and the example code that is on the Spin Semi site is AMAZINGLY good. I had figured out most of these techniques a few years back (by...um...doing a LOT of tests on a Midiverb 4), but to see them in print, complete with delay numbers and a theoretical explanation, is an embarrassment of riches. The other people associated with the company are also very helpful and informative - there is a forum on the Spin Semi site that a few people from the DIYstompboxes forum have started to visit.

My FV-1 development board should arrive by Friday. I am pretty excited. I hope to design a few of my own reverb algos, as well as some cool modulation effects.

Sean

[Jack]

Has anyone tried looping or even a hold reverb yet?

[SeanCostello]

Has anyone tried looping or even a hold reverb yet?

Yes. The 2nd one.

In rev_pl_1.spn, change line 54 from

sof 0.6, 0.3

to

sof 0.7, 0.3

This allows for the gain in the allpass ring recirculation to go up to 1.0, which results in infinite reverb.

No preconditioning of the signal yet to remove transients, but I just got the compiler running about 1/2 hour ago, and just started poking around.

This is fun...

Sean Costello

[PharaohAmps]

Has anyone tried looping or even a hold reverb yet?

As for looping, I've messed around with a delay program with a nearly infinite hold, but with only 1 second of delay time (have to change the crystal that comes on the eval board to 32.768khz) there's not much joy to be had for the looper.

I'm working on a code snippet to use one of the POT inputs for a tap-tempo function.  If you use an external mixer, you can get a really nice tap-tempo delay with feedback and delay time controls onboard, with no extra funny business.  Total cost in parts is under $20, not including stomp switches and enclosure.

Any interest in a "generic" FV-1 board?  Peter was talking about one with his Lotus board, but I've got to do some boards for my own production, and 100 boards isn't much more expensive than the 30 I need.  If people want 'em I'm happy to sell.

Matt Farrow

[David]

Any interest in a "generic" FV-1 board?

Why, yes!  How thoughtful of you to ask!  What feature set did you have in mind?

[PharaohAmps]

Any interest in a "generic" FV-1 board?

Why, yes!  How thoughtful of you to ask!  What feature set did you have in mind?

Well, the idea would be to implement everything that's on the datasheet:  3-bit program selector, 3 pot pins (these are CV, not audio, and you can use any pot you want,) and stereo in / out.  One easy option is to use one of the signal inputs as a control input, and also to do the signal mixing in hardware to free up software controls.  It's up in the air right now but I'll keep everyone posted.

Matt Farrow

[SeanCostello]

Will the generic board have a place for an EEPROM?

I'm interested...

Thanks,

Sean Costello

[Peter Snowberg]

That sounds pretty similar to what I have in mind. Keeping the mix control in hardware allows for the dry signal to be totally analog which is desirable for lots of people and it keeps the mix control from using up one of the three parameter controls. Since so much of the overall design is based around the FV-1 capabilities, if we all try to keep our boards similar, we'll end up with E2PROMs that can plug in anywhere.

[PharaohAmps]

Peter, I'm thinking it might be better if we come up with something together, rather than trying to develop on our own.  PM me if you're interested.

Matt Farrow

[MR COFFEE]

I vote for the MIX control in hardware IF it works.

Someone with a demo board needs to check on the effect of the reported frequency-dependent phase shift before designing a PCB with a hardware mix pot.

I'd buy a bunch of the PCBs if they had a space for the external EEPROM and a internal\external switch line.

[SeanCostello]

Even if the FV-1 adds frequency-dependent phase shift, the MIX control would be useful. For flanger and phaser effects (where the phase of the mix is most critical), the programs could be developed such that the mixing of the modulated and straight signal happens within the FV-1 itself, and the MIX control should be advanced to fully wet signal in order to hear the effect. For chorus, having a nonlinear phase response of the delayed signal may not be such a bad thing. For reverb, any phase shift of the ADC/DAC is inconsequential, as reverb creates a highly nonlinear phase response, so having a little added by the ADC/DAC will not be noticable.

Sean Costello