FV-1 Development Board (w TAP SW)

phi1

Well-known member
I finally finished my FV-1 Dev board. It's been really fun seeing what some of the other folks on here are doing with the FV-1 boards, so here's my contribution.

I have added a momentary switch for tap tempo (or other functions), and an LED controlled by the DACR output for monitoring tempo. There is also a switch to select between internal and external Eeprom chips.

So far I've gotten lots of patches I've found online to work on it. For using the tap switch, I've gotten the Babelfish taptempo delay working, after a few tweaks. It can easily be configured to do dotted 8th delay. I also got a patch working for infinite reverb with the tap sw held, while you play over top (sort of like the EHX freeze). This was actually crudely achieved with the "Reverb+Pitch+LP" patch on the SpinSemi website by basically just swapping pot0 and pot2 in the code. I plan to edit it more in the future to make it more suitable for this use.

Here's a bit of detail on the mods, and I attached my schematic page.

The "Tap->" toggle switch configures CTRL3 for either normal operation (full pot range, no tap sw input) or tap operation. The pot is still effective in tap mode, but it has a limited range (from about 0.25 to 0.75 instead of 0 to 1). The tap switch pulls CTRL3 to 3.3v (value of 1). Another cool beneift is that without any patch editing, the FTSW maxes out CTRL3 regardless of the pot setting. For example, the shimmer verb can have CTRL3 low, and then the FTSW brings in the shimmer when you want it.

The "EXT->" toggle switch toggles between an EEPROM chip installed on the inside of the pedal (chip labeled LAB in the gut shot photo), and a socket on the side of the enclosure. This will allow me to keep this pedal closed up and quickly switch EEPROMs, or flash an EEPROM for my Arachnid pedal.

The TAP LED signal is taken from DACR output, so I had to solder a wire onto that pin of the FV-1. The FV-1 DACR puts out max 3.3V, and the load minimum from the datasheet is 10kohm. I think technically it would be possible to directly drive an LED with this, but probably not very bright. So, using a transistor to drive the LED seems like the better route. The Babelfish tap tempo code flashes this LED with the tempo.

Not only was this my most involved build yet, but I also experienced an abnormally high number of problems (maybe you'll spot some in the gut shot haha). Fortunately, everything was recoverable, and now everything performs as desired.
 

Attachments

  • front shot.JPG
    front shot.JPG
    418.2 KB · Views: 315
  • gut shot.JPG
    gut shot.JPG
    432 KB · Views: 343
  • usb side.JPG
    usb side.JPG
    395.2 KB · Views: 278
  • ext.JPG
    ext.JPG
    394.1 KB · Views: 207
  • dc.JPG
    dc.JPG
    391.5 KB · Views: 179
  • left iso.JPG
    left iso.JPG
    411.7 KB · Views: 155
  • right iso.JPG
    right iso.JPG
    418.7 KB · Views: 185
  • Schematic.pdf
    609.6 KB · Views: 294
I like the external Eeprom a lot so far haha. With the momentary sw, I'm still kinda getting used to it. I like the idea of tap tempo delay, time will tell if it gets used much (the group I play with at church doesn't always use click, so... tap tempo isn't that helpful). We'll see about other uses in the future.

Since others might be interested in doing the tap sw, I'll give a bit more background about the circuit in case it's not obvious from my post:
-The 47kohm from 3.3v to pot_lug_3 is there to intentionally limit the pot max to 0.75 in the code. That way, it will be easy for the program to read the switch being pressed (value of 1) even if the pot is maxed.
-The 47kohm from pot_lug_1 to ground is there to prevent the switch from shorting 3.3v to ground when the pot is all the way down. This has a side effect of limiting the the pot min to 0.25 in the code.

Fortunately, the pot values can be programmed to whatever you want, so it shouldn't matter. If you're writing all your own patches, the toggle switch wouldn't be necessary since you could just manipulate the pot value in the code. But I kept my options open to make it easy to test other's code, and for developing my own code for my arachnid build, which just has normal ctrl3 operation.
 
Ok update / heads up for anyone thinking of doing a tap sw.

At certain settings on my build, the program is slow to detect or slow to release the tap switch. For example, when the ctrl3 pot is low, a can't get the tap tempo delay program to read a tempo.

I believe this is due to the hysteresis / filtering on the internals of the fv-1 pot inputs (I read about it on the architecture page of the spin semi website).

I did a bunch of testing with code with the tap sw just turning on the tap led. The optimal setting seemed to be ctrl3 pot high (reads near 0.75 in the code) and the detection threshold within the code at 0.85. This setting works great for the tap tempo delay program. If the ctrl3 pot is turned down significantly, the program is slow to read the tap. If the threshold in the code is set higher, it is slow to read the tap. If the threshold is lower, like 0.80, then the program is slow to release the tap.
 
I found this gem from almost 9 years ago, it's a code snippet that defeats (well, reduces) the filtering on the pot inputs. I incorporated in my code for pot 2 and now the tap tempo works flawlessly regardless of the ctrl 3 position.


Found out emailing with pedalpcb that SpinCAD has a pot speed up which is the same code.
 
I finally finished my FV-1 Dev board. It's been really fun seeing what some of the other folks on here are doing with the FV-1 boards, so here's my contribution.

I have added a momentary switch for tap tempo (or other functions), and an LED controlled by the DACR output for monitoring tempo. There is also a switch to select between internal and external Eeprom chips.

So far I've gotten lots of patches I've found online to work on it. For using the tap switch, I've gotten the Babelfish taptempo delay working, after a few tweaks. It can easily be configured to do dotted 8th delay. I also got a patch working for infinite reverb with the tap sw held, while you play over top (sort of like the EHX freeze). This was actually crudely achieved with the "Reverb+Pitch+LP" patch on the SpinSemi website by basically just swapping pot0 and pot2 in the code. I plan to edit it more in the future to make it more suitable for this use.

Here's a bit of detail on the mods, and I attached my schematic page.

The "Tap->" toggle switch configures CTRL3 for either normal operation (full pot range, no tap sw input) or tap operation. The pot is still effective in tap mode, but it has a limited range (from about 0.25 to 0.75 instead of 0 to 1). The tap switch pulls CTRL3 to 3.3v (value of 1). Another cool beneift is that without any patch editing, the FTSW maxes out CTRL3 regardless of the pot setting. For example, the shimmer verb can have CTRL3 low, and then the FTSW brings in the shimmer when you want it.

The "EXT->" toggle switch toggles between an EEPROM chip installed on the inside of the pedal (chip labeled LAB in the gut shot photo), and a socket on the side of the enclosure. This will allow me to keep this pedal closed up and quickly switch EEPROMs, or flash an EEPROM for my Arachnid pedal.

The TAP LED signal is taken from DACR output, so I had to solder a wire onto that pin of the FV-1. The FV-1 DACR puts out max 3.3V, and the load minimum from the datasheet is 10kohm. I think technically it would be possible to directly drive an LED with this, but probably not very bright. So, using a transistor to drive the LED seems like the better route. The Babelfish tap tempo code flashes this LED with the tempo.

Not only was this my most involved build yet, but I also experienced an abnormally high number of problems (maybe you'll spot some in the gut shot haha). Fortunately, everything was recoverable, and now everything performs as desired.



Hi all,

I would be very grateful if any of the PedalPCB Fv1 dev board folks with successful dev board builds please comment/advice/ ideas

I have mostly populated the Dev board pcb with the Spin FV1 and CH341 chips presoldered. See pics. Im currently seeing a dead short between 3.3v and ground.



So far I have :-



Inspected and cleaned FV1 and CH341 chip smd contacts.



Eliminated the regulator as the problem.



Pots not problem as not yet populated



Inspected all solder joints and vias - no visible issues.



Component check - bad decoupling caps ?



Many thanks,

Tim
 

Attachments

  • B3C691B7-138C-4896-8C5B-B4961E415F03.jpeg
    B3C691B7-138C-4896-8C5B-B4961E415F03.jpeg
    786.2 KB · Views: 60


What’s that little guy? Also pull that little 15pf ceramic disk cap out and see what happens, those things are prone to fail, and when they go they short out instead of being open.
 
Something looks funny under the 3.3v regluator, can't quite see what's going on.

Noticed a couple other things, but these wouldn't cause the 3v3 to gnd short.
1. Looks like you're missing a 47u cap in the bottom left (unless you put it on the underside of the board laying down. You should be able to do that since there's no pot right there.
2. Check that the yellow cap under the Fv-1 is indeed 2n2. A MLCC 2n2 should be fine, just checking since you used greenies for the other nano range caps.

I'm sure you've checked, but I'd be very careful with all the standing legs on the 1uF caps to make sure they aren't shorting anything. De-soldering isn't fun, but if it doesn't seem reliable/stable, it might be worth getting the MLCC 1uFs to replace those.

Worst case, if nothing turns up visually, you may have to go through the schematic and lift all the components that directly connect to 3v3, then check for continuity after removing each one. But I'd give it a few more scans visually first just to make sure. Feel free to post a pic of the underside too.
 
Thanks so much for replying, very helpful :-
See answers to your points below.


Something looks funny under the 3.3v regluator, can't quite see what's going on.



Yes I lifted the 3.3v leg to check - regulator fine





Noticed a couple other things, but these wouldn't cause the 3v3 to gnd short.

1. Looks like you're missing a 47u cap in the bottom left (unless you put it on the underside of the board laying down. You should be able to do that since there's no pot right there.



Same again took this out to check - cap fine but as expected 3.3 to gnd shorted



2. Check that the yellow cap under the Fv-1 is indeed 2n2. A MLCC 2n2 should be fine, just checking since you used greenies for the other nano range caps.



Yes the yellow cap is indeed 2.2nF



I'm sure you've checked, but I'd be very careful with all the standing legs on the 1uF caps to make sure they aren't shorting anything. De-soldering isn't fun, but if it doesn't seem reliable/stable, it might be worth getting the MLCC 1uFs to replace those.



Yes ive checked the legs seem fine but worth mentioning. I had to use these caps as Tayda out of MLCC caps ?



Worst case, if nothing turns up visually, you may have to go through the schematic and lift all the components that directly connect to 3v3, then check for continuity after removing each one. But I'd give it a few more scans visually first just to make sure. Feel free to post a pic of the underside too.



Thats what im doing currently a royal pain !?



Ive posted a pic of underside - not the best soldering ?but seems ok

Let me know what you think.





Thanks again for your help !



Tim
 

Attachments

  • 282903C0-BF0B-4DF5-8B46-888F801AADEB.jpeg
    282903C0-BF0B-4DF5-8B46-888F801AADEB.jpeg
    1,019.3 KB · Views: 23
Building an FV-1 development board. Most of the parts I have in my (too numerous) junk boxes. However, the USB connector is new. Can you tell me where you got it? Going even further out on a limb is there a BOM available for this board? Ideally this would be one on Mouser or DigiKey.
 
I got it mine from tayda. A type B usb female from any site should work, I assume the pin layout is standard, see if the pictures where you buy it look like it’ll fit with the hoard holes.
 
Back
Top