Unable to make FV-1 work..

[amz-fx]

T0 must be connected when you power up the board - it can't be floating.  Pull it up to 3v3 with a 10k resistor and power up to see if the clip LED goes out.

If T0 is grounded, then the 24LC32 MUST be connected.

Measure the dc voltages on the pins and let us know.

regards, Jack

[cloudscapes]

T0 is connected, pulled up to 3.3v via 10k resistor (it was just out of frame when I took the photo on page 2). I ground it when I try external memory, but that's not something I'm worrying about at the moment. I just want to get it to work bare-bones.

I'll do more proper tests later this week. Including pin voltages.

[snap]

Sorry if you've already answered this, but are you 100% sure your vias have continuity?  Does the board draw any current at all?

Also tested for shorts? (sorry for the obviousness).

[cloudscapes]

Things like shorts, grounding/power and continuity from part to part are the first things I test on each of the failed boards. It's the easiest to test so I do it first.

I didn't really want to ask for advice/help unless I did at least that.

[Digital Larry]

Photo of actual board including visible cap markings?

[snap]

Have you tried grounding both inputs through a capacitor, so it doesn`t pick up stray noise?(especially because there are no LP-filters on the PCB).
If the electrolytic in your picture is left floating on it`s negative pole, there could be a static voltage building up on the Spin`s input-pin, which could cause the LED (which of course has its limiting resistor) to light up. Just a thought.

[Ice-9]

It would possibly be a good way forward if you were to post the rest of your off board/breadboard schematic as looking over the FV-1 daughterboard again everything looks fine, so that leaves me with the thought that the problem is somewhere in the rest of the circuit. 

[cloudscapes]

Hey, thought I should post an update. I had shelved it for a couple months because the frustration of repeated failure was getting to me.

It's fixed now! Picked it up again a couple days ago. Remembering what was mentioned about the flux, I took one of the failed test boards, the one in the best condition, and almost literally gave it a bath in electronics cleaner. I normally never use that much on a project since the stuff's toxic. Usually cleaning a board with a qtip is enough. But for this I completely doused it, until it was so clean it looked like a freshly fabbed board (with components). I applied power, and I finally had a clock signal!

I tested the board further. Made sure it could pass sound (which it did). Also found that it could function without much of the support circuitry that I had taken pains to make sure was there for debugging (such as pullups/pulldowns on the s0-2, pots in the pot inputs, etc). Most of everything could be left floating and the FV-1 would still function.

I had one original unpopulated board left over, the first one that was meant to actually be the pedal, so I took the risk to assemble that. Only FV-1 support circuitry first, and cave it a cleaner bath. It worked right away. It looks like octfrank was right, it was the flux. Of the 6-7 boards I assembled, it looked like the only difference between the working and non-working ones was the strong cleaning chemicals. For my other projects, I would somewhat clean the boards with qtips, as I did originally with these ones months ago, and usually that was enough. Even for projects I would have thoguht would be more sensitive to stray capacitance and conductivity, with clock signals between microcontrollers and codecs many MHz higher than what the FV-1 deals with. I find it pretty surprising that those higher speed "trickier" projects are more robust than a relatively slow 32khz clock source. The more you know, I guess.

After I made sure the original first board worked with a minimum for FV-1 circuitry, and assembled the rest. Works fine, looks like the boards worked from the start.

[bloxstompboxes]

I'm glad you got it working and very pretty enclosure.

[octfrank]

Most likely some flux was left around the crystal pins or the crystal pins of the FV-1, crystals can be very sensitive to any capacitance or resistance between the pins and not work.

[anshuman2512]

Exactly the same issue I faced with my hand made phenolic PCB using FV-1 chip, but I managed to fix it in an hour time. I'd say I'm bit lucky here as I was checking for any dry solder joint while powered online and touched the pin 9 & 10 with hot soldering iron and FV-1 booted up with amazing reverb sound.
But when I restarted the power, FV-1 found quiet with the clipping LED on and applied the same procedure with hot soldering iron tip and again FV-1 started working.

First I thought it is something with the clock frequency and a historical case came into my mind.... similar problem I faced before 7-8 years while building a voltmeter using ICL7407 and with a phenolic board, all 7 segment displays continued to fluctuates in random manner without any input signal. That time I found a blog that suggested to use glass epoxy board and to use the minimum amount of flux as the internal resistance of GE board is higher than the phenolic one and less sensitive to stray capacitance.
The same design I built in a glass epoxy board and it was a success.

But here what I did is- used 22pF capacitor for both pin 9+10 (where the 32.768kHz Xtal is connected)  to ground with the same phenolic board, cleaned the board with isopropyl alcohol, waited 10 mins allowing the alcohol to evaporate completely, took a hair dryer to blow hot air on the soldered side for few seconds and that's it. The problem is no longer there and FV-1 is working as expected.
Strange thing is that nowhere in the Spinsemi site it's mentioned as caution and one worse is FV-1 doesn't boot up without the crystal, however some useful info from Spin site http://www.spinsemi.com/knowledge_base/xtal.html

I'd like to thanks cloudscapes, octfrank and all members for opening this thread and posting their suggestions, this would help other hobbyists to fix a problem. Much appreciated !

Have fun building the FV-1 project; it's a great reverb chip. I've one more FV-1 chip which I'm planning to build a Dev board. My next project would be Coolaudio V1000 DSP along with external ADC/DAC.

[lovellmusiclab]

Hey guys. Long shot here
Anyone else around had made further discoveries on the FV1s not booting up?
I've been dealing with a pretty similar situation of some chips randomly working and some other not.
I've tried changing xtals, caps, chips, and cleaning them as lastly suggested but nothing had worked.
Perhaps something more had been discovered since the topic was posted?

Cheers!

[Digital Larry]

Hey guys. Long shot here
Anyone else around had made further discoveries on the FV1s not booting up?
I've been dealing with a pretty similar situation of some chips randomly working and some other not.
I've tried changing xtals, caps, chips, and cleaning them as lastly suggested but nothing had worked.
Perhaps something more had been discovered since the topic was posted?

Cheers!

Where did you get them?  Apparently fakes abound.