Spin Semiconductor

Hi, I have been building an FV-1 based pedal but have been having a little problem with distortion. What is happening is that after powering up the circuit it sounds perfect for a few minutes then develops a distortion sound which I can only get rid of by powering down the circuit and then re-applying power, then it works perfectly again for a some time and the problem starts again.

Has anyone else came across this problem and found out any cause ?

The circuit I have built is following the datasheet with simple opamp buffers for in and out. I'm thinking it might be a problem around the crystal and cap to ground or possibly the 3.3v power line which is using a l78l33 reg.

I can draw up a schematic if it will help and also post a picture of the PCB.

Thanks

I have never seen that before, problems with the crystal are usually a noise at power up then sounds fine due to crystal starting at a harmonic of its frequency.

I would use a scope to look at the inputs and outputs of the FV-1 to see where the distortion is coming from. What program are you running, an internal or external one? How stable is the 3.3 supply? Needs to be stable for the converters.

Schematics and PCB images always help us debug things.

Thanks frank for the suggestions, I don't have access to a scope at the moment but i'm very sure the signal into the FV-1 is clean leaving me thinking the 3.3v line must not be stable causing a problem with the converters. I will test with a battery instead of a PSU in case that may be at fault.

All programs are external eeprom programs and this is happening with every hex I load, hex that I know to be working 100% in other builds.

Picture of pcb but no schematic at the moment. I will draw one up after checking the 3.3v line.

Try with a pure through program, read ADC write to DAC, no other effect.

Is this a new design or a board you have used many times in the past? Any solder bridges? I've seen a board where the problem was a short under the FV-1 because 2 traces were very close together and there was no solder mask.

It is a new PCB design but that particular FV-1 chip has been soldered into about 8 different pcb's (it's my test chip)and could possibly have developed some damage from being re-used so many times.

I have 3 new chips so I may removed that one and check the traces under it for shorts and replace it with a new one but first I will do as you suggest and flash an ADC to DAC read write.

I tried using a battery instead of a wall wart and also added extra 100n cap on the 3.3v line along with the 10uf that is there but that didn't help.

I will get a schematic up tomorrow.

Problem solved now, the problem lay in a bias resistor in the output buffer which had gone short so after a short time began to throw the bias out on the opamp. Sometimes you can't see the obvious until you walk away and have a think about it.

Thanks for the help, now onto finishing this off and getting a final PCB design. When finished all details will be posted on the usual DIY effects forums if anyone is interested in making one.

ice-nine wrote:Problem solved now, the problem lay in a bias resistor in the output buffer which had gone short so after a short time began to throw the bias out on the opamp. Sometimes you can't see the obvious until you walk away and have a think about it.

Thanks for the help, now onto finishing this off and getting a final PCB design. When finished all details will be posted on the usual DIY effects forums if anyone is interested in making one.

How are you making the programming of the SMD eeprom?

Sweetalk wrote:
How are you making the programming of the SMD eeprom?

I am using a pickit2 which connects to a header I have on the PCB, if you look at the picture above you can see a set of 4 solder pads just above the Eeprom. The pads are Vdd, Vss, data, and clock, by using the 4 pins there is no need to power up the pedal to program the Eeprom as the pickit2 supplies the power to it.

Next time I will not solder a connector to these pads as the pickit2 can just push into the pad holes without having to remove the PCB from the pedal.

You will want to be careful when using a pickit2 for in-circuit programming. It is 5V Vdd default I believe and the FV-1 is 3.3V only so make sure to set the programmer to 3.3V.

frank wrote:You will want to be careful when using a pickit2 for in-circuit programming. It is 5V Vdd default I believe and the FV-1 is 3.3V only so make sure to set the programmer to 3.3V.

Yes I should've mentioned that, or alternatively if the pedal is powered then the Pickit2 will detect that and just use the pedals own power, but yes if using power from the pickt2 programmer then it would be wise to set it to 3.3v

That's how I do it, programming with the pedal powered, that way you don't need the power connection from the PICkit2, meaning only 3 connections.
I've done a couple of pedals with stereo 1/8" jacks in them for programming using a lead that takes SDA, SCL and ground from the PICkit, this allows programming without opening the pedal.

slacker wrote:That's how I do it, programming with the pedal powered, that way you don't need the power connection from the PICkit2, meaning only 3 connections.
I've done a couple of pedals with stereo 1/8" jacks in them for programming using a lead that takes SDA, SCL and ground from the PICkit, this allows programming without opening the pedal.

Excellent idea to use stereo jack for a programming port.