MISTAKE ON DIRTYSAM loop pedal!! (ISD chip) do this modifications!!

[mathflan]

hi,

the dirtysam project is a loop pedal based on the ISD 2560 chip like mobius trip looper and loop junkie effect from ZVEX...
It has been created by Guruzeta http://humano.ya.com/guruzeta/Guitarra/dirtysam/
I didn't arrived to work this pedal and I recently found that there's a mistake on the PCB and shematic:
there a mistake on the input fx which goes to the opamp TL072

so do this modifications: ( you have to cut trace on pcb ..)

[mathflan]

Do not pay Attention to the TL 072 chip on modifications schematic
the modification is on pin 1,2,3 of the TL074

[mathflan]

I'am wondering if the signal is in phase when the signal input is at input - of the opamp?
therefore, I always see in schematics, signal input to input MINUS of th opamp

[ethrbunny]

My guitar signal comes through ok.. but there is alot of noise in the background during playback. Could this be the same problem?

[mathflan]

It's strange that your signal is ok with the original pcb...
I checked many times mys PCB but it's ok.
I tried to back to the original PCB, input to  PIN 3 ( interveting input + ), PIN 2 (interveting input -) and PIN 1 ( output)  weld together but No more sound of my input , When I put my guitar, I have to play very loud to have some distortion sound...
maybe my TL074 is different...


I have too, a lot of noise but I guess it's due to the ISD chip...
I don't think that the noise is due to this PLUS or MINUS input, since the signal which is record by the chip is take before the OPAMP.

I'll try change the LM386, a guy said me that this chip is not appropriate for speaker output.
I added a compander NE570 like the PT-80 to reduce the noise of the ISD, and it reduce a lot the noise but I have less low frequecies ( I will change some capacitors...) an the compander " eat" a little bit the sound, I will reduce a little bit the ratio of the compander...

Is there someone who could tell me which ratio, resistor' value and capacitor's values I have to add to the compander to works the best with a ISD chip??

[ethrbunny]

I have too, a lot of noise but I guess it's due to the ISD chip...
I don't think that the noise is due to this PLUS or MINUS input, since the signal which is record by the chip is take before the OPAMP.

How much noise and of what 'type'? IE periodic? static? I tried an audio probe to see where its coming from. Near as I can tell its some problem with the ISD chip. I corresponded with Winbond about it though they were not too helpful. Im going to try separating out the +5 dig and ana to two diff 7805s too see if there is noise coming in that way. Winbond did suggest that what I was hearing was the 'charge pump'. I have some recording equipment now- Ill try to get an .mp3 of the problem too see if that better illustrates it or if its just the same thing you are hearing.

[mathflan]

ok, I'll do the same...

[Jaicen_solo]

I'd definitely be happy to hear some sound samples of the dirty sam. I've always been put off by the apparent complexity of the design, and my inability to read spanish, hence my alternative design. I've also just created a variable power supply to use with these chips which should provide a nice vibrato effect. It can be built on a small daughter board to allow it to be added to the dirty sam.
I'd very much like to hear what this design sounds like, so I can compare it to my own in terms of noise floor etc. (when I build it ).

Can you just clear something up for me, i'd like to know why you've included CMOS control to utilise the OVF pins and control LE & PD. It's not really necessary when using the ISD25xx series as that's all built into the chip. As long as the PR pin is held high, the chip will loop indefinitely whenever it reaches an EOM marker, or reaches the end of the capacity. If you take PR low during playback, the sample will play to the end and then stop.
Similarly, you can just use a latching DPDT switch for selecting between play/record.

[mathflan]

Hi,

I put a Pot to control Voltage of VSS of the ISD to add a Pitch down Effectn very cool, How do you add a vibrato effect, with a LDR-LED combo??

You said "Can you just clear something up for me, i'd like to know why you've included CMOS control to utilise the OVF pins and control LE & PD
It's not me that created the dirtysam, I think that Gurutheta put a 4070 to use only one SWITCH to record and play, instead of two switch ( one for ENABLED and one for PLAY/RECORD).
I use a momentary switch, I push the switch, it record and I release the switch and it play, but I have a noise when I activate the switch...

I'll try today to put two voltage regulators to see what happen on the noise...

ethrbunny, do you have some TICK noise or GRound noise when you activate the BYPASS switch and the PLAY/RECORD switch?

[Jaicen_solo]

Oh, I see now.
My layout uses a DPDT switch which sends a reset signal when switched from record to play. That way, it starts looping as soon as you take it out of record (well, a few milliseconds after). The reset pin also needs to be tied to one lug of a 3DPT bypass switch so that the loop plays back as soon as the pedal is switched on.

[mathflan]

I have some news....

First, THe modifications that I Posted is not important, I 'll try again to back to the original schematic and it works . I don't why, perhaps bad TL074...
Now I have again the same problem Noise on the play volume sample.
In fact I'am wondering if it's not because of the 3 MEGA ohm before the input of the opamp of the Analog IN , I think that this resistor reduce a lot the sound and so, We have a very bad signal-to-noise ratio.
I put a 220k for try and the I have now a good signal-to-noise ratio, very less noise, but the play volume give too much volume...
I compare the schematic of the dirtysam withthe Mobius trip looper and there a lot of differences on the schematics BEFORE analog IN.

I'll try to do some modifications...

[ethrbunny]

My heavens! You say you had the noise problem and this resistor change *fixes* it? (throws child in lap to the floow, pushes wife out of the way, runs into the garage and holds a match under the solder iron to get it to warm up faster....)

Ill try it ASAP and let you know what happens... be still my beating heart!

[ethrbunny]

It's not me that created the dirtysam, I think that Gurutheta put a 4070 to use only one SWITCH to record and play, instead of two switch ( one for ENABLED and one for PLAY/RECORD).

ethrbunny, do you have some TICK noise or GRound noise when you activate the BYPASS switch and the PLAY/RECORD switch?

Mathflan - how tough would it be to break out the 'enable' from the 'record' switching? IE make it so you could have the thing enabled but not playing, then hold down a momentary to get the sample, release it to play once or twice and then stop it?

The noise I get is a grinding sort in the background. Hard to describe. Its definitely periodic. Ill try the input resistor swap to see if it makes a diff.

[ethrbunny]

Hmm.. Well I took out my 3M2 and replaced it with a 1M. Different type of background noise but there is still alot of noise there. Instead of a grinding sound Im now getting a buzzing. Its a very regular pattern at about 30-40Hz (just guessing). Ill try the 220K next I guess.

[mathflan]

I think your noise is just a Voltage Noise,
CAhat do you use, 9 Volt battery or voltage Adapater??

I use a Voltage adaptater that I putted to 9 volts but I had a VOltage noise in continue like you describe...
So I Putted to 12 VOlts And no more noise!

I think I will change the schematics berfore the Analog IN to see if it better or not.
I'm sure that it's the 3 Mega ohm which reduce the sound in the Analog IN and that give a very bas signal to  noise ratio.

I still have this "pop" noise when I push the two switch....

[ethrbunny]

Voltage noise? errr.. ok. Ill try it at 12V. Ill have to redo my LEDs though.

What value for the resistor (instead of the 3M2) did you finally settle on?

[mathflan]

ethrbunny, can you record a sample of your dirty sam?
I could know what is it really this noise in back ground...
I gonna work on the distysam this evening..., I'll keep you informed about that

[ethrbunny]

Here is a short sample of the problem.

http://lupinemedical.com/mp3/problem_sound.mp3


Its a v short loop.

[Jaicen_solo]

That's interesting. The sound is actually related to the loop, there's a slight pause everytime the sample re-starts.
This suggests to me that the noise is coming from the analogue in circuitry. Perhaps some of the sampling clock is getting into the audio. It definitely sounds like some sort of high frequency square wave pulse, so it's certainly possible. You should make sure that there's nothing connected to pins 17, 18, 19, 21 and 11. Also, you should check that the filter caps on the power supplies are oriented correctly, and that your supply voltage is clean. Makes sure you are getting +4.5 - 6v at both the Vcc, and that Vcc_D and Vcc_A are as isolated as possible. You should not connect signal grounds (ie, jacks) to Vss_D (Pin 12), only Vss_A (Pin 13). You could try adding a 0.1uf Cap from pin 28 to pin 12 right next to, or even underneath the 2560 chip itself, to makes sure it's properly decoupled. Do the same for pins 13 & 16.
Finally if this doesn't help, remove the 386 chip and tie the output of the 2560 directly to a speaker and see what happens! 
One last thing, I can't see the layout at the minute, but you shouldn't ground either of the SP +/- pins.

[ethrbunny]

Jaicen_solo: The noise changed quite a bit when I changed the 3M2 resistor to a 1M. Does that mean anything to you?