Little Angel problem

[bluebunny]

The more I think about it, the more I believe the 100n that is shown on the layout connected between I and O of the regulator is an error. That 100n was meant to go between I and G.

Good spot!  I guess it would let noise on the power line get through to the 5V output, and on to the PT.  I guess that other builders with quiet power supplies wouldn't have noticed any issue.  But you definitely should have the I and O pins bypassed to ground via caps (just like Merlin has done ^^) - it's exactly what the datasheet tells you.

[Elijah-Baley]

Ok!
I have to try it!

Thank you.

[Elijah-Baley]

Tried.
I desoldered the 100nF cap (film) through I and O of the regulator and I soldered two pin socket through I and G.
With and without the cap the background noise like a FFFFFFF when I play is always there. Not so loud, but is there. If I can eliminate is better, else I have to hold it.

[bluebunny]

It's worth remembering that the PT isn't a hi-fi part.  And by all accounts, some are born less hi-fi than others.  Have you tried swapping out your PT for another - preferably from a different supplier?

[Elijah-Baley]

I am almost convinced that the problem is the noisiness of the PT2399, anymore.
I'm reading several topics in the web about this IC and eventually noises, especially in delay circuits where it is often used.

I haven't other PT2399 at the moment, and I bought this from Tayda, I could buy it from Musikding, but it is more expensive unfortunately, and I afraid of another failure. LEss expensive in another shop in my country, but is the same brand to Tayda.

I could buy one sometime a try it. But I wonder, assuming the problem is really my PT2399, if this is noisy on this circuit, probably it will be noisy in other circuit, some delay for example, right? You know, I'm plaining a easy cave dweller delay circuit.

[bluebunny]

. . . but is the same brand to Tayda.

They're all the same brand - Princeton Technology.  But they do appear to have variable quality.

[Elijah-Baley]

Thanks, I didn't know it.

[...] But I wonder, assuming the problem is really my PT2399, if this is noisy on this circuit, probably it will be noisy in other circuit, some delay for example, right? You know, I'm plaining a easy cave dweller delay circuit.

Somebody has an opinion about this?

[anotherjim]

They do have variable quality, and it seems to have changed with time. I wonder if the lowest quality chips go in the DIL packages while the good ones go in surface mount which the big volume customers are probably buying.

[Elijah-Baley]

They do have variable quality, and it seems to have changed with time.

Indeed, some delay circuit has noise problem with long time delay or several repeats. Just read it.

[Elijah-Baley]

Update, finally.
I tested the circuit with further 3 PT2399. No changing, the background noise when I play remains. Tried another 78L05, and nothing.
I checked the voltage on ICs and 78L05 and are ok. Just the battery slightly diying.

And now?

PS Testing the circuit (after a long time) I noticed when the speed pot is to 0 a quick detune. On every setting of the depth.
I can't notice it anymore qhen the speed is higher, enough obviously I think.

[...]
I desoldered the 100nF cap (film) through I and O of the regulator and I soldered two pin socket through I and G.

Now it is soldered as early. Maybe a bad solder?

I'm waiting the new solder tin.

[Elijah-Baley]

Hello guys.
I'm still stuck with my Little Angel (or devil?). Background noise in chorus mode.
I'm looking those caps on the right side of the PT2399. There is a "New Year's Edition" and those caps have different values.
I noticed that the noise rise up when the treble are high.
Could those caps filter treble or something?

I have decide if save this circuit (it at least is good as vibe) or I have throw it away.

Thank you.

[ElectricDruid]

The way I read that LA layout, the 100nF cap across regulator I & O pins is giving any high frequency noise from the power supply a fast, no waiting route into the circuitry instead of bypassing it to ground.

+1 agree. That looks like a mistake to me. The should go from I to G, not O. The board will work anyway (as you've discovered) but it'd be better with the supply properly bypassed.

Tom

[lethargytartare]

I really think your best starting point is to breadboard a new build of the same circuit -- that will let you eliminate your soldering work from the equation.  Or it might confirm that what you're getting is as good as it gets.  It will also let you experiment with the suggestions more easily -- as you desolder and resolder things over and over, you increase the risk that you're going to fry something else, etc.  If you breadboard it and get a better result, you can then compare pin voltages between your soldered version and your breadboard version.

I'm in the same boat on another build...it's frustrating, but it's the nature of DIY sometimes!

[Elijah-Baley]

The way I read that LA layout, the 100nF cap across regulator I & O pins is giving any high frequency noise from the power supply a fast, no waiting route into the circuitry instead of bypassing it to ground.

+1 agree. That looks like a mistake to me. The should go from I to G, not O. The board will work anyway (as you've discovered) but it'd be better with the supply properly bypassed.

Tom

I tried to move that cap with no changing. The layout is verified, sometime the PT2399 is the problem, but I changed 2 or 3.

It is a bit hard for me for breadboard it.

[ElectricDruid]

The way I read that LA layout, the 100nF cap across regulator I & O pins is giving any high frequency noise from the power supply a fast, no waiting route into the circuitry instead of bypassing it to ground.

+1 agree. That looks like a mistake to me. The should go from I to G, not O. The board will work anyway (as you've discovered) but it'd be better with the supply properly bypassed.

Tom

I tried to move that cap with no changing. The layout is verified, sometime the PT2399 is the problem, but I changed 2 or 3.

It is a bit hard for me for breadboard it.

To be honest, I wouldn't expect moving that cap to solve anything much. It looks like an error where it is, but it's not going to stop the circuit working, or fix it when moved if it doesn't.
The rest of the layout looks ok, although I'm not a fan of those bodged ground wires linking everything together. The Little Angel chorus is probably fussy because it uses the chip in a way it was never designed for, and the PT2399 isn't exactly hi-fi at the best of times. So if it makes some noise, I'm not really surprised. I'd be reluctant to suppose that meant that you'd actually made any error, even if other people might have got lucky and built circuits that make less noise than yours does.

Sorry. No magic bullet to offer.

Tom

[Elijah-Baley]

Thank you, ElectricDruid, this is the point!
If there's any error, somewhere, I really can't found, so it could be there's no.
I know PT2399 is "noisy", indeed I tried 3 different, perhaps, but no solved.

In "Vibe Mode" it is noiseless.
Could somebody tell me why? What I bypass in VIbe Mode, and what I include in Chorus Mode that could be responsible of the noise?
This is the schematic, more or less. This is not the exactly version I built, but it has the switches.



Just another clue: I missed, in the first test, the diode. I noticed the battery was warm. Could some components be damaged or something?

[duck_arse]

what type opamp are you using? maybe check that 47uF on the V/2 line. does it still noise if you disconnect the depth pot so it doesn't modulate?

[Elijah-Baley]

I built this:


It is a bit different from the schematic, and I follow the layout, so the IC is a NE5532P.
I'm sorry, what 47uF cap you meant? Near the 78L05, the 1N4148 or the opamp?

About the depth pot what I should to do? There's the toggle switch connected, it is a bit difficult.

And I had joint Speed 1 & 2, and Depth 1 & 2. I had seen this connection in another layout and I found weird to have disconnect those lugs, so I thought was correct joint them.

[anotherjim]

Use such as TL072 or RC4558 instead of NE5532 in this one. The 5532 is an excellent hi-fi amp, but becomes very noisy with high resistor values. I wouldn't think the 5532 is the best choice for the LFO either. Perversely, even the MC1458 or LM358 would be better than the 5532 in this particular effect.

The vibrato mode is all PT2399 output, so less noise then is probably due to the filtering in the delay lines own op-amps making the hiss less noticeable.

[PRR]

> battery was warm.

PT399 sucks almost 100mA.

This is a LOT for a 9V battery!!

(I do not know why it is such a pig.)

The pulled-down battery voltage *may* be related to your hiss problem.

Try it with a solid 9V source.