Ground issues powering several diy oscillators

[eh la bas ma]

Hello,

A friend asked me for help with his diy project.

He 's trying to feed several oscillators with one power supply and a daisy chain.

It's working when all little circuits are powered with its own power supply, but it's not when he tries to use a single power supply and a daisy chain.

I suggested to connect each negative DC jack's pole to the enclosure, but... could it work ?

here is some pics :












The oscillator is triggered by a photoresistor, an led lights up when the oscilator is engaged. The idea is to use eight of these together and a stroboscope with the photoresistors to create some good old bass drums.

I am not really sure about this, any help or ideas would be welcome !

[Rob Strand]

The oscillators should not care.  So perhaps the first thing to check is the polarity of the DC jacks.   Maybe there's a problem with the DC jack wiring.

It would be a good idea to get two running off daisy chain.   If that works try replacing one of pair.  Repeat that for all channels.  Maybe there's only one that isn't working and that unit is shorting out the power to the bunch when they are all powered together.

Beyond that, I actually suspect the problem is related to the way the LEDs or opto's are wired.   If the switching has a common that common needs to be a ground. 

Without a schematic I can only guess.

[eh la bas ma]

Thanks for your help, here is a schematic :

[eh la bas ma]

Isn't it strange that the DC jack's positive pole is connected to the NE555 's GND ?

[ElectricDruid]

The oscillators should not care.

The oscillators should not care, but they often do.

You don't say exactly how "it's not working", so it's difficult to know - no sound comes out? waveforms get distorted? what exactly happens?

One common problem with multiple oscillators run off the same power supply is that (especially when tuned very closely together...like you mostly would!) they will often "lock" and all oscillate as one. You think you'll get a fat multi-osc sound, and you don't. This happens because there's enough feedback through the power rails to sync them together. Some designs are worse than others, but it's a common issue. Solutions involve choosing oscillator designs that are less susceptible, or isolating the power supplies to each oscillator either by filtering or separate regulators.

HTH.

[Rob Strand]

Thanks for your help, here is a schematic :

From the schematic, the problem I see is the audio out.   It's got a common positive rail.  Also a resistor on the positive rail.  It all looks a bit odd.   If you connect the audio outs to a common ground you will get problems.

So what I would expect is:
- Audio ground to pin 1 (NE555 ground)
- Series resistor on pin 3 (NE555 output) then audio output on the other-side of the resistor.

One common problem with multiple oscillators run off the same power supply is that (especially when tuned very closely together...like you mostly would!) they will often "lock" and all oscillate as one.

Yes, that does happen.  Injection can be used for good but it can also be bad.   Back before 2000 there were a few guys on the sci.electronics.design newsgroup talking about oscillators in the same room synchronizing or at least being affected by each other.

[ElectricDruid]

So what I would expect is:
- Audio ground to pin 1 (NE555 ground)
- Series resistor on pin 3 (NE555 output) then audio output on the other-side of the resistor.

Yes, +1 agree with Rob. This all looks very wrong. Get the signal referenced to ground, and take the output from the output!!

My comments about oscillators syncing don't look to be relevant here - it seems like there are much bigger problems.

[eh la bas ma]

I am trying to draw a new schematic according to your instructions :



Is it allright ?

[Rob Strand]

I am trying to draw a new schematic according to your instructions :
Is it allright ?

Looks better. 

Power switch should be in series with the power.

Not sure about polarity of DC Jack but you will soon work that out.

So, the thing I'm not sure about is,
- Did schematic just have a few errors but the actual unit was wired OK?, or,
- Was the unit wired like old schematic and now the wiring needs to be updated to see if it has fixed the issue?

[eh la bas ma]

I will give the schematic to my friend, and report back as soon as he gives me some news. Tomorrow I hope.

Power switch should be in series with the power.

Isn't it the case ? "in series" means a direct connection, with a wire, from the positive DC jack's pole to one outter switch's terminal, am I wrong ?

Edit :

Sorry, I am still confused about parallel and series...



Is the switch in series with the power jack now ?
About pin 4 : should it be connected to both DC jack and the switch or only to the DC jack ?

[Rob Strand]

I will give the schematic to my friend, and report back as soon as he gives me some news. Tomorrow I hope.

I'm sure he will know exactly what he's done and be cable to comment on the schematic.

Quote from: Rob Strand on Today at 10:13:43 PM


    Power switch should be in series with the power.


Isn't it the case ? "in series" means a direct connection, with a wire, from the positive DC jack's pole to one outter switch's terminal, am I wrong ?

The way it is drawn is parallel.   The both switch contacts are across the DC Jack.

For a series connection one terminal of a part connects to one terminal of the next part.  The parts form a loop like people holding hands in a circle.

To modify the switch connection for series:
- removed the wire on the  DC Jack + terminal (leave the lower switch terminal connected to the circuit)
- removed the wire between the DC Jack - terminal and the upper switch terminal
- add a wire from the DC Jack + terminal to the upper switch terminal

[Rob Strand]

Maybe this helps visualize it,

[eh la bas ma]

Did I read you correctly ?

[Rob Strand]

Did I read you correctly ?

That's it.

[amptramp]

I did a tone generator for an organ at one time using a dozen Fairchild µA2240 devices which are 555 astable oscillators much like yours with an 8-stage divider having open collectors which were pulled high using a SIP 10 K resistor package for each one.  When you have this many oscillators together, you need regulation on the board to keep them from syncing or interfering with each other.  The 555 is notorious for demanding a large pulse of current when the output changes state so decoupling and voltage regulation are necessary even though the frequency is not affected all that much by supply voltage.  Some people use the CMOS version of the 555 just to get away from this, even if it means using some form of amplifier at the output to increase the current output.

https://datasheetspdf.com/pdf-file/1305444/Fairchild/UA2240/1

This is a fun device.