Electrical Noise

Hi guys,

I mentioned this in my post in The Workshop (thread: http://mutable-instruments.net/forum/comments.php?DiscussionID=132) but didn’t get much joy, so I thought I’d continue it in the Electronic Workshop.

My Shruti-1 makes quite a lot of noise, even when it’s not playing notes. Just turning it on creates a noticeable jump in noise. The output volume (a linear 100K pot from Maplin which I was told would be fine as a replacement for the logarithmic 100K pot that came with the kit - I killed my pots with bad soldering and had to buy replacements) controls the amount of noise, but also has an effect on the quality of it. As you turn the pot down it turns from more of a “white” kind of noise into more of an electrical hum. Touching a wire between the body of the output pot to the GND pin on the hack_me port doesn’t seem to have any effect (likewise with the input pot, or with both pots simultaneously).

Has anyone else noticed this? How should I go about tracking down and eliminating the source of the noise?

If it will help with diagnosis, I can try to record the noise and post it up somewhere.

Sorry for the double post - nobody seems to have any ideas, so I went back to play with it again with some new tests of my own. I think I made i worse :frowning:

I wondered if the noise could be coming from the power supply (the supply I’ve got seems to put out the right voltages and stuff, but the plug feels ever so slightly “loose” in the socket), so I tried powering the Shruti with a 9V battery instead. The Shruti powers on fine with the battery, but the electrical noise is still there.

It was at this point I discovered that I seem to have killed the LCD screen somehow. The light comes on when the Shruti powers on, but I no longer see anything on the display. Any ideas what I might have done, and how I can test and rectify (or replace!) it?

The buttons, knobs and LEDs still work despite the dead screen, and I can still play the test tone, and notes from the keyboard, and edit the sound - just nothing on the screen. Another thing I noticed that I hadn’t before is that when any of the buttons are pressed down, the sound of the electrical noise changes - becomes more “buzzy” - and then changes back when the button is released. Any idea what that could be?

To rule out any problems with the interface board, I removed it, and tried just powering on the main board whilst it’s connected to the amp. Can’t play a test tone because there are no buttons plugged in (I didn’t have the keyboard plugged in at the time, but I presume it would have played notes) but I was only really listening for noise anyway. Still there. So there’s a problem with the main board (or with the switch or pots).

I took a look at the main board under a good light and a magnifying glass. My soldering isn’t the greatest in the world, but I couldn’t see any obviously dry joints - can you see anything in the following picture?

(Oh, the shame…)

Grounding the pots hasn’t helped, but they still seem like they might be the likeliest suspects. Is there a way I can short them, to get an idea what the Shruti would sound like without them, or would I have to physically remove them?

Any help with any of these questions would be great. Right now I seem to be breaking the poor thing more every time I go back to try to diagnose why it’s so sick in the first place :frowning:


Regarding the shorting of the pots, you can have a look at this picture to show where the shorting has to be made:

The LCD screen problem is likely to be a bad connection… Have you checked continuity between the pin 4 of the ATMega328p and the RX pin of the LCD?

It is not possible to test without the main board, because the analog input used to read the pots is not connected to anything, so you’ll get random values input into all the parameters. You could probably try that by connecting the “>ANL” pin to +5V but then, at startup, the 4 parameter of the filter section will get set to their maximum value.

I think I’ve found the problem with the LCD - the connection between the ATMega and the RX pin on the LCD connector is fine, but there’s no continuity error between the plug that goes into the connector and the point where the RX is soldered onto the LCD. I’m not able to break out the iron right now, but that looks like it’ll be quite straightforward to fix.

Shorting the input volume pot didn’t seem to make any difference at all (but then I’ve not been testing with any input going into the Shruti so that doesn’t surprise me). However, shorting the output volume pot actually makes the noise worse. What can I conclude from that? Problem with the pot, or perhaps with the wire I’ve used to attach it (it’s the same type of wire I used to make the short)?

When you say you have shorted the output pot, you mean that you have totally desoldered it, and have put instead a cable like on this photo? A bad solder point on the audio connector that would make the sleeve of the audio jack not connected to ground?