Correct lights, unresponsive - did I break something?

I think I broke something. LEDs 3 and 7 are lighting up but nothing else is happening… :frowning:

I haven’t put the LCD on yet.

The LEDs and controls seemed to be working properly. I wanted to make sure I was getting sound OK before doing putting the LCD on. I checked the sound and there was nothing coming through the filter board, although when turning it off there’s a blip like the one my SQ80 used to do, so I’m guessing the filter is at least somewhat working. Then I tried probing the oscillator output, and that was working fine, including via MIDI.

I thought maybe I just needed to adjust the cutoff trimpot, and I wanted to be able to hear the sound while I adjusted it, so I tried to connect the two using the ISP cables included in the kit. But I have my “switch” cables connected with an alligator clip, and it bumped something on the control board.

I fear that I noticed it too late. I unplugged it and plugged it back in, tried probing the oscillators again, and got this (kind of cool) nasty XOR-type sound, but it wouldn’t respond to MIDI. I unplugged and plugged it in again, and now it’s not doing anything. It’s not responding to MIDI or any kind of button/encoder action. LEDs 3 and 7 are lit, and sometimes 7 and 8 will light up together seemingly randomly.

Also, all LEDs light up when it powers up, which it didn’t seem to be doing before. And the 7805 is very hot, which I didn’t notice before.

Do I need to order a new Atmega644P? :frowning:

Also I just noticed that the LEDs slowly dim. Did I fry a capacitor on the control board?

  1. First, disconnect the two boards and check for shorts between GND and +5V.
  2. All LEDs being lit up for a short amount of time at boot is normal.
  3. You don’t need to solder the LCD to the board to test it. It takes a bit of practice to do this trick, but if you solder the 16 pins connector to the LCD, and then insert this on the board, and bend it a bit to establish a connection between the pins and the board, you can boot the Shruthi with the LCD without soldering it. Adjust the contrast trimmer until you see something on the screen. If you’re seeing a line of squares, it’s likely that you have damaged the ATMega.

> Also I just noticed that the LEDs slowly dim. Did I fry a capacitor on the control board?

You mean after some time the brightness goes down? Doesn’t look good. There might something really wrong on the power supply then (maybe wrong enough to trigger the brown out detection on the ATMega). Do you something to check how stable the 5V rail is?

OK now the LEDs are lighting up with the correct timing, but it’s a seemingly random arrangement of LEDs. The dimming definitely gives me a feeling that something analog is wrong.

I think I will just go ahead and order of each of the ICs on the digital control board.

Is there anything else on the board that would have been hurt from getting touched by the wire going to the switch? If so, I’ll order that too. Hopefully the answer isn’t “everything”… :frowning:

The most fragile thing on the board in the ATMega644p.

> You mean after some time the brightness goes down? Doesn’t look good.
Yes :frowning:

> There might something really wrong on the power supply then (maybe wrong enough to trigger the brown out detection on the ATMega). Do you something to check how stable the 5V rail is?

Do you mean like the first tests on the control board before inserting the ICs? I will give that shot. Would the 7805 getting hot be symptomatic of those kinds of problems?

> Would the 7805 getting hot be symptomatic of those kinds of problems?

It can run quite hot on a correctly running unit too. Yes, check the voltage of the +5V rail, and check for shorts between the power rails and ground.

I just ran those tests again. The ground and +5 points are working fine, but the -5 points are at +0.68 volts! I will double check my solder work. I was getting about -11 before, and then I fixed a couple joints and it was back to about -5. Maybe something got loose somehow.

Also now that the digital board is disconnected, the 7805 is not hot, barely even lukewarm.

I can’t find any bad joints. The negative voltage is still reading +0.68. What could be causing this?

Maybe a damaged LT1054.

Wow, I am dumb. I didn’t even have the LT1054 socketed. Now all the points are working correctly.

Since I’m pretty sure I broke some stuff on the digital filter board, I thought I would try to use the SMR4 board as a standalone filter for now, as described here: http://mutable-instruments.net/forum/comments.php?DiscussionID=341 But I’m not getting any sound except pops when I turn it off and on. In another embarrassing noob mistake, I plugged in the LM13700 in IC7 backwards when I first turned it on. Would that have damaged it? Other than that, the trim pots don’t seem to have any effect.

I’m going to put in a Digikey order this week for the Atmega644p and I’m trying to figure out what else I might have damaged. I might need to order an LM13700 too, and some of the other ICs on the digital board.

Sorry I am so dumb. I probably should have tried to build something simpler first.

I probably should have asked here before I placed my order, but is it possible to program the Atmega644p using an Arduino Uno? It sounds like Programmer2 or Mega ISP will do it, but then I read this:

> The main difference between the Shruti-1 and Shruthi-1 bootloaders is that the Shruti-1 is Arduino compatible (update by serial port using the STK500 protocol), while the Shruthi-1 is not.

I guess the clock differences might be problematic too? Hmm.

I have never used an arduino to bit-bang the STK500 or ISP protocol, so I don’t know how well it works… But just because the chip is clocked with a 20 MHz crystal doesn’t mean that it needs to be programmed (as in: transfer of program data from a computer to the MCU’s flash) at 20 MHz.

Ah, OK. I will see how it goes. I might have to grab a real programmer too.

Thanks for all your help so far.

FYI/FWIW I actually can get sound out of the OSC output (still no filter joy) using MIDI, but it stops working - clicking/popping as though the oscillator speed suddenly dropped very low and then silence - if there are more than a couple of notes pressed. CC1 (modwheel) works, but I tried a few of the other CCs and they give the same result as multiple notes pressed.

Also I could not get the LCD working, and only button 6 works (no encoder action or potentiometers either, as far as I can tell), and only if the board hasn’t been turned on for more than a few seconds.

So it’s possible that there’s some other components broken, like the shift registers, but it’s probably the Atmega644p. I guess I’ll see.

Alright, I got my replacement parts in today. I haven’t programmed the MCU yet but I thought it would be straightforward enough to get the analog board working.

I bought 3 new TL074s and 3 new LM13700s, so all the ICs in the analog board (except the LT1054, which seems to be working because all the voltages are correct) are fresh. I definitely got the orientation right this time, and I don’t see any cold solder joints.

However, I’m still getting nothing but a square-ish pop on power up when I follow these instructions to use the board as a standalone filter: http://mutable-instruments.net/forum/comments.php?DiscussionID=341

I noticed when I had the digital board connected before when it was semi-working that on power down, I would get a slightly more interesting pop. So it seemed like something was happening with the filter. No such luck now.

I don’t see any cold solder joints. I am, however, confused about the trim pots. I’ve turned the “Lowwww” (I’m assuming that’s cutoff?) all over the place and I’m not getting any sound. Can you offer any suggestions on those or anything else that needs to be adjusted to get the filter working standalone?

Alright, I just realized I made another stupid mistake - I forgot to bridge R1 and R2 on the input/output jacks. Done - BUT… that didn’t solve the problem yet. Still no filtering of audio input. Now I just have noise (quite a lot of noise, actually) and a pop on power off now in addition to the pop I had earlier on power down.

Any ideas?

EDIT: Also I just noticed that there’s kind of a quiet, gravelly rumbling when I turn the resonance pot, but not the cutoff pot.

EDIT: When the power is off, I hear a distorted, highpass-filtered version of the input at the output jack.

Have you tried tracing the signal through the board to see where the chain is broken?

I haven’t. I’m not exactly sure how to do that.

Should I use the image of the PCB on the SMR-4 page? This is confusing to me because early in the path it looks like the signal goes from amp 4 / pin 14 of IC3 to the reso trimpot, which doesn’t make sense to me.
Should I do this with power to the board?
Should I use my multimeter or audio probes?

If these are stupid questions, could you give me a link to a page describing the process, or tell me what to google for? Thank you.

> Should I use the image of the PCB on the SMR-4 page? This is confusing to me because early in the path it looks like the signal goes from amp 4 / pin 14 of IC3 to the reso trimpot, which doesn’t make sense to me.

You can, or use the points highlighted on the troubleshooting page: http://mutable-instruments.net/shruthi1/build/troubleshooting

There’s nothing confusing. The signal doesn’t go from IC3:14 to the reso trimpot… The output of the op-amp and the output of the resonance VCA (through the resonance adjustment trimmer) are both fed into the first filter stage (pin 3 of the first OTA).

> Should I do this with power to the board?

Of course. How the signal will flow through the board if none of the IC are powered?

> Should I use my multimeter or audio probes?

The best if to take a basic mono audio cable, cut it, connect the ring to the ground of the Shruthi-1 filter board, and connect the tip to the point of the circuit you want to probe. The jack at the other side of the cable connected to your mixer/speakers.