www.megadrum.info

Hi everyone!

I searched this forum but didn't find the topic I need.
I have found that some guys has problems with usb connection too (I have done all recomendations according to this topic), but I think I have another problem, may be related to interference with other electrical devices.
I assembled my megadrum according to "Latest MegaDrum schematics. 32 inputs".

The megadrum is detected, all is ok, all is working, i was really happy. But after some time usb midi signal is lost. The solution is simple: close Nuendo, switch off and on usb cable, restart Nuendo - all works. You all understand that this is very bad solution.
Some facts:
- no matter is it Nuendo or any other program (FL studio), all are the same;
- midi activity led is working, lcd display shows correct signals;
- the moment of usb midi signal losting is unknown, but it's clear that plugging off-on the fridge will direcly cause the problem (plugging off-on of all other electrical devices also can cause the problem, but fridge is most stable);
- when I switch off-on the light or other electrical devices, megadrum occasionly produce strong signals (may be 5-6 drums, may be all drums ex. hihat pedal), when all peds unpluged, no false signals;
- when I unplug all peds (40pin connector), during "fridge test" no signal lost was happen;
- I have read all this topic and all forum and indeed I didn't find the answer;
Sorry for my English

So do I have to change my flat or is there a more simple solution?
Hope you will help me

1. Add more decoupling capacitors close to Atmega, PIC18F2550 and multiplexers
2. Add a 10uF capacitor between Vcc and GND close to USB.
3. Look at this http://www.megadrum.info/forums/viewtopic.php?f=3&t=268 . It may be related.

1. I soldered 100nF capacitors between Vcc and GND on multiplexors. No results. Punctually, no changes.
2. Done. First think I done I soldered 10uF capacitor as close to usb as I could. And no result.
3. I have already read that topic. It is close, but I think not related. I have done all recomendations posted there.

I suggest that pads are catching some electromagnetic waves produced when switching on-off lamps or old fridge. I tested precisely all electric devices I have: iron has no effect, 2kw heater (installed close to pad) too, soo only high-frequency noise is a problem (but old fridge - ???). On wikipedia I read that piezo elements are insensible to electromagnetic fields and radiation... Just no ideas!

maks wrote:1. I soldered 100nF capacitors between Vcc and GND on multiplexors. No results. Punctually, no changes.
2. Done. First think I done I soldered 10uF capacitor as close to usb as I could. And no result.
3. I have already read that topic. It is close, but I think not related. I have done all recomendations posted there.

These suggestions were to address the problem with loosing USB connectivity. Is the problem still there?

I suggest that pads are catching some electromagnetic waves produced when switching on-off lamps or old fridge. I tested precisely all electric devices I have: iron has no effect, 2kw heater (installed close to pad) too, soo only high-frequency noise is a problem (but old fridge - ???). On wikipedia I read that piezo elements are insensible to electromagnetic fields and radiation... Just no ideas!

I think it'll be very hard to eliminate this problem. Even my Toshiba digital TV looses picture/sound when a fridge turns on/off and becomes totally unwatchable when a kitchen mixer is being used.

Sounds like you might want to investigate "power conditioners". If your appliances are really noisy, you're always going to have problems. Unfortunately, there's a lot of hype and BS in this market so it's hard to know what you're really getting unless you do a bunch of research yourself.

In the audio world, Furman is a respected brand that a lot of studios install. Comes with a "respected" price tag though. See http://www.furmansound.com/

Unplug the sensors (pads) and put your brain in a Farady cage.

A shoe box with a tiny hole for the (quality) USB lead, then wrapped in foil earthed to the radiator should eliminate any bad vibes

Maybe try different cables between pads and MD... Hook up one input at a time, see if its one input or all.
Just my .02 //

Brian

Talk about non-trivial problems. I wonder how many of you MD users experienced this rather unpleasant inconvenience?

I get USB connection lost upon, for example, turning off my soldering station. And, of course, random MIDI activity on unconnected channels at the same time (or even when nothing is connected to the input).

The problem doesn't solved yet.
When I use MIDI connection (to sequencer), the connection is ok during "fridge test", but random signals are heard. So, I guess, problem is in usb pic controller.
But using only midi connection isn't a good solution, because modern sound cards with midi ports are quite expensive.
So, any advices?

So you don't loose the USB connection any more but still get random signals due to electromagnetic noises, right?