Problem with firmware getting erased by random!!!

[denis]

I did not want to open a new topic for a similar topic so I'm using slayer666's topic.

Along time ago I have ordered two kits (20mhz v2.4) from synthex one for me and an other for my friend. As you know a mounth ago I had firmware erase problem with one of them, reprogramed it and fixed the problem by your help. Then I completed my drumkit with the other kit (with the one working without a problem).

Anyway Today I gave the working box to my friend and set the other board that I reflashed a mounth ago into a enclouser also replaced the max's with new 74hc4851 chips (I ordered them form synthex inorder to solve the overspill problem) and headache started again.

Module is recognised by the computer truely but while trying to set the settings for the kit my pc stopped receiving midi signal so I disconnected and plugged it again. But it's strange that when pc stops to receive midi megadrum seems working ok, midi led is flashing when I hit a pad. This problem happened a few times and then it was opened with firmware update message as it was happened a mounth ago. ( I was not using this board since that time)

1- I build two board and one of the board has no problem other then overspill. Same pc with two megadrums one is having problems other don't and they are same kits without soldering problems. Other atmega was used with this board for a while without a problem also.

Do I really need to do pullup resistor modification? Or this problem is related with a problematic atmega?


2- -edi- overspill problem is solved wiht new ics.

3- Please help I'm working on this edrum for two mounths and getting bored now. Kit looks fantastic now and I want to finish it as soon as possible.

Do I need a pullup resistor or is it better to order a new atmega. ( Huh this module cost me alot )

[Synthex]

1- I build two board and one of the board has no problem other then overspill. Same pc with two megadrums one is having problems other don't and they are same kits without soldering problems. Other atmega was used with this board for a while without a problem also.
Do I really need to do pullup resistor modification? Or this problem is related with a problematic atmega?

You can add these resistors.
The new v2.7 PCB design have these pullup resistors.

2- With my short experience with 74hc4851 chips, It seems they reduce but don't solve the overspill issue. My piezzo are pad are functioning god with Admirs edrum and roplad pm16 but when it comes to megadrum they are too hot.
Isn't it possible to compress or limit the over input voltage to multiplexers? In my opinion board needs a modification in this way.

I have never had problem of overspilling with the 74HC4851. Are you sure of your results??
If your pad are really too hot, you can add a simple 1K (or less) resistor between GND and the input (or 2 resistors to make a divisor of tension).

I think than the board not needs a modification in this way, it really depend of the type of pad used !
To add this type of component on the PCB would take many place, while it is easy to add it on connector (for example),
for only the persons who have the problem.
But, I have never had problem of overspilling with the 74HC4851.

[denis]

Ok then I'll add the resistors. But it's still a quesiton that why oneof the boards have this problem while the other one doesn't. Isn't it pointless?

[Synthex]

Ok then I'll add the resistors. But it's still a quesiton that why oneof the boards have this problem while the other one doesn't. Isn't it pointless?

Of course !

I think it is parasite's problem on high impedance inputs of the atmega.
Adding pullup resistors is the best.

[denis]

Ok 47k resistors from pins 6,7,8 and 9 to pin 10 is the modification needed wright?

[Synthex]

Ok 47k resistors from pins 6,7,8 and 9 to pin 10 is the modification needed wright?

Yes, 47k resistors from pins 6,7,8 and 9 to pin 10 (Vcc).

[slayer666]

Ok 47k resistors from pins 6,7,8 and 9 to pin 10 is the modification needed wright?

Yes, 47k resistors from pins 6,7,8 and 9 to pin 10 (Vcc).

I'm experiencing "lazy buttons" on my mega. It's like the unit doesn't register every time I push a button.
I'm thinking 47k pullup is too high?
Could this be the case?

Maybe 5-10k is better?

Edit: I bought new buttons that "click". My lazy button problem seems to be gone.

[slayer666]

Module is recognised by the computer truely but while trying to set the settings for the kit my pc stopped receiving midi signal so I disconnected and plugged it again. But it's strange that when pc stops to receive midi megadrum seems working ok, midi led is flashing when I hit a pad. This problem happened a few times and then it was opened with firmware update message as it was happened a mounth ago. ( I was not using this board since that time)

I had this problem when my megadrum was connected to my PC through a USB-hub. The problem dissapperared when I skipped the USB-hub.

[denis]

I'm not using a usb hub. There is only 2 usb ports on my laptop and one goes for the sound card while the other one for megadrum.

[slayer666]

I'm not using a usb hub. There is only 2 usb ports on my laptop and one goes for the sound card while the other one for megadrum.

ok. But the problem is definitely Usb related. I tried connecting both usb and midi out, and when usb midi stopped working i switched midi interface in cubase without restarting the megadrum and the megadrum was still sending midi to cubase.