Fried IC9?, possibly more issues

Discuss x0x construction and related issues

Moderators: altitude, adafruit_support_bill, adafruit, phono, hamburgers

Please be positive and constructive with your questions and comments.
rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Fried IC9?, possibly more issues

Post by rememberence »

Hi guys!
I've completed my X0xb0x, however i ran into some trouble.

At first, nothing worked after pluging in the power, i did some measurements and found out that my IC9 was put in place the wrong way :oops:

I then decided to disconnect the power and simply flip it around, at first around 30 LEDs was turned on, but no functionality or sound from the b0x. So i disconnected the power again and connected it, nothing, and that's where i am right now.

I'm aware that IC9 might be totally fried, but i tried to measure pin 3 on IC9 and pushing some notes in keyboard mode, but all i got was a constant 0,58v and no action when pushing notes

Is there anyone who can give me some hints or tips on how to proceed with the troubleshooting?

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

Is the digital side working properly? Can you select the different modes; and do the keyboard buttons light up when being pressed in keyboard mode?

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:Is the digital side working properly? Can you select the different modes; and do the keyboard buttons light up when being pressed in keyboard mode?
I haven't really tested the digital side properly, not sure how to honestly.

Nothing happens when selecting different modes, and nothing happens when pressing buttons in any mode.

If it helps, i hear a "pop" when the power is plugged in, and a constant "hum" in my headphones.

Thanks for the quick reply by the way!

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

rememberence wrote: I haven't really tested the digital side properly, not sure how to honestly.
Nothing happens when selecting different modes, and nothing happens when pressing buttons in any mode.
Well, that probably means the digital part isn't working at all. Simple test: Put the xoxbox into "Pattern Sync Out" mode. The Tempo LED should flash; the "white keys" should select patterns (and the associated LED's should light up) and you should be able to select one of the banks.

If neither of this works the processor probably isn't running properly - and then you wouldn't get any useful measurements on IC9 anyway.

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:Well, that probably means the digital part isn't working at all. Simple test: Put the xoxbox into "Pattern Sync Out" mode. The Tempo LED should flash; the "white keys" should select patterns (and the associated LED's should light up) and you should be able to select one of the banks.

If neither of this works the processor probably isn't running properly - and then you wouldn't get any useful measurements on IC9 anyway.
I put it in "Track Sync Out" and got nothing, no lights or anything.

I was thinking i could measure the voltage for all the pins on the ATmega and se if it gets powered properly?
If it does get the correct power, could i assume it's busted?

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

rememberence wrote: I put it in "Track Sync Out" and got nothing, no lights or anything.

I was thinking i could measure the voltage for all the pins on the ATmega and se if it gets powered properly?
If it does get the correct power, could i assume it's busted?
This really depends. I have seen processors where the internal program was somehow corrupted and re-flashing it solved the problem.

The good thing is that the digital part is relatively simple. There are very few components that can be faulty.

First thing to check is if you got +5v on pin 40. Then check if you get the RESET signal on pin 9. And last but not least if there is a clock on pins 18 and 19 (this is very hard to see, even with a good oscilloscope). In fact debugging the digital part is best done with the help of an oscilloscope.

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote: This really depends. I have seen processors where the internal program was somehow corrupted and re-flashing it solved the problem.

The good thing is that the digital part is relatively simple. There are very few components that can be faulty.

First thing to check is if you got +5v on pin 40. Then check if you get the RESET signal on pin 9. And last but not least if there is a clock on pins 18 and 19 (this is very hard to see, even with a good oscilloscope). In fact debugging the digital part is best done with the help of an oscilloscope.
OK!
Thanks a lot for all the info, i've been totally lost :)

I will measure the pins on IC3 to see what could be wrong when i get home from work!

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:First thing to check is if you got +5v on pin 40. Then check if you get the RESET signal on pin 9. And last but not least if there is a clock on pins 18 and 19 (this is very hard to see, even with a good oscilloscope). In fact debugging the digital part is best done with the help of an oscilloscope.
So, i checked pin 40 on IC3 and i only got 3,91v.
Do i even need to check the RESET, XTAL1 or XTAL2?

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

If I'm reading the datasheet correctly the processor should theoretically function with 3.91v.

HOWEVER:


If the digital 5v supply is that far off it hints at a more serious problem (in my opinion). So I wouldn't bother doing anything with the CPU until we find out why the voltage has dropped that far.

Just to be sure measure again. If the voltage is still significantly lower than 5v you may want to separate the IOboard from the Mainboard. Remove the ribbon cable.

With the IOboard powered on, measure the voltage at the innermost pin of IC20 (the pin closest to the ribbon connector; you'll see a trace going from that pin to the ribbon connector).

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:If I'm reading the datasheet correctly the processor should theoretically function with 3.91v.

HOWEVER:


If the digital 5v supply is that far off it hints at a more serious problem (in my opinion). So I wouldn't bother doing anything with the CPU until we find out why the voltage has dropped that far.

Just to be sure measure again. If the voltage is still significantly lower than 5v you may want to separate the IOboard from the Mainboard. Remove the ribbon cable.

With the IOboard powered on, measure the voltage at the innermost pin of IC20 (the pin closest to the ribbon connector; you'll see a trace going from that pin to the ribbon connector).
OK!
First of all, thanks for the reply :)

And second: i measure 4,95v on the pin closest to the ribbon (and 5,04 on j4 itself) so i think the mainboard gets the correct voltage.

Today i was thinking that maybe IC9 might be ruined by my noob-mistake and therefore causing the low voltage?

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

rememberence wrote: And second: i measure 4,95v on the pin closest to the ribbon (and 5,04 on j4 itself) so i think the mainboard gets the correct voltage.
Just to clarify the situation: The xoxbox actually uses 4 different supply voltages:

5.33v
12v (approximately)

Those are present on J4.

5v

This is the digital supply for processor and a few more components; this is going to the mainboard via J3 (the ribbon cable).

6v

This is the supply for the headphone amp. It is connected to the mainboard via J5.

4.95v for the digital supply sounds comparatively close; however since this is generated from a regulator I would personally expect something closer to 5.00v. But it should do for now.
Today i was thinking that maybe IC9 might be ruined by my noob-mistake and therefore causing the low voltage?
If it is ruined or not I don't know; but IC9 is not run from the 5v digital supply (it's supplied from the 5.33 rail).

As far as I know the 5v digital supply serves IC1 to IC8 and IC16 to IC18; it is also used for the various switches.

You may want to measure the 5v supply again with IOboard and mainboard connected, but the processor removed (you can measure from pin 20 to pin 40 on the socket).

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:
rememberence wrote: And second: i measure 4,95v on the pin closest to the ribbon (and 5,04 on j4 itself) so i think the mainboard gets the correct voltage.
Just to clarify the situation: The xoxbox actually uses 4 different supply voltages:

5.33v
12v (approximately)

Those are present on J4.

5v

This is the digital supply for processor and a few more components; this is going to the mainboard via J3 (the ribbon cable).

6v

This is the supply for the headphone amp. It is connected to the mainboard via J5.

4.95v for the digital supply sounds comparatively close; however since this is generated from a regulator I would personally expect something closer to 5.00v. But it should do for now.
Today i was thinking that maybe IC9 might be ruined by my noob-mistake and therefore causing the low voltage?
If it is ruined or not I don't know; but IC9 is not run from the 5v digital supply (it's supplied from the 5.33 rail).

As far as I know the 5v digital supply serves IC1 to IC8 and IC16 to IC18; it is also used for the various switches.

You may want to measure the 5v supply again with IOboard and mainboard connected, but the processor removed (you can measure from pin 20 to pin 40 on the socket).
Hi!
Thanks for the clarification and your patience :)

I measured the following pins

Code: Select all

Pin        Voltage
40         4,41
39         3,58
38         3,59
37-34      0
33         4,76
32         4,33
31-21      0
And, a question, does it matter if i use the ground on ic20? (i've been using it the whole time)

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

rememberence wrote: I measured the following pins

Code: Select all

Pin        Voltage
40         4,41
39         3,58
38         3,59
37-34      0
33         4,76
32         4,33
31-21      0
And, a question, does it matter if i use the ground on ic20? (i've been using it the whole time)
Just to make sure: Those measurements are without the processor installed?

The ground should be (theoretically) the same for the whole system so it shouldn't matter where you obtain it.

rememberence
 
Posts: 9
Joined: Mon May 21, 2012 4:28 pm

Re: Fried IC9?, possibly more issues

Post by rememberence »

MMM wrote:
rememberence wrote: I measured the following pins

Code: Select all

Pin        Voltage
40         4,41
39         3,58
38         3,59
37-34      0
33         4,76
32         4,33
31-21      0
And, a question, does it matter if i use the ground on ic20? (i've been using it the whole time)
Just to make sure: Those measurements are without the processor installed?

The ground should be (theoretically) the same for the whole system so it shouldn't matter where you obtain it.
Yes, these measurements is without the processor installed.

Well, i tried the ground-pin (pin20) and i got totally different voltages.
So i tried to see if ground-pin on J4 is the same as pin20 on IC3 (beeping-mode on my multimeter) and i wont get any beeping.

User avatar
mmm
 
Posts: 517
Joined: Tue Apr 24, 2012 1:51 am

Re: Fried IC9?, possibly more issues

Post by mmm »

rememberence wrote:[
Well, i tried the ground-pin (pin20) and i got totally different voltages.
So i tried to see if ground-pin on J4 is the same as pin20 on IC3 (beeping-mode on my multimeter) and i wont get any beeping.
This is a bit tricky to explain. There *might* be a problem here but since there are so many different things that could go wrong it's not easy to say right now.

The mainboard has two different grounds: "analog" ground and "digital" ground. The "analog" ground is supplied from J4 (together with the "analog" voltages 5.33v and 12v). The digital ground is supplied from J3 (the ribbon cable) together with the "digital" voltage 5v (and various additional signals). Those two grounds are not connected to each other on the mainboard itself.

HOWEVER:


The grounds are tied together on the IOboard.

So with either J4 or J3 NOT connected you will get no continuity between pin 20 of the processor (or processor socket) and the ground pin of J4. But with both J4 and J3 connected you should definitely get continuity.

Locked
Please be positive and constructive with your questions and comments.

Return to “Making x0x”