MB-6582 Partial Success with Testtone

MIDIbox SID
1

Hi all,

I’m having some problems passing the testtone stage. Maybe it’s coincidence but there is some symmetry that might be worth reporting. No SIDs in at the moment, just jumpering pins 8 and 27.

SID 1: completely silent, SID 2 - loud and clear. SID 3 - loud and clear, SID 4: completely silent

SID 5: quiet but audible (“1/2 strength”), SID 6 - loud and clear. SID 7 - loud and clear, SID 8: quiet but audible (“1/2 strength”)

Note my numbering system of SIDs is straight right-to-left, i.e., “SID1” above is J2_SID1 whereas “SID 2” is J1_SID1.

I checked for bad soldering and shorts. Was unable to see any shorts. Redid a few solder joints that looked crummy. Will try one more time tomorrow but think I got the big ones and these were all in the lower part of the board possibly under currently unused jumpers etc.

There is a thread here that mentioned checking the 9V jumpers on J1+J2_SID1 et al. These have connectivity/are jumpered. J1+J2 are connected to J73 which has pins 2&3 bridged for PSU A. J4 has no bridges. J71 has 2&3 jumpered.

<___base_url___>/index.php?/topic/11293-mb-6582-very-low-output-and-hanging-notes-sounds-familiar/page__p__97798&do=embed&embedComment=97798&embedDo=findComment

Another thread mentioned the amp circuit being worth checking/testing. I’m having a tough time locating the amp circuits on the R2 schematics and I’m unsure how to test it when (if!) I find it.

I tried to get started on the SID interconnection tests. However, I am not sure which pins are A0, A1, etc. Just for fun I uploaded the app to Core 1 and used at the default pin A0 on “dead†SID 1. I got +5V on pins 8 and 25. I got +9V on pin 28.

I am using PSU Option A. LCD powers up. I passed all the previous checks about testing for +5V in various places. I read everything on the “6582 testtone” + “sid testtone” searches to no avail. Thanks!

2

To start with, for the ones that are completely silent, you can try jumpering pin 8 of a known good SID socket with pin 27 of one you want to test. Also do the inverse (pin 8 of a “silent” SID socket to pin 27 of a “good” SID socket).

This will help you validate that all the PIC->SID socket tracks are good and the problem really is in the audio buffers for the SID socket.

It might also be a good idea to connect your amp or whatever to the J23_SIDx headers and validate the audio sockets are OK.

Alternately, you could use the testtone app and connect SID socket pin 8 to the OUT pin of J23_SIDx with a 1K resistor (to be safe) and test the audio sockets that way… I haven’t ever tried this but it should output a 1khz tone but probably not the same volume as through the audio buffer.

It’s possible that you’ve fried a few transistors, or maybe just didn’t install components correctly. Check the polarity of all the electrolytic capacitors in those areas, and that the resistors are correct values.

3

Hi Wilba,

Thanks for the suggestions. Here’s some findings.

For good Sid 2 pin 8 to bad Sid 1 pin 27, it’s fine.

For bad Sid 1 pin 8 to good Sid 2 pin 27, nothing. Guessing bad input to Sid1 somewhere.

For good Sid 3 pin 8 to bad Sid 4 pin 27, it’s fine except that the stereo channel switched over as compared to Sid 3 by itself.

For bad Sid 4 pin 8 to good Sid 3 pin 27, nothing. Bad input again(?)

For good Sid 5 pin 8 to bad Sid 6 pin 27, it’s 1/2 volume. Bad output(?)

For bad Sid 6 pin 8 to good Sid 5 pin 27, it’s fine.

Occasionally when I test this it’s full volume for a split second and then goes down to half volume.

For good Sid 7 pin 8 to bad Sid 8 pin 27, it’s fine except stereo channel switched.

For bad Sid 8 pin 8 to good Sid 7 pin 27, nothing. Bad input to Sid8(?)

Resistors check out (R70-R77, R2’s, R3’s, R22’s, R23’s, R4’s, R24’s) and all caps in the area have correct polarity.

I was able to follow the bad output track and check connectivity all the way through to J23B and on to the audio output.

It looks like I don’t have any 1K resistors so I will have to go to the store and get back on that one. I have some 10K ones - wasn’t sure if this was going to mess something up.

4

OK - I tried the 1K resisitor test. It replicates what I got before. Sids 1, 4 & 8 when connected to their J3s/J23s make no output and Sid 6 makes half output. The rest generate normal testtone output.

5

Note my numbering system of SIDs is straight right-to-left, i.e., “SID1” above is J2_SID1 whereas “SID 2” is J1_SID1.

Let’s start using actual part labels on the PCB… right-to-left this is U2_SID1, U1_SID1, U2_SID2, U1_SID2 etc.

Thus there are four groups of two SIDs (right-to-left), U1_SIDx is the LEFT SID and outputs on the LEFT channel, U2_SIDx is the RIGHT SID and outputs on the RIGHT channel.

Each pair of SIDs is connected to the PIC (U1_COREx)

U1_COREx pin 24 connects to U1_SIDx pin 8.

U1_COREx pin 27 connects to U2_SIDx pin 8.

For good Sid 2 pin 8 to bad Sid 1 pin 27, it’s fine.

For bad Sid 1 pin 8 to good Sid 2 pin 27, nothing. Guessing bad input to Sid1 somewhere.

This suggests something wrong between U1_CORE1 pin 27 and U2_SID1 pin 8.

For good Sid 3 pin 8 to bad Sid 4 pin 27, it’s fine except that the stereo channel switched over as compared to Sid 3 by itself.

For bad Sid 4 pin 8 to good Sid 3 pin 27, nothing. Bad input again(?)

“stereo channel switched over” doesn’t make sense. The output of “SID 4” aka U1_SID2 will always be the left channel. The output of “SID 3” aka U2_SID2 will always be the right channel.

This suggests something wrong between U1_CORE2 pin 24 and U1_SID2 pin 8.

For good Sid 5 pin 8 to bad Sid 6 pin 27, it’s 1/2 volume. Bad output(?)

For bad Sid 6 pin 8 to good Sid 5 pin 27, it’s fine.

Occasionally when I test this it’s full volume for a split second and then goes down to half volume.

This doesn’t match the earlier statement “SID 5: quiet but audible (“1/2 strength”), SID 6 - loud and clear.”

Let’s assume you just mixed up the good/bad labels.

This suggests something wrong between U1_CORE3 pin 27 and U2_SID3 pin 8.

For good Sid 7 pin 8 to bad Sid 8 pin 27, it’s fine except stereo channel switched.

For bad Sid 8 pin 8 to good Sid 7 pin 27, nothing. Bad input to Sid8(?)

This doesn’t match the earlier statement “SID 7 - loud and clear, SID 8: quiet but audible (“1/2 strength”)”

This suggests something wrong between U1_CORE4 pin 24 and U1_SID4 pin 8.

I think you should focus on just one SID pair for now and fix that, it’s getting a bit confusing trying to work out what is and isn’t working when you lump all of the tests into one report.

eg for the first two SIDs (U1_SID1, U2_SID1), it seems the audio buffers are OK, pin 27 on each socket to the audio jacks seems OK, so it’s something wrong before there.

6

Hi Wilba,

Yes, you’re right about my getting the 5 and 6 labels flipped the first time around. Anyway, sorry for the confusion and I’ll use the PCB labels. Also, about the stereo – sorry for the bad description. I meant what you said re: the output of “SID 4” aka U1_SID2 will always be the left channel. Just ignore this if necessary.

This suggests something wrong between U1_CORE1 pin 27 and U2_SID1 pin 8.

I’ll work on this first. I looked at the diagram and it appears there is nothing in the way between these two in terms of connectivity. So then I tested connectivity and it’s there. I did the test by touching the pin on the PIC and the socket of U2_SID1 to rule out bad soldering. Also U1_CORE1 pin 24 is connected to U1_SID1 pin 8.

* “This suggests something wrong between U1_CORE2 pin 24 and U1_SID2 pin 8.

This is connected.

So I’m thinking it’s the PICs. By the way, PIC3 -> Sids ‘5’ and ‘6’ are now fine (not sure why but it’s an improvement). I went back to MIOS V 1.9F as I had upgraded them to V 1.9G. Didn’t change anything.

To really test the PICs, I removed PIC3. I put PIC 1 where PIC 3 was (i.e., the formerly good spot). I also out PIC 4 in PIC 1’s spot.

PIC 4 gives good output on U2_SID1 (a.k.a. ‘Sid 1’) but nothing on U1_SID1 (a.ka. ‘Sid 2’). This is the pattern it exhibited in its original spot and moreover is the reverse of what the previous problem was (namely bad ‘Sid 1’, good ‘Sid 2’).

PIC 1 gives nothing on U2_SID3 (‘Sid 5’) and is fine on U1_SID3 (‘Sid 6’). This is also consistent with its behavior in its original socket and inconsistent with the previous Ux_SID3 results.

To finish the test, I put PIC 2 in the original PIC 1 socket, and PIC 3 in the original PIC 2 socket. (PIC 1 went to the PIC 3 spot and PIC 4 went back to it’s original PIC 4 spot, but that was immaterial here). PIC 2 sounded on U2_SID1 but not U1_SID1, which again puts the spotlight on the PIC as opposed to Ux_SID1. Good PIC 3 did fine on both channels of Ux_SID2, again corroborating all the above.

So I guess new PICs are in order(?) Is there any way of reviving them? Kind of weird that 3/4 of them went out somehow. I remember doing de-static on my hands before working with them, but guess that wasn’t good enough.

7

I do not know if this could be related but I found myself in the same situation this week end.

I fried a PIC by accidentally touching pin 28 (SID socket) while bridging pin 8 & 27 to perfrom the test tone. Pin 28 is 9/12V. As a result I got no more sound on one channel and half volume on the other one. Also I can still upload mios and apps to this PIC but it shows strange behaviors (mios locked in rebooting loop, messed up characters and the like…).

8

Ouch!

That’s why I always recommend using a 1K resistor to bridge pins, which should limit current in case of accidents like that.

Kyo: just out of interest, did you use a resistor and the PIC still got damaged? or was it a piece of wire? :tongue:

9

I’m almost positive I got socket 28 too at some point. And sockets 8 & 9 at some point.

The weird thing is that PIC 3 came back. Kyo, were you able to figure out anything to revive the PIC? It would be great if I could avoid buying 3 more PICs.

10

No sowwy. I reuploaded boot loader from scratch then mios & app and still no sound and corrupted behaviors.

11

(Make that pins *7* & 9)

12

Bummer…

Well anyway, for the next person who reads this thread, what I could have done (and will do after another Smash run) was put electrical tape/painters tape on the sockets making sure to leave only the relevant places uncovered. That should cut down on problems.

13

FWIW this was resolved by replacing the three bad cores with new ones from Smash. Fortunately Core 3, which was a little sketchy there initially, appears to be doing ok for now. Electrical tape on the socket worked out also as a failsafe.

14

FWIW this was resolved by replacing the three bad cores with new ones from Smash. Fortunately Core 3, which was a little sketchy there initially, appears to be doing ok for now. Electrical tape on the socket worked out also as a failsafe.

Damn It seems like I am having the same problem

testtone app no sids

U1 SID 1 no sound

U2 SID 1 good

U1 SID 2 good

U2 SID 2 good

U1 SID 3 no sound

U2 SID 3 good

U1 SID 4 good

U2 SID 4 good

however, with 6582a’s installed through the test tone application i get

U1 SID 1 good

U2 SID 1 no sound

U1 SID 2 good

U2 SID 2 good

U1 SID 3 really quite

U2 SID 3 no sound

U1 SID 4 really quite

U2 SID 4 no sound

I have really messy joint on R4 SID3 . actually there is a missing patch on the bottom side of the pcb, but it’s not the leg going to T1 SID3 and C4 Sid 3

On U1 SID1 I have similar problem but this time with C4 SID1.

It thought they were the problem but after reading this point I am more and more convinced that the 3 PIC’s (00; 02; 03) out of four are my problem (and of course my poor soldering skills).

Could you or somebody please suggest a possible troubleshooting sequence before I spend any money on new PIC’s?

Thanks!

15

I can’t see how you can get “U1 SID 1 no sound” in the first test (without SID) and then “U1 SID 1 good” in the second test (with SID).

Try the “good” PIC (01) in each slot, and two “good” SIDs, do both tests again. Swap SIDs as well to prove it’s not the SID.

Repeat with another PIC, if the results differ, then the PIC is the problem, on top of other problems.