Roland Supercube 100 -- input jack advice needed (probably)

[Enzo]

I think you misunderstood me.  I have zero problem with the suggested ways to fix it.  I was concerned about this line:

avoid the recurrence of problems like this.

We should do whatever it takes to FIX it, but it doesn't need reengineering, because recurrence is not really an issue.   Too many times an OP will see any failure as a design flaw.  "My 12AX7 got noisy, how can I prevent that in the future?"

[niftyprose]

I will try contact cleaner before I go in with the wirecutters, Enzo, honest. But what I'm hearing suggests that contact cleaner won't be enough. Could someone review my previous ;post and see if I'm talking rubbish? Best, NP

[phatt]

Best I can make out is R101 was never installed hence the unused pads between the inputs can be seen in the pic. the amp will work without it and may even add a little more signal to pass.

I've edited the input schematic of the Cube 60 for reference.
sw1-sw4 are the internal switches.

How it works, maybe?
With no plug in either input sw1 grounds out R101 270k which mutes the normal input while sw2 grounds the output of Q1 at (point X)
with *Normal input* in use sw4 opens and signal passes through R10 with R9 running to (point X) which is still grounded via sw2.

With *OD input* in use, sw1 and sw2 open.
sw1 lifts R101 off ground which adds to the value of R10 forming a voltage divider at the input of C6.

sw2 removes the mute at point X allowing signal to pass throuh R9.
As sw4 is then closed it forms a voltage divider out of r10 and R101.

I'm not sure about the extra internal tab (the red ? mark)
I can only assume it contacts the sleve of the plug ,, which seems pointless as the sleeve is already grounded by the threaded metal socket???
unless maybe the metal thread uses a plastic isolation washer???

Hey nifty a Q, are you saying that the Od input and the Boost Switch are on that PCB? If so then my wire numbers I mentioned earlier are wrong.
 
I'm still trying to decipher Joe's idea but thought a drawing and explanation might help.
Phil.

[Enzo]

I think your description works.  As to the , The metal bushing on a TRS jack does not short the ring.  That is a grounded ring contact, so if you plug a TRS into the jack, it grounds the ring.  Say a balanced signal.  If you plug a common TS plug in, then of course the ring contact is redundant.

The jack is poorly drawn.  Usually we expect the tip to be farther from the bushing than ring, but here the positions seem reversed.

[niftyprose]

Enzo and particularly Phil, thank you so much! I had figured that something like Phil's analysis would apply, but I never actually had to anatomize such a circuit before and needed some guidance. Right...

I applied contact cleaner to a 1/4" plug and worked it in and out of both inputs, then did a partial reassembly on the workbench. Amp to speaker, amp to reverb tank, jacks back in speaker chassis. I kind of hoped that would sort things out, but actually the problem is worse than before. The jacks are really noisy, and wiggling the plug creates lots of full-volume speaker abuse. I think this is because the sub-board on which they're mounted has been shaken up so much over the last few days.
=>
I think I have to go ahead and disconnect the sub-board from the main amp board, install a new input jack, and make Joe's switch-mounted-in-spare-jack-hole thing work.

It will take me 24hrs to fully digest Phil's breakdown (like a python swallowing a goat). But it sounds to me like I can put a DPDT switch into the unused jack drilling, and use it to route the signal from the guitar to either 1&4 or 2&3 on Phil's diagram (which may or may not be labelled that way in real life, but we'll cross that bridge when we come to it). I'm concerned that I may be omitting to earth something somewhere!

In answer to Phil's question: "Hey nifty a Q, are you saying that the Od input and the Boost Switch are on that PCB? If so then my wire numbers I mentioned earlier are wrong." -- I'll do my best to give a straight answer. The 1*3" PCB sub-board that appears on my first gutshots carries both input jacks, a five-cable ribbon, and separate black and white leads to the overdrive footswitch. The five cables in the ribbon *must* correspond to the inputs in your diagram, surely, and they're numbered 1,2,3,4,5 where they join the main board. The time I spent with the multimeter and trying to work out the PCB gave me the distinct impression that those number assignments are different from those in the schematic (remember this is a Supercube 100, not the Cube 60 from the schemo). I think they're just reversed. I can presumably confirm that by tracking where the connections go to on the main board. Give me time.

One small point from my previous note that didn't get answered. If I'm joining the cables from that five-way ribbon connector to the new DPDT, I will need either to pare them apart with a craft knife (messy) or lug them into some kind of breakout board. Any recommendations?

Best NP

[joecool85]

In looking at the schematic again, it is either wrong, or very poorly drawn.  It shows on the Normal circuit that the tip goes to ground (tip goes to pin 2, pin 2 is tied with 1, 2, 3, 4, and 5, and sent to ground right next to pin 3.  There are some dashes in this line connecting them, which may indicate that under curtain circumstance (unplugged lead etc) they are not all connected.  Regardless, as drawn I don't see how the Normal channel could pass signal at all since it is sent straight to ground.  I think that this may look more complicated than it is.

Here is my proposed solution:

Use a mono jack.  Send sleeve to ground.  Send tip to DPDT switch.  On this switch it will either send it to R1 for Overdrive or pin 2 (I think) for the Normal channel.  On the other side of the DPDT you can ground out the opposing channel.  IE: If switched to Overdrive, send Normal channel's input to ground via 270k resistor (R101 on schematic).  If switched to Normal, send Overdrive channel input to ground (R1).

[niftyprose]

Wow! Joe, I think you did a bunch of my 'digestive' work for me. I have to wait until the DPDT switch and replacement jack get here, so I wlll consider carefully and may come back with further queries. Thanks again, guys. NP

[Enzo]

Joe, step back and look again.  That line "connecting" 1,2,3,4,5 and has the dashes in it is not a trace, it is the border of the board.  It is not a conductor.

Ignore the dotted line, and reread phatt's description.

If you follow the dotted line around it would also"short" all three legs of Q11, or whatever transistor is at pins 6,7,8.  Clearly it is not a circuit trace.

[joecool85]

Joe, step back and look again.  That line "connecting" 1,2,3,4,5 and has the dashes in it is not a trace, it is the border of the board.  It is not a conductor.

Ignore the dotted line, and reread phatt's description.

If you follow the dotted line around it would also"short" all three legs of Q11, or whatever transistor is at pins 6,7,8.  Clearly it is not a circuit trace.

Wow, very poorly drawn indeed.  But you are right, definitely a PCB edge.  Confusing since each pin label has a circle, typically representative of a node/connection.

The good news, my plan still works exactly as described in my previous post 

[Enzo]

Those are open circles unlike the solid dots of joints within the schematic.  Reminds me of the drawings for FX pedals, considering it is a Roland and so are Boss pedals, I guess that makes sense.

[phatt]

Obviously the top OD input SKT has failed and ideally you need to replace that socket but as it might be hard to source the exact part then here is an option using only one input socket,, probably similar to what *Joe* mentioned.
Now as it's possible the input for Supercube 100 might be different to the cube 60 schem I'm working from I can't be sure this would work so you will have to tread careful and double check.
This requires an N/C switched Socket and a DPDT switch to work.
N/C means Normally Closed,, opens when you insert.

As to your Q re those ribbon wires, you can just gently use your fingernails to split them, might have to nic between them to get it started but they should seperate fairly easy.
If you need more length then use a piece of strip board and then add more wires.
With out a pic of how the ribbon connects to PCB it's hard to give advice.
Hope it helps, Phil.

[joecool85]

Obviously the top OD input SKT has failed and ideally you need to replace that socket but as it might be hard to source the exact part then here is an option using only one input socket,, probably similar to what *Joe* mentioned.
Now as it's possible the input for Supercube 100 might be different to the cube 60 schem I'm working from I can't be sure this would work so you will have to tread careful and double check.
This requires an N/C switched Socket and a DPDT switch to work.
N/C means Normally Closed,, opens when you insert.

As to your Q re those ribbon wires, you can just gently use your fingernails to split them, might have to nic between them to get it started but they should seperate fairly easy.
If you need more length then use a piece of strip board and then add more wires.
With out a pic of how the ribbon connects to PCB it's hard to give advice.
Hope it helps, Phil.

Looks good to me, Phil.  Your drawing adds in the switched jack so that when nothing at all is plugged in, both inputs are grounded.  Nice touch.

To niftyprose, the DPDT Phil shows does *not* need any connections between the top pins on left and right.  What he is showing is the switch in the overdrive position.  Also remember that wires traces that cross don't connect unless there is a filled black circle connecting them where they meet.

Good luck!

[niftyprose]

Guys, can't thank you enough. My DPDT switch arrived two days ago but I'm still waiting for the jack, and I have to go and take care of my mum for a few days. Look for an update in about 10 days. Best, NP