Membrane keypads again.

Haven't had time to work on this project recently due to illness. I'm back at it.

I'm still trying to figure out the reason that this synthesizer's membrane buttons keep being non-responsive. Again, there's a certain multi-wire header that plugs into the main board where if I tug on it, things function normally. If I let it go slack, the buttons become non-responsive again. I've cleaned all the connections, reflowed the solder joints to the headers but still no luck.

Can anyone suggest a methodical way to isolate where the problem is?

--
----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/
 Click to see the full signature
Reply to
bitrex
Loading thread data ...

Connector in question is 67-74 on the mainboard schematic in the following service manual:

formatting link

--
----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/
 Click to see the full signature
Reply to
bitrex

From your description, it is mechanical. Is there a way, using a dental tool or similar that you might test each wire in the loom as it goes into the connector? I expect you will find the problem there.

Peter Wieck Melrose Park, PA

Reply to
pfjw

Sure. How would the test procedure go precisely?

Reply to
bitrex

I have a suspicion this might not be a mechanical issue. The connector I'm actually yanking on is 34-52, on page 18 of the service manual, labeled "Cartridge Board."

I swapped out a resistor that's connected to ground from the LCD display to increase the contrast, changed from 5.6k to 1k. That resistor is right next to IC20, which is connected to the "Cartridge Board" read/write lines.

I wonder if that bollocked something up, because the synth didn't have this issue prior to the LCD swap.

I don't see any obvious solder bridges but I'll look again...

Reply to
bitrex

Find which wire on the connector is not making a connection. Power off. Attach wire hooks or tack solder some wires to opposite sides of the connector. Attach an ohmmeter. Now, wiggle things around until you determine where things are intermittent.

A close photo of the connector area would be helpful if you want me to make a guess(tm).

I couldn't find 67 - 74 on the main board schematic (Pg 18). That are the row and column coordinates or maybe a large nearby component?

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

I think the problem may not be mechanical, as I've noticed that now the keyboard is having issues triggering as well, which is not connected to the same area.

I think I screwed something up swapping R66 on page 18 near the LCD, and the processor is having trouble with read/writes or something.

Reply to
bitrex

Well, it wouldn't hurt to eliminate the keyboard as a possibility. Put an ohmmeter across a row and column on the keyboard and measure the contact resistance with an ohmmeter. Try all the button combinations. If the resistance seems resonable, move on to looking for another culprit.

5.6K goes to Vo on the LCD display, whatever that does. If it's carbon comp, check the resistance with an ohmmeter. They change over time and absorb moisture. However R66 doesn't seem to go anywhere except the display making changes to keyboard and buttons rather unlikely. Dunno.
--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

R66 controls the contrast of the LCD. I had to swap it out as the new LCD display didn't seem to be 100% compatible, and it was invisible with the stock 5.6k resistor. 1k makes it legible again.

R66 is sitting about half a millimeter away from the deglitching cap for IC20. I'm looking to see if I might have bridged it somehow.

Reply to
bitrex

Press down on each separate wire as it goes into the MPC. A spade-type dental tool is excellent for this purpose. The point being that if the connector holds when under stress, but fails when there is no stress, it is very most likely mechanical.

And, of course, it could be two-or-more things if you, in fact, bridged something replacing a resistor.

Peter Wieck Melrose Park, PA

Reply to
pfjw

I have a suspicion this might not be a mechanical issue. The connector I'm actually yanking on is 34-52, on page 18 of the service manual, labeled "Cartridge Board."

I swapped out a resistor that's connected to ground from the LCD display to increase the contrast, changed from 5.6k to 1k. That resistor is right next to IC20, which is connected to the "Cartridge Board" read/write lines.

I wonder if that bollocked something up, because the synth didn't have this issue prior to the LCD swap.

Hang on, this is the first we have heard of an LCD swap.

Did you get the new LCD from Roland, and/or is it a part that it known to work properly in this keyboard?

I smell a Red Herring.

Gareth.

Reply to
Gareth Magennis

Cool! The dental-spade trick seems to have found a candidate (I used a tiny tiny phillips head screwdriver for manipulating each wire on the header individually.)

Purple wire. When I wiggle that one around the keypad becomes responsive. The others have no effect.

I did some reading and I think the reason the keyboard and the sound engine is behaving goofy is because I swapped the RAM backup battery, and it's become corrupted. The processor has no internal memory, and uses the external battery-backed RAM for both its heap/stack and storing patch data.

There's a startup key combination that instructs the processor to wipe the memory and re-initialize, but of course I can't execute it when the membrane buttons aren't working...: )

Reply to
bitrex

Well, it wouldn't hurt to eliminate the keyboard as a possibility. Put an ohmmeter across a row and column on the keyboard and measure the contact resistance with an ohmmeter. Try all the button combinations. If the resistance seems resonable, move on to looking for another culprit.

5.6K goes to Vo on the LCD display, whatever that does. If it's carbon comp, check the resistance with an ohmmeter. They change over time and absorb moisture. However R66 doesn't seem to go anywhere except the display making changes to keyboard and buttons rather unlikely. Dunno.

R66 is most likely contrast or viewing angle adjustment. It might be a good idea to check all the pullup resistors (RA3) are all present and correct.

It is probably another good idea to just resolder everything around that part of the circuitry, including the Gate Array, just to eliminate dry joints as the problem. That's a 5 minute job at most.

And, if you (the OP) have replaced R66, go look at it again with a loupe or magnifying glass and check you haven't inadvertently shorted something out, or there is maybe some solder debris sitting on the Gate Array pins.

Gareth.

Gareth.

Reply to
Gareth Magennis

Cool! The dental-spade trick seems to have found a candidate (I used a tiny tiny phillips head screwdriver for manipulating each wire on the header individually.)

Purple wire. When I wiggle that one around the keypad becomes responsive. The others have no effect.

I did some reading and I think the reason the keyboard and the sound engine is behaving goofy is because I swapped the RAM backup battery, and it's become corrupted. The processor has no internal memory, and uses the external battery-backed RAM for both its heap/stack and storing patch data.

There's a startup key combination that instructs the processor to wipe the memory and re-initialize, but of course I can't execute it when the membrane buttons aren't working...: )

Here's what I put in my Service Report on an Alpha Juno 2 I repaired January this year:

"Check Battery. Reinitialise RAM. Replace Power Supply capacitor. Resolder all joints on Power Supply PCB. Clean all key contacts and contact PCB"

The system here was corrupted, though the battery was good. You need to do the reinitialise procedure. And have a good look at the Power supply too.

Gareth.

Reply to
Gareth Magennis

Thanks! All power supply filter caps have been replaced with fresh Nichicons; one of them looked like it was starting to get the plague. All the voltages are within tolerances and the HV inverter is putting out around ~90 VAC.

All the keys were removed and the membrane contacts were cleaned. That was a pain in the ass job.

Reply to
bitrex

Thanks! All power supply filter caps have been replaced with fresh Nichicons; one of them looked like it was starting to get the plague. All the voltages are within tolerances and the HV inverter is putting out around ~90 VAC.

All the keys were removed and the membrane contacts were cleaned. That was a pain in the ass job.

OK, so it looks like these are the common ailments of the Alpha Juno 2!

Gareth.

Reply to
Gareth Magennis

Apparently both the EL backlight and the inverter are prone to failure. The inverter looks like a pretty simple blocking oscillator configuration, either the filter cap or transistor can fail. If the transformer fails then you're probably SOL.

The EL material fades out over time.

The inverter on mine was putting out good HV as is, but the EL material had faded. I happened to have some EL material lying around so I fitted that, but in the process the old LCD display basically crumbled as I removed it. Fortunately I found an unlit display module that was pretty much an exact replacement on eBay - 3 for $3. I just had to scrape off the reflective backing. The contrast resistor needed adjustment though.

Some people disconnect the inverter and just replace it with a standard

16x1 LED display, fed from the +5 rail, that mod is apparently drop-in. A few people have replaced their AJ screens with OLED displays, but they aren't 100% compatible and need a hacked firmware ROM to support them.
Reply to
bitrex

Apparently both the EL backlight and the inverter are prone to failure. The inverter looks like a pretty simple blocking oscillator configuration, either the filter cap or transistor can fail. If the transformer fails then you're probably SOL.

The EL material fades out over time.

The inverter on mine was putting out good HV as is, but the EL material had faded. I happened to have some EL material lying around so I fitted that, but in the process the old LCD display basically crumbled as I removed it. Fortunately I found an unlit display module that was pretty much an exact replacement on eBay - 3 for $3. I just had to scrape off the reflective backing. The contrast resistor needed adjustment though.

Some people disconnect the inverter and just replace it with a standard

16x1 LED display, fed from the +5 rail, that mod is apparently drop-in. A few people have replaced their AJ screens with OLED displays, but they aren't 100% compatible and need a hacked firmware ROM to support them.

Yes, I recently replaced the LCD on a Juno G, which was no longer available. I got the updated LCD from Roland, complete with firmware upgrade, that would enable the Juno to properly talk to the new LCD.

Gareth.

Reply to
Gareth Magennis

Yes, I recently replaced the LCD on a Juno G, which was no longer available. I got the updated LCD from Roland, complete with firmware upgrade, that would enable the Juno to properly talk to the new LCD.

Oh, and after supplying the serial number of this Juno G, Roland supplied the new LCD and firmware update for free.

The customer was delighted, as was I, even though I made no profit on the parts.

It was an uneconomical to repair, otherwise.

Gareth.

Reply to
Gareth Magennis

Question is now that I've located the problem, what's the most effective way to make an expedient repair? I can't seem to get the wire to make firm contact no matter how I poke at it, and it seems impossible to pull out of the header.

I'm thinking about just clipping it off, heat shrinking a longer section to it and then soldering it directly to the underside of the board, but that seems a bit barbaric.

Reply to
bitrex

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.