600XL Mechanical Keyboard Repair

[+joeatari]

I have a 600XL with an AWC 100001 mechanical keyboard that I am trying to repair.  I have cleaned everything and re-flowed solder to the connectors.  I re-seated the 4051 chips.  

Everything now works with the exception of the A S D E keys.  I'm so close!  

In researching this I have noticed others placing some jumper wires to repair these.  I'm not experienced with this.  Any hints on what is the best way to trouble shoot this keyboard?  The traces look like they are in good shape.  I don't see any physical damage.  I have used a jumper wire to check to make sure it is not the key switches.  These keys do not work when jumpered, but the others do, so I do not think it is the switches.  Thanks!

[bfollowell]

When you mentioned the problem keys and said A S D, it looked promising, as all of those are on the same row in the keyboard matrix, then you threw E into the mix and that screwed up that theory.

 

Something else you can try is to swap your 4051 chips and see if the problem keys move to other keys and your original bad keys start working. If so, your problem is as simple as a bad 4051, readily available and fairly cheap. It's worth a shot.

 

[erichenneke]

I've said this many times and will still stand by it 100%. It has yet to fall me... 

 

Get a multimeter, and check continuity from each key contact on the flex mylar circuit to the edge that plugs into the connector. 

Anywhere there is a break/open you start to narrow it down, closer and closer until you pinpoint the exact open.

 

Then use some copper trace paint (I just get mine from automobile rear-window defroster repair kits, permetex works great!) to bridge the gap. Sometimes there are many!  I use a toothpick to paint the trace. 

 

Be patient and use the multimeter... pinpoint, paint it. And repeat.  Until you have a full continuity reading on your multimeter for each mylar key contact.

Plug it in, and you are good to go.

 

I started using this method many years ago, intending it just to be temporary until I could get new mylar.  But I've done this to at least 8 systems over many years, and NONE of them have ever failed again!

 

Eric

 

 

[+joeatari]

2 hours ago, bfollowell said:

When you mentioned the problem keys and said A S D, it looked promising, as all of those are on the same row in the keyboard matrix, then you threw E into the mix and that screwed up that theory.

 

Something else you can try is to swap your 4051 chips and see if the problem keys move to other keys and your original bad keys start working. If so, your problem is as simple as a bad 4051, readily available and fairly cheap. It's worth a shot.

 

I swapped the 4051s (great idea) but unfortunately no change.  I checked all the ribbon cable contacts for continuity from end to end and that checks out.  Tomorrow I'll pull and try a working keyboard from another XL to rule out problems on the motherboard.  Eric:  This keyboard does not have the mylar membrane sheet.  It is just a circuit board with the key switches attached.  I'm hoping I can find what may be broken in the circuit and add jumper wires to fix the problem, but it is a lot of traces to follow and check.

[+joeatari]

I checked with a working keyboard (the mylar version) and it works just fine with the motherboard in question.  It is definitely a problem with this particular mechanical keyboard.

[Piotr D. Kaczorowski]

@joeatari,

 

 

Check the tutorial prepared by @Peri Noid:

 

 

 

[+joeatari]

8 hours ago, Piotr D. Kaczorowski said:

@joeatari,

 

 

Check the tutorial prepared by @Peri Noid:

 

 

 

Nice! I appreciate this.  The switches are not the problem in this case, but I certainly learned something!

[+joeatari]

This didn't take long.  Just as you would expect, I followed the traces with an OHM meter and ran a jumper wire to the "dead" location.  I'm not really sure why they are dead, because the traces look just fine.  The top white wire fixed the "E" and the bottom blue wire fixed "A S D."  The other wires were applied to the keyboard by a previous owner.  Another 8-bit Atari saved.  This makes eight for me in 2023.  Happy New Year everyone!

[Peri Noid]

Probably some micro-cracks. It's possible. Notice, that the PCB works and bends a bit. Good job.

[TZJB]

2 hours ago, joeatari said:

This didn't take long.  Just as you would expect, I followed the traces with an OHM meter and ran a jumper wire to the "dead" location.  I'm not really sure why they are dead, because the traces look just fine.  The top white wire fixed the "E" and the bottom blue wire fixed "A S D."  The other wires were applied to the keyboard by a previous owner.  Another 8-bit Atari saved.  This makes eight for me in 2023.  Happy New Year everyone!

 

Nice fix.

If the top shield is missing, this would allow the keyboard to flex and cause the micro-cracks that @Peri Noid suspects.

In my 600XL, it has two rubber supports/insulators under the keyboard to act as spacers against the shield. This prevents flexing of the board from occuring and is very effective.

[Piotr D. Kaczorowski]

@joeatari Congrats!

[_The Doctor__]

On 12/29/2023 at 3:00 PM, TZJB said:

 

Nice fix.

If the top shield is missing, this would allow the keyboard to flex and cause the micro-cracks that @Peri Noid suspects.

In my 600XL, it has two rubber supports/insulators under the keyboard to act as spacers against the shield. This prevents flexing of the board from occuring and is very effective.

Yep as stated many times ad-nauseam,  the shielding is electrical/structural/emi/rf for Atari machines. The number of machines with problems have inevitably been used for long periods of time without shielding structure... always showing micro-fractures from stress on the PCB and keyboards. Almost never seen when shielding is present.