[GB] Melody96 Type-C + QMK

[mtt]

Got some weird issue with my board.

The majority of the underglow LEDs cant be controlled most of the times (It very rarely works).

E.g. FN Q > turns on underglow LEDs. If I press it another time, only half of the LEDs turn off  (see attached picture). Same goes for every other command - such as FN E to change the RGB modes. Only a part of the LEDs react. I asked Vincent already but I haven't heard back yet.

Maybe one of you guys know whats the issue here? Is it QMK?

I have reset the EEPROM and reflashed the default hex a few times already to no avail.

Any help is much appreciated 

[masterhail]

sorry for the dumb question, but do i need to program one of my keys to be the FN key?

My bottom row is currently ctrl/win/alt/space/alt/ctrl/arrows

[driftingbunnies]

Yeah you need to flash it. Just follow the instructions given by Vincent in the manual

[BearonicMan]

Got my board yesterday and it's a lot heavier than I expected. The brass plate is scuffed\scratched up fairly badly so I guess I'll need to buff or polish it a bit and it looks like the case was dropped pretty hard in transit or something because it actually deformed the aluminum corner and took a chunk out of it... Overall happy with the board though. I still need to get it hooked up and test the keys as I imagine the drop may have possibly broken some of the soldering. Can't wait to get a set of caps on it and get it up and running though.

[KaosJ]

Got my board yesterday and it's a lot heavier than I expected. The brass plate is scuffed\scratched up fairly badly so I guess I'll need to buff or polish it a bit and it looks like the case was dropped pretty hard in transit or something because it actually deformed the aluminum corner and took a chunk out of it... Overall happy with the board though. I still need to get it hooked up and test the keys as I imagine the drop may have possibly broken some of the soldering. Can't wait to get a set of caps on it and get it up and running though.

damn son 

if you got pics of the package, just contact vincent, this kind of shipping might be insured. 

[masterhail]

Yeah you need to flash it. Just follow the instructions given by Vincent in the manual

Appreciate the help!

I was able to get everything working fine except the L_GUI key. For some reason I can't get it to register as the windows key.

I've re-flashed the firmware and the layout multiple times now but haven't been successful. The switch is working, as I can map it to something else and it works fine. it just doesn't want to work as the windows key.

Any advice on how to fix the issue?

[driftingbunnies]

I'll take a look when I get home but the standard json file from the drive seem to work for me. All I did was switch fn and Ctrl for my setup

https://drive.google.com/open?id=1oJmSfjgs_DugsvloO9VaO5KWvwS4T2OX

[masterhail]

I'll take a look when I get home but the standard json file from the drive seem to work for me. All I did was switch fn and Ctrl for my setup

https://drive.google.com/open?id=1oJmSfjgs_DugsvloO9VaO5KWvwS4T2OX

Yeah that was the same json I was using. I had just changed a few keys on the numpad and the fn/ctrl switch. Other than that it's just using that standard Json.

Going to try and totally wipe and reflash again and see if it makes a difference.

[Broseph]

Yeah you need to flash it. Just follow the instructions given by Vincent in the manual

Appreciate the help!

I was able to get everything working fine except the L_GUI key. For some reason I can't get it to register as the windows key.

I've re-flashed the firmware and the layout multiple times now but haven't been successful. The switch is working, as I can map it to something else and it works fine. it just doesn't want to work as the windows key.

Any advice on how to fix the issue?

I had the same issue, it was because bootmagic was disabling left windows. Holding space and the left windows key while plugging in the board is what fixed it for me.

https://docs.qmk.fm/#/feature_bootmagic

[masterhail]

I had the same issue, it was because bootmagic was disabling left windows. Holding space and the left windows key while plugging in the board is what fixed it for me.

https://docs.qmk.fm/#/feature_bootmagic
[/quote]

Thanks for that info! Been reading up - didn't realize bootmagic had so many options.

Unfortunately I still can't get the LGUI key to register when holding down space+windows while plugging it in.

I can get the board to reset all options by holding space+esc on boot, but can't seem to get this dang windows key to fire up.

[Thumper_]

So my board arrived (beautiful and top quality)and upon building it I noticed that 9/16 rgb leds light up and no key / switch registers. Even with a SS316 bridge there is no key press.
For the rgb led I’ll upload an image tomorrow maybe there is an issue I don’t see.
For the Keycaps problem, i followed the pdf manual step by step but somewhere is a bug ...

[Jkush463]

If anyone has a gold top or whole case they want to sell bang my line

[Thumper_]

So my board arrived (beautiful and top quality)and upon building it I noticed that 9/16 rgb leds light up and no key / switch registers. Even with a SS316 bridge there is no key press.
For the rgb led I’ll upload an image tomorrow maybe there is an issue I don’t see.
For the Keycaps problem, i followed the pdf manual step by step but somewhere is a bug ...

So, nearly everything works now.

Rgb light is still at 9/16 LEDs. I tried to resolder the last working and first not working led. Will replace them later with a new led.
And my 8 key is not registering. Neither in the switch nor on the pcb. How can I troubleshoot this?

[BearonicMan]

damn son 

if you got pics of the package, just contact vincent, this kind of shipping might be insured.

Thanks dude, I'm going to mess around with it this weekend here, I'll try to upload a pic or two this evening. It's nothing that can't be fixed though.

[jihadu]

Have the EU orders started to ship already? We got an Email from saying it would arrive soon, but it's been 3 weeks since that

[BearonicMan]

Just those pics I talked about earlier, nothing too crazy and nothing that's not fixable, just looks like it's been through the ringer.

[Astralou]

Just wanna share mine : Cyan and Cyan plate with acrylic bottom and SA Miami

[Jay5ash]

Have the EU orders started to ship already? We got an Email from saying it would arrive soon, but it's been 3 weeks since that

Hi jihadu
Are you still waiting for you board to be shipped?

[jihadu]

Have the EU orders started to ship already? We got an Email from saying it would arrive soon, but it's been 3 weeks since that

Hi jihadu
Are you still waiting for you board to be shipped?

Just got the shipping notification yesterday

[WHeisenberg]

Mine just arrived! Looks awesome in person. Truly eye catching. 

Now I just need to read up on how to assemble and program it. Saw some useful tips earlier in this thread, so thanks for those!

[Kibort]

I got my melody96 from the Candykeys proxy a few days ago. The board looks absolutely stunning. Tested all the keys with a tweezer and all good. Half of the SMD leds on the underside didn’t light up at first. After some testing and looking I saw that 1 of them in he corner got pulled of during shipping. Was easy enough to solder in a new one that came extra with the board. All good now. Now just waiting on a keyset FeelsBadMan

[mousetrap00]

Mine just arrived! Looks awesome in person. Truly eye catching. 

Now I just need to read up on how to assemble and program it. Saw some useful tips earlier in this thread, so thanks for those!

What colors are those?

[Uhmmmm]

How can i turn off leds without Fn? I didn't like FN button

[jihadu]

How can i turn off leds without Fn? I didn't like FN button

If you want too, you can program any key to led control. That's the beauty of programability

[Tasteless]

Just those pics I talked about earlier, nothing too crazy and nothing that's not fixable, just looks like it's been through the ringer.
(Attachment Link) (Attachment Link) (Attachment Link)

Just a heads up, yours looks exactly like mine did before removing the plastic film cover.  Have you removed the weight and made sure you don't have the film there?

[KaosJ]

Just those pics I talked about earlier, nothing too crazy and nothing that's not fixable, just looks like it's been through the ringer.
(Attachment Link) (Attachment Link) (Attachment Link)

Just a heads up, yours looks exactly like mine did before removing the plastic film cover.  Have you removed the weight and made sure you don't have the film there?

I was thinking the same. Mine was like that before removing the plastic on it. 

[notFREEfood]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):





Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

[merlin64]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

Have you tried reflashing the board and maybe putting the Fn somewhere else?

[Dumblescore]

Apologies if it’s out there and I didn’t see it — is the plate design available to download somewhere?  I’m looking to get a busted Melody and try a completely different layout with it but I’m going to have to make a new plate.  Thanks!

[Uhmmmm]

How can i turn off leds without Fn? I didn't like FN button

If you want too, you can program any key to led control. That's the beauty of programability

Can i use macro to setup rgb toggle button? I don't see any option in order to control lightning in macro tab.
BTW: i changed right Alt into Fn 

[KaosJ]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

What layout you flashed in the first place?  I'm not an expert with the multimeter so on this i can't help, but i think you are trying to do the "last thing to try" and i don't know if you tried the "first things to try".  E.g. do you even have a key to switch layers setted? You tried to do an hard reset?

[azachhh]

my two melody96s!

[Uhmmmm]

I want to add leds for cap lock and num lock in order to recognize that i pressed them or not. However, i don't know how to do that 

[notFREEfood]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

What layout you flashed in the first place?  I'm not an expert with the multimeter so on this i can't help, but i think you are trying to do the "last thing to try" and i don't know if you tried the "first things to try".  E.g. do you even have a key to switch layers setted? You tried to do an hard reset?

I've already tried flashing the default layout, and if I'm reading the layout correctly that's they key to switch layers.  If I plug it into a computer it immediately starts behaving like a handful of keys are held down, and I've confirmed it is in layer 1 as the LED controls work.

I should be seeing a zero for when I have my two leads on those pads as there should be absolutely no path for current, but what I am seeing indicates that there is some component bridging the gap between the two lines.  Deleting the keys from the keymap to test this out is probably a good place to start, but trying to find motivation to work on this when I've already got a number of other headaches is difficult. I'm just hoping it's something stupid that somehow I'm overlooking, because the other reasons I can think of I really dislike (stab lube is slightly conductive, bad chip).

[KaosJ]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

What layout you flashed in the first place?  I'm not an expert with the multimeter so on this i can't help, but i think you are trying to do the "last thing to try" and i don't know if you tried the "first things to try".  E.g. do you even have a key to switch layers setted? You tried to do an hard reset?

I've already tried flashing the default layout, and if I'm reading the layout correctly that's they key to switch layers.  If I plug it into a computer it immediately starts behaving like a handful of keys are held down, and I've confirmed it is in layer 1 as the LED controls work.

I should be seeing a zero for when I have my two leads on those pads as there should be absolutely no path for current, but what I am seeing indicates that there is some component bridging the gap between the two lines.  Deleting the keys from the keymap to test this out is probably a good place to start, but trying to find motivation to work on this when I've already got a number of other headaches is difficult. I'm just hoping it's something stupid that somehow I'm overlooking, because the other reasons I can think of I really dislike (stab lube is slightly conductive, bad chip).

You should write Vincent, his contacts are in the first page

[azachhh]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

What layout you flashed in the first place?  I'm not an expert with the multimeter so on this i can't help, but i think you are trying to do the "last thing to try" and i don't know if you tried the "first things to try".  E.g. do you even have a key to switch layers setted? You tried to do an hard reset?

I've already tried flashing the default layout, and if I'm reading the layout correctly that's they key to switch layers.  If I plug it into a computer it immediately starts behaving like a handful of keys are held down, and I've confirmed it is in layer 1 as the LED controls work.

I should be seeing a zero for when I have my two leads on those pads as there should be absolutely no path for current, but what I am seeing indicates that there is some component bridging the gap between the two lines.  Deleting the keys from the keymap to test this out is probably a good place to start, but trying to find motivation to work on this when I've already got a number of other headaches is difficult. I'm just hoping it's something stupid that somehow I'm overlooking, because the other reasons I can think of I really dislike (stab lube is slightly conductive, bad chip).

If you can't get it figured out, I have an extra PCB I could sell you. $20 + whatever it costs to ship.

[MiradoBlackWarrior]

A little late to this but I couldn't get around to building the keyboard until now. Anybody else have an issue with the LED's? I have everything flashed and tried the FN+QUERTYUI and nothing will toggle the bottom LED's on. I even tried changing other keybindings on the FN(1) layer to make sure the FN was working properly but to no avail. Just looking for any advice because I bought the acrylic bottom so it's pretty pointless if the underglow doesn't work.

[YiZZo]

I am in the same boat as you man. Acrylic bottom and some of the LEDs on the bottom row dont work. I tried reflowing the last un-lit LED and next lit LED and I was able to get one LED to come back, but the rest still don't light up. So I tried reflowing all the un-lit LEDs to no avail. I don't know if I need to replace those LEDs all together or the board traces for the LEDs are bad.

Does anyone have spare LEDs or know the size/model of the LEDs that I can buy?

[superuser41]

The traces are unlikely to be bad. The LEDs are standard 5050-size WS2012b/SK6812s. The datasheet has the pinout, for troubleshooting: https://www.seeedstudio.com/document/pdf/WS2812B%20Datasheet.pdf

[notFREEfood]

So layer 1 is permanently activated on my board, and I think that this is the source of my issue (seeing this on multiple keys for this row):

Show Image

Show Image

Anyone have any idea as to what is causing this?  Not using screw-in stabs and I can't see any solder bridges.

What layout you flashed in the first place?  I'm not an expert with the multimeter so on this i can't help, but i think you are trying to do the "last thing to try" and i don't know if you tried the "first things to try".  E.g. do you even have a key to switch layers setted? You tried to do an hard reset?

I've already tried flashing the default layout, and if I'm reading the layout correctly that's they key to switch layers.  If I plug it into a computer it immediately starts behaving like a handful of keys are held down, and I've confirmed it is in layer 1 as the LED controls work.

I should be seeing a zero for when I have my two leads on those pads as there should be absolutely no path for current, but what I am seeing indicates that there is some component bridging the gap between the two lines.  Deleting the keys from the keymap to test this out is probably a good place to start, but trying to find motivation to work on this when I've already got a number of other headaches is difficult. I'm just hoping it's something stupid that somehow I'm overlooking, because the other reasons I can think of I really dislike (stab lube is slightly conductive, bad chip).

If you can't get it figured out, I have an extra PCB I could sell you. $20 + whatever it costs to ship.

Thanks for the offer; I'm going to wait to hear back from Vincent, but if that goes nowhere I'll probably take you up on it.

Anyone have any thoughts on replacing the microcontroller?

[superuser41]

For what reason? That's the least likely culprit.

[notFREEfood]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

[the_high_5]

I wasn't a part of the group buy, but I was lucky enough to pick up a kit from Vincent. Thought I'd share my build with you all. Really happy with this!

- Black aluminum case (top and bottom) with brass plate
- Outemu Sky v2.2 lubed with Tribosys 3204
- JTK Photo Studio

[Jkush463]

still looking for a gold melody if anyone wants to give one up!

[superuser41]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

Well, that's annoying. I would offer to replace the 32u4 for you but shipping two ways may make it less appealing than just getting that new PCB that was offered.

[notFREEfood]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

Well, that's annoying. I would offer to replace the 32u4 for you but shipping two ways may make it less appealing than just getting that new PCB that was offered.

You were right that it wasn't the chip...apparently one of the switch pads is incredibly easy to bridge to ground, to the point were you can bridge it, desolder the switch and the bridge will still remain

Pictures of the evil pad in question to come tomorrow when I'm not falling asleep

EDIT: yup, that was the culprit, a single leg shorted to ground...now I need to figure out how to fix the two LEDs that I murdered in the process of fixing the switches

EDIT 2: how do you desolder one of the rgb leds with a conventional soldering iron? I removed all of the visible solder and it's still stuck...

Here's the bad pad in question.  You can't really tell from this, but the ground is exposed a fraction of a millimeter away from the pad...

[Chennasaurus]

I've been having some similar issues with my board as well, it being pre-soldered. Half of the PCB LEDs do not light up and it's also permanently in layer 1.

I don't have the time or the expertise to debug and fix this, so I'm offering to sell the board to anyone who wants it. I've also learned my lesson about getting things soldered, rather than doing it myself.

It's a gold case with acrylic bottom, kailh brown switches, and had the PBT Gold and Blue switches that Vincent offered. Everything works fine besides the above defects (couldn't do anything about the LEDs, but I've reprogrammed everything to layer 1 of the board as a temporary fix).

[superuser41]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

Well, that's annoying. I would offer to replace the 32u4 for you but shipping two ways may make it less appealing than just getting that new PCB that was offered.

You were right that it wasn't the chip...apparently one of the switch pads is incredibly easy to bridge to ground, to the point were you can bridge it, desolder the switch and the bridge will still remain

Pictures of the evil pad in question to come tomorrow when I'm not falling asleep

EDIT: yup, that was the culprit, a single leg shorted to ground...now I need to figure out how to fix the two LEDs that I murdered in the process of fixing the switches

EDIT 2: how do you desolder one of the rgb leds with a conventional soldering iron? I removed all of the visible solder and it's still stuck...

Here's the bad pad in question.  You can't really tell from this, but the ground is exposed a fraction of a millimeter away from the pad...

Show Image

Glad you found the short. For the LED, I typically use either low-temperature solder or a hot-air reflow station. It's definitely going to be trickier without those. Did you use wick?

[notFREEfood]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

Well, that's annoying. I would offer to replace the 32u4 for you but shipping two ways may make it less appealing than just getting that new PCB that was offered.

You were right that it wasn't the chip...apparently one of the switch pads is incredibly easy to bridge to ground, to the point were you can bridge it, desolder the switch and the bridge will still remain

Pictures of the evil pad in question to come tomorrow when I'm not falling asleep

EDIT: yup, that was the culprit, a single leg shorted to ground...now I need to figure out how to fix the two LEDs that I murdered in the process of fixing the switches

EDIT 2: how do you desolder one of the rgb leds with a conventional soldering iron? I removed all of the visible solder and it's still stuck...

Here's the bad pad in question.  You can't really tell from this, but the ground is exposed a fraction of a millimeter away from the pad...

Show Image

Glad you found the short. For the LED, I typically use either low-temperature solder or a hot-air reflow station. It's definitely going to be trickier without those. Did you use wick?

I tried using wick, but the LED still wouldn't come off

I was thinking hot air is the way to do it, but I really don't want to drop a few hundred on one right now...

[superuser41]

For what reason? That's the least likely culprit.

Every single IO pin has a voltage drop going towards one IO pin, which is the pin that takes in the row input for the function key.  Flipping the leads results in an open circuit, which means that there is at least one diode in between this IO pin and all of the other IO pins.  Unfortunately with switches soldered in I can only see half of the board, but there's no obvious defects that I can see, especially one that would cause this pin to get connected to every single other pin.  I didn't completely mangle my board, which should be obvious from the picture, and there is no circuitry that I can find outside of the chip that could produce the results I see from my multimeter.  I can think of reasons why this circuitry would exist inside of the chip however, and if the internal diode went bad and is now allowing bidirectional current that would explain everything.

The only reason I'm hesitating is because I have very limited experience with smt soldering (my Iris was the first time I had ever done it).

Well, that's annoying. I would offer to replace the 32u4 for you but shipping two ways may make it less appealing than just getting that new PCB that was offered.

You were right that it wasn't the chip...apparently one of the switch pads is incredibly easy to bridge to ground, to the point were you can bridge it, desolder the switch and the bridge will still remain

Pictures of the evil pad in question to come tomorrow when I'm not falling asleep

EDIT: yup, that was the culprit, a single leg shorted to ground...now I need to figure out how to fix the two LEDs that I murdered in the process of fixing the switches

EDIT 2: how do you desolder one of the rgb leds with a conventional soldering iron? I removed all of the visible solder and it's still stuck...

Here's the bad pad in question.  You can't really tell from this, but the ground is exposed a fraction of a millimeter away from the pad...

Show Image

Glad you found the short. For the LED, I typically use either low-temperature solder or a hot-air reflow station. It's definitely going to be trickier without those. Did you use wick?

I tried using wick, but the LED still wouldn't come off

I was thinking hot air is the way to do it, but I really don't want to drop a few hundred on one right now...

Since the LED is busted anyway you can carefully lift each leg as you heat it, or get some low-melting-point solder.