Memory gone in Crescendo firmware

5 posts / 0 new
Last post
ControlTheController
ControlTheController's picture
Offline
Last seen: 6 days 3 hours ago
Joined: 02/15/2018 - 12:31
Posts: 105
Location: Under a layer of phosphor
Memory gone in Crescendo firmware

I have been using my S2+ with Mtn17DDm driver and 3*SST-20 for quite a while. It has Crescendo firmware and I always keep memory on. The button finally gave up so I replaced it with a new one. But now the firmware is acting weirdly. It no longer memorizes the current level (even after toggling memory on/off). Is it because the memory in the Attiny25 has been worn down? I have a spare MCU lying around, but I doubt that’s the problem.

The person who always struggles between practicality and lumens.

Edited by: ControlTheController on 05/15/2019 - 12:43
Lightbringer
Lightbringer's picture
Offline
Last seen: 2 hours 42 min ago
Joined: 08/30/2016 - 14:12
Posts: 8054
Location: nyc

Soooooo… it works fine, you replace the switch, now it doesn’t work fine?

Certainly seems to me the switch might be the problem.

Maybe it’s “leaky” in some sense, from contamination across the contacts or bridging traces. Same way adding a lighted switch can play Hell with modes/memory/whatever.

Take off the tailcap, bridge the end of the battery to the case. Jiggle it to set different modes. Ie, let it act as if you were working the switch. If the light works as expected, then the switch is to blame.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

ControlTheController
ControlTheController's picture
Offline
Last seen: 6 days 3 hours ago
Joined: 02/15/2018 - 12:31
Posts: 105
Location: Under a layer of phosphor

I just tested it with a piece of wire and our suspicion is spot on. Facepalm I should’ve tried it before lol. However the new switch is not a lighted switch. It’s just an Omten switch soldered onto the original switch’s position. I’ve checked it and I can’t see any debris.

The person who always struggles between practicality and lumens.

Lightbringer
Lightbringer's picture
Offline
Last seen: 2 hours 42 min ago
Joined: 08/30/2016 - 14:12
Posts: 8054
Location: nyc

Could be anything. Solder flux that absorbed crap and is ever-so-slightly conductive, similar with grease, anything.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

ToyKeeper
ToyKeeper's picture
Offline
Last seen: 26 min 46 sec ago
Joined: 01/12/2013 - 14:40
Posts: 9312
Location: (469219) 2016 HO3

It definitely sounds like there’s current leaking through somewhere, preventing the driver from ever fully losing power. That’ll break the offtime measurement, so it probably thinks every button press is a short one.