Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I have an out-of-warranty dasKeyboard Professional (with Cherry MX Blue key switches). I acquired this keyboard as a new replacement in a warranty keyboard swap. Their warranty service is very slow and they do not have an excuse.

Recently the newer keyboard (now over 1 year old) has had trouble producing the b v f and g keys. Sometimes it is just b and v; sometimes it is all four keys that are dead. The problem is intermittent and is not fixed by replugging the USB nor is it fixed by rebooting. The problem has been seen on both Ubuntu 10.10 and Windows 7.

This is not a fast typing problem. When it occurs, it may persist. The problem keys may start working after a while. I would describe the problem as 'truly dead' to distinguish it from the "dead key" problem that has to do with an operating system international usage feature that is actually a user misunderstanding.

The problem keys being in the same area, I suspected a solder joint problem so I re-flowed the solder for a few of the keys in that area but that did not resolve it.

How else might I diagnose and fix this problem?

Edit: Lest you think that a pattern indicates it is me that is defective not the keyboards, the unit Metadot received from me had quite a different defect that had to do with the wrong codes being emitted. The pattern is that two different failing units came from the same maker. It's no excuse but I remember scanning some very clicky $5 key switches (single unit price) with RCA 1802 assembler code around 1979 and it was a little surprising how often my anti bouncing algorithm was not up to the task. It was just a high school project and I would expect better from a manufacturer 30 years later.

Edit 2: It's interesting that in the cases of both defects, the problem took months to show up. May be it is a combination of wear and a marginal algorithm that is a common cause. However, my current impression continues to be that the 'bvfg' dead key problem is quite different. Perhaps b and v have a common current path that is unique to them. If somebody has a dasKeyboard schematic, please share it.

Edit 3: There may be spelling problems because my keyboard is acting up. This is not a mechanical problem with individual Cherry keys. The 4 problem keys were swapped (desoldered/resoldered) with function keys 5 through 8. The problem remains so it must be something in the dasKeyboard design.

share|improve this question
Could be an internal corroded wire or something... –  soandos Jul 31 '11 at 2:13
Have you checked for physical issues like a piece of fuzz under the problem keys? –  Yitzchak Jul 31 '11 at 3:14
soandos: It's unlikely individual keys are wired. Yitzchak: There is no fuzz or obstruction. Actually on a dasKeyboard, you would probably would not have to take a look because you would be able to hear if there were an obstruction –  H2ONaCl Jul 31 '11 at 5:45
Take the board apart and follow the traces of the connections on the trouble keys all the way to whatever chip they connect to. Look for any damage of any sort, such as physical disconnect or discoloration. Follow all of the traces, take high-res photos and zoom in on a computer if necessary. You're looking for something that might result in an intermittent or poor connection. –  MrDoom Sep 21 '14 at 4:29

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.