29-Apr-2015, 07:38
(This post was last modified: 29-Apr-2015, 07:43 by Rufus McDufus.)
Hi - I'll tell you the whole story. I contacted Devialet after finding that the unit seemed to be getting more and more muffled. It seemed worse after it'd been on for an extended time. It was quite severe - pretty much unlistenable it was so bad. A new problem started happening a couple of times too - it just started outputting clicking sounds during music playback - no music, just clicking, and only cured by turning it off and on again. This was with a USB source, nothing to do with the AIR problem.
I opened the unit up and first thing I noticed was the Class D board stuck to the plastic bottom cover of the D200. I didn't try and free it, The blue (Blu-Tack like) heatsink material was sticking it to the cover. There was a fair bit of fluid in the tray where the Class D board sits. I also noticed some solder connections - a couple on the power input board where the Schurter filter is - looking a bit 'weepy'. I didn't know if this was heat damage, failing joints or what.
The other thing I noticed was it sounded a lot better when I put the Class D board back in again and put it back together. It was like a completely different unit!
OK so back to Devialet. The sticky Class D board is unfortunately very common. The heatsink material gets a bit sticky. It also 'sweats' a bit and damp-looking patches can appear. This is not a problem but they cleaned it up.
They changed the board with the Schurter filter as well. I don't know why. Whether it was just a matter of precaution or not due to the slightly poor solder joints I don't know.
Anyhow upshot is it sounds a lot more consistent. I haven't experienced any of that muffled soundingness since.
The only snag is I was investigating other things in my system which could've been responsible too! I stopped using a dedicated earth rod which could well have been an issue. Its impedance stayed very constant but didn't have great earth loop impedance readings (about 50 Ohms, which is OK but to provide a benefit it should really be better). I'd also been grounding my Entreq Cleanus filter on the same grounding box that the Devialet was on. A few people had noticed this led to inconsistency so I removed the ground connection. I've also noticed recently that the port trunking (2 ethernet cables, balanced load balancing) I'd used on my Qnap (for speed - my wife is a photographer and it improves the upload speed a fair bit) was detrimental to the sound quality. Making it a single connection seemed to remove some inconsistency. Why, I have no idea!
The question still remains in my head why it sounded so much better after opening the unit, replugging the Class D board, and putting it back together. Was it just reseating the board that helped? The explanation could actually be related to earthing - just the simple procedure of unplugging it from the mains and replugging.
And for the clicking sound - no idea... Maybe the Schurter filter board is a known issue and that solved everything? Seems strange though as it's a very common component used in a lot of devices and not really prone to going wrong. I don't think there's much else complicated attached to that board.
I opened the unit up and first thing I noticed was the Class D board stuck to the plastic bottom cover of the D200. I didn't try and free it, The blue (Blu-Tack like) heatsink material was sticking it to the cover. There was a fair bit of fluid in the tray where the Class D board sits. I also noticed some solder connections - a couple on the power input board where the Schurter filter is - looking a bit 'weepy'. I didn't know if this was heat damage, failing joints or what.
The other thing I noticed was it sounded a lot better when I put the Class D board back in again and put it back together. It was like a completely different unit!
OK so back to Devialet. The sticky Class D board is unfortunately very common. The heatsink material gets a bit sticky. It also 'sweats' a bit and damp-looking patches can appear. This is not a problem but they cleaned it up.
They changed the board with the Schurter filter as well. I don't know why. Whether it was just a matter of precaution or not due to the slightly poor solder joints I don't know.
Anyhow upshot is it sounds a lot more consistent. I haven't experienced any of that muffled soundingness since.
The only snag is I was investigating other things in my system which could've been responsible too! I stopped using a dedicated earth rod which could well have been an issue. Its impedance stayed very constant but didn't have great earth loop impedance readings (about 50 Ohms, which is OK but to provide a benefit it should really be better). I'd also been grounding my Entreq Cleanus filter on the same grounding box that the Devialet was on. A few people had noticed this led to inconsistency so I removed the ground connection. I've also noticed recently that the port trunking (2 ethernet cables, balanced load balancing) I'd used on my Qnap (for speed - my wife is a photographer and it improves the upload speed a fair bit) was detrimental to the sound quality. Making it a single connection seemed to remove some inconsistency. Why, I have no idea!
The question still remains in my head why it sounded so much better after opening the unit, replugging the Class D board, and putting it back together. Was it just reseating the board that helped? The explanation could actually be related to earthing - just the simple procedure of unplugging it from the mains and replugging.
And for the clicking sound - no idea... Maybe the Schurter filter board is a known issue and that solved everything? Seems strange though as it's a very common component used in a lot of devices and not really prone to going wrong. I don't think there's much else complicated attached to that board.
