At 19:15 23-9-02 +0100, you wrote: >I'm looking for power supplies suitable for the VIA EPIA (Eden) 5000 boards, >but the problem is that they must be small and fanless. The only solution >I've found on this list, google searches and the like is to buy a Morex Cubid >2677R (http://www.mini-itx.com/reviews/2677R/) and strip out the DC-DC >converter and the external PSU, but there must be a more elegant (and >cheaper) solution out there. > >I know this list had a discussion that touched on fanless PSU's back in May >this year, but there was no definitive outcome visible in the archives. >Anyone made and progress since then? > >We don't want to use any of the case options we've found as they all include >space for hard drives, cdrom's, etc, which we just don't need. We're >compromising from the openbrick (openbrick.org) for cost reasons, but would even if it means bolting it into a >slightly customized aluminium enclosure ( RS Components standard part type)
>like to get back to that sort of size,> >Thanks > > >Steven Edgar Hi Steven, I have done some investigation about a year ago on fanless PSU's. Found an URL but their PSU's provided too little power for a 'normal' comp. (i.e. motherboard, HD, CDR, floppy, etc.) so I dropped the project. I also dropped some of the ideas I had. And I had never heard of LTSP and diskless clients at that time. Your question made me set up a limited experiment. My original idea was to use an oversized PSU (in terms of power) and simply disconnect the fan. PSU's are designed to work in high temperature environments so, in our climate (Northern Europe), no need for fans, was my assumption. This is what I did today: Took an old 230 watt (220volt) PSU and disconnected the fan. Removed the case so I can sense the temperature on the thyristors cooling sinks. The heat can easily escape. Connected the PSU to a 486 without floppy and HD. Connected monitor and keyboard and mouse. On power-on everything works fine. Here is my report: It's now 07.30 hours. Start of experiment. Room temp. 18 degrees Centigrade. So the PSU temp is 18 C. 08.00: temp 26 C immediately above the heat sinks. 08.15: temp29 C imm. etc. 08.30: 32 C imm. etc. Not bad for a start. Time for some experimentation. So I put the cover back on the PSU and put it in a way the air can travel trough it (kind of chimney). Temperature meter at the top side where the heat comes out (about 2 inch away from the heat sinks) 10.30: Temperature dropped to 27 C (ofcourse). Took the power off the PSU and manually felt the heat sinks. Warm, but not exessive. 11.30: Turned PSU and ventilation holes are now on the side. 12.00:Manually feel temp. on the cover: Very moderate. Feeling the sinks, hot, but very touchable. 12.00: Now I put the temp. meter on the heat sink and cover the PSU with isolating material. 'No' heat can escape. The final test! 13.00: Temp. in the PSU: 35 C. Sinks are hot but touchable. End of experiment. I would follow your idea: >... bolting it into a slightly customized aluminium enclosure ( RS Components standard part type). When you take an 'oversized' (in power) PSU and make it part of a kind of chimney design to promote natural air flow, IMHO it can succeed. Anyway, I learned a lot! And David, thanks for the resistor tip. I would suggest experimenting with the highest possible value (in ohms) that makes the PSU start. Good luck, dirk __________________________________ Dirk Schouten Media Action Projects: http://utopia.knoware.nl/users/schoutdi Tools for teachers: http://213.84.240.66 ________________________________ ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _____________________________________________________________________ Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto: https://lists.sourceforge.net/lists/listinfo/ltsp-discuss For additional LTSP help, try #ltsp channel on irc.openprojects.net
