Hi If you have enough room for a heatsink that is the size of a couple of shoe boxes, then by all means use a convection heatsink.
There is zero difference between cooling by fan vs big heatsink as far as device current consumption is concerned. This or that level of cooling takes the baseplate to this or that temperature. At that baseplate temperature, the device pulls X watts (regardless of how you get to that temperature). If your “natural convection” approach does *not* increase the current drain then it’s not actually cooling the baseplate in any meaningful way. The current drain goes up to heat the oven. The baseplate cools to protect the rest of the device. Indeed the fan its self pulls some current. That current is typically quite low compared to the power you are trying to manage. One cute thing with a fan is that you can servo it. Baseplate control of well under 1C is fairly easy to do. Fun !! Bob > On Sep 1, 2020, at 4:32 PM, Taka Kamiya via time-nuts > <[email protected]> wrote: > > Will you elaborate on fan? Just about everything I've read, use of fan was > discouraged. Natural convection over heat sink fins were recommended. I've > done limited testing. When fan cooling, indeed temperature will go down, but > current consumption stays higher indicating heating is often conducted to > keep it hot inside. How do I reconciliate these opposing conditions in > actual use? > > --------------------------------------- > (Mr.) Taka Kamiya > KB4EMF / ex JF2DKG > > > On Tuesday, September 1, 2020, 2:08:04 PM EDT, Bob kb8tq <[email protected]> > wrote: > > Hi > > In this case there may not be much that can be done. In the more general case > ( = a unit with a surface that mates to a heatsink) , it’s well worth doing > something. > > I have an unfortunate lot of empirical evidence (on FRK’s and LPRO’s ) > showing that > a “no heatsink” setup is one with a short life. With no heatsink, a couple of > years is > doing well on a surplus unit. With a heatsink, > 10 years is quite possible. > > How much heatsink? Is the next question. If you can get the baseplate down > into the > 40’s (C) that’s doing pretty well. A fan plus heatsink may be needed to get > this done. > If a fan is involved, mag field could become an issue ( = Rb’s are sensitive > to mag > field, as the fan speed changes the field *may* change). > > ===== > > Indeed this is not always the official word from the Rb manufacturers. I’ve > spent > considerable effort drinking beer with many of their marketing and > engineering > guys. The conversation is not *quite* the same when marketing it not at the > table. > I believe that they simply don’t have a body of long term data “in house” to > study. > > The fact that I’m not the only one seeing Rb’s die early is backed up on > eBay. > From time to time, you can find pretty large batches of “for parts” Rb’s. > Date codes > often are *not* very far back in time. > > If they get cheap enough (like ~ $30) they can be fun to play with. You can > pretty > quickly work out that the electrolytic caps are the most likely failure > points. Cap > swap outs may only fix 2/3 of them, but it’s quick and easy. You now have a > pile > of $45 Rb’s …. > > Bob > > > > >> On Sep 1, 2020, at 9:23 AM, Matthias Welwarsky <[email protected]> >> wrote: >> >> On Dienstag, 1. September 2020 14:25:31 CEST Richard Katsch wrote: >>> Many thanks Matthias,Hugh. >>> >>> The case doesn’t seem to have any direct method of fastening to a heatsink >>> but something can be arranged I’m sure. The manufacturer’s documents don’t >>> suggest anything. They do say that the base temperature can be up to 85 >>> degrees C which seems awfully hot. >>> >>> I haven’t measured the temperature other than a finger test that indicates >>> that more than 5 secs is uncomfortably hot. At that point I switched it off. >> >> I wouldn't say "uncomfortably hot" is a cause for alarm. The lamp housing in >> the LPRO-101 runs at 100°C, the resonance chamber at roughly the same temp. >> You need Rb vapor at around that temperature (if that Rb is a traditional >> construction) for the system to work. >> >> Anyway, what else can you do? If it's broken, the chances of repairing it >> are >> slim. If it works, no way to find out if it's in spec unless you run it. >> Just >> watch for smoke ;) >> >> BR, >> Matthias >> >>> >>> Best regards >>> Richard >>> >>> On Tue, 1 Sep 2020 at 19:02, Hugh Blemings <[email protected]> wrote: >>>> Hiya, >>>> >>>> >>>> I'd wondered this also - my (limited) experience of Rbs are the >>>> >>>> LPROs and they definitely need a heatsink (I have a nice story to >>>> >>>> share about getting some made one day which I must write up, but I >>>> >>>> digress) >>>> >>>> >>>> Curious, I dug around a bit and only data I can find is a two >>>> >>>> pager on the manufacturer website - >>>> >>>> https://www.quartzlock.com/product/Rubidium/rubidium-oscillators/E10-MRX >>>> >>>> >>>> The packaging itself doesn't appear to be designed with >>>> >>>> heatsinking in mind, but perhaps there is an implicit expectation >>>> >>>> of convection cooling at least to keep it manageable ? >>>> >>>> >>>> Any idea what case temperature you're seeing @Richard ? >>>> >>>> >>>> >>>> >>>> Cheers, >>>> >>>> >>>> Hugh >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> On 1/9/20 5:01 pm, Matthias Welwarsky >>>> >>>> wrote: >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Dienstag, 1. September 2020 07:15:34 CEST Richard Katsch wrote: >>>> >>>> >>>> >>>> >>>> >>>> Hello All, >>>> >>>> I have acquired a Quartzlock E10-MRX Rb 10 MHz standard. It appears to >>>> lock >>>> >>>> and produce a nice sine wave that stays in phase with my Trimble Tbolt for >>>> >>>> a time which exceeds my attention span!!! >>>> >>>> It does however get hot in operation. As this is my first experience with >>>> a >>>> >>>> Rb standard I don’t know whether this is normal or indicates that I need a >>>> >>>> heat sink. >>>> >>>> Any comments would be appreciated. >>>> >>>> >>>> >>>> >>>> >>>> >>>> They're supposed to run hot, but excess temperature will degrade the >>>> stability >>>> >>>> and shorten the lifetime of the electronics. It's hard to recommend what >>>> >>>> heatsink they require if you have no documentation. Typically you'll need >>>> one. >>>> >>>> My LPRO-101 need a heatsink with < 2K/W of thermal resistance to keep the >>>> >>>> temperature in check. >>>> >>>> >>>> >>>> BR, >>>> >>>> Matthias >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> Regards >>>> >>>> Richard Katsch >>>> >>>> VK2EIK >>>> >>>> _______________________________________________ >>>> >>>> time-nuts mailing list -- [email protected] >>>> >>>> To unsubscribe, go to >>>> >>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow >>>> >>>> the instructions there. >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> >>>> time-nuts mailing list -- [email protected] >>>> >>>> To unsubscribe, go to >>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com >>>> >>>> and follow the instructions there. >> >> >> >> >> >> _______________________________________________ >> time-nuts mailing list -- [email protected] >> To unsubscribe, go to >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com >> and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
