4 ms·
In a previous life I worked for a company that made extremely precise tooling for airplane manufacturing. These things were often upwards of 200ft long and 60f
by dan_quixote 6y ago
In a previous life I worked for a company that made extremely precise tooling for airplane manufacturing. These things were often upwards of 200ft long and 60ft tall. And they are so precise, that a difference of 5F could make the steel structures grow enough to be out of spec. Most aerospace factories have tightly controlled HVAC for this reason. Well, anyway, we installed a set of tooling in a factory in Japan where the summers can be quite hot. They didn't have HVAC in the factory and this resulted in a 30F gradient from bottom to top. This meant that we had to redesign some connections between parts to progressively account for thermal expansion. This crazy problem would not have existed had they just had HVAC!
- Hello71 6y agowouldn't it be cheaper to buy a few dozen box fans from the local electronics store?
- closeparen 6y agoCirculation is not cooling.
- foota 6y agoTechnically, parent says the issues was the temperature gradient, so this may have helped?
- nix23 6y agoNormally, high precision measuring rooms have zero airflow, otherwise have a constant temperature in the room and on the piece is not possible.
- foota 6y agoSorry, could you clarify? It seems like zero airflow would inhibit a constant temperature?
- nix23 6y agoFor example you putt those pieces like 24 hours (depends on thickness) before measuring in that room, the piece should have from it's core to the edges the same temperature. Airflow inhibits a precise measurement, imagine you putt something in that room that changes the airflow just a bit, so more warm air from the ceiling hit's one side of the piece. Always when you try to measure something, no outer energies should interfere (no temperature change and no force like airflow or even driving cars outside of the building), it could even be, that the airflow creates vibrations in the piece, like rolling cars do.
- bumby 6y agohttps://en.m.wikipedia.org/wiki/Convective_heat_transfer https://en.m.wikipedia.org/wiki/Convective_heat_transfer
- dan_quixote 6y agoYou'd need much more serious fans. And this method lacks any kind of real control. It would reduce the gradient though.
- ip26 6y agoDo you think it would have been cheaper in the long term, e.g. 10-20 years, for them to install & run the HVAC continuously? Was the redesign terribly difficult & expensive?
- dan_quixote 6y agoIt's been a decade so my memory is fuzzy, but it was likely on the order of $1-3million (USD) and 3-6 months. Not sure what HVAC would cost over that time. The real problem is that these types of adaptations are only in spec over some range. Any deviation from that range can be problematic and non-linear.
- nix23 6y agoYeah no, it really depends on the working piece (geometry), material, manufacturing and so on...its incredible complex, some milling machines work 24/7 (even without a piece in it) just to uphold the overall machine temperature. A HVAC could make stuff even worse when your piece or machine comes in to the cold flow.