KnightMB,
Oil cooled is meant as per the description of the said mods, not any other way. Go read them, bottom of the first page.
Yes, you understood correctly that oil is to be used as a heat transfer medium, and that this use of oil as heat transfer medium is what makes the controller oil cooled. This is by opposition to keeping it air cooled such as it was in stock form.
I won't ask you again what makes you think that an active cooling system is required if oilcooled, when you suggest passive air cooling would otherwise be more then adequate, since I've asked twice allready and all I got was a nonsensical argument that does not adress the question at hand. You may as well say the computers are a bad example since they don't support your theory... Cooling a computer is not much different from a transformer, or a controller, it's just that cooling of computers is better documented on the net, and they mostly have built-in temp sensors to keep things accurate.
A computer in oil tank stays content beeing passively cooled, whereas removing the fans in an aircooled computer will make it malfunction within seconds and pop within minutes. Also, to get comapareable results to a passive oil system using air requires an active system (fans). This demonstrates that passive air cooling is quite drasticly less effective then passive oil cooling, more detail on this later on. But I like your idea of a fan in the case, that ought to work, though I don't see how it could be made to fit in there.
I do plan to cool the entire controller with oil, as per the mod's description, bottom of the first page. The buss bar is not melting it's the solder around it, look at the pics? And yeah, thicker wire I've allready thought of. So vegetable oil as a poor choice for cooling, well it was the first that came to mind in describing the mod, having seen a computer run fanless while dunked in the stuff. There's a crapload of other fluids that would be suitable, but at this point mineral, silicon and PAO are the top 3 oil candidates for various reasons. And though I've been reading up for days, I've not come across a description of this mineral oil used in X-rays and partical accelerators that you're talking about, may be interesting. If you were thinking about flourinert such as used in some larger computers though, I'm not considering it due to ridiculous evaporation rate and price.
The circuit board conducts heat (PCB .25 W/mK, Copper 400 W/mK). Air is an insulator (.025 W/mK) compared to the board, so heat builds up on the board and it's components. Heat is likely to be why the cap was cramming, but also the blackening of the PCB due to heat means carbon, this results in a conductive path that leads to erratic behaviour and premature failure. This is a bad thing. Now air is an insulator (0.025 W/mK) compared to oil (0.180 W/mK). And so, oil does a much better job then air at removing heat from components/circuit board, transfering it onto the case to be dissipated. So the entire case should feel warm and be a suitable place to put a heatsink.
In other words, having the controller filled with air keeps heat around the components more effectively then oil, this explains why mine stayed cool with just warm spots where the fets were clamped on, and where the pcb got near 200 degrees. Except of course for the time the solder finally went, then the entire case got hot...
Even having agreed that air is an insulator compared to oil, you still hold the idea that an oil cooled controller will keep the controller's components hotter then an aircooled equivalent, so I've thought of a mod for you to try: Line the inside of your controller real thick with pink fiberglass insulation, that'll optimise the cooling of the components so much that the entire case will stay cold, no more heat issues!
Air is better then oil kinda like the way fibreglass insulation is better then air...
-------------------------------------
Latest iteration of the controller mod has AWG12 coming in, splitting into AWG14 and the shortened thinner stock cables. In other words, the bussbar has been mostly circumvented, with the double effect of neutralising that heat source and removing this controller's amp limiting capabilities...
Used lead-free solder for pretty much everything, it does take much more heat to melt. My 40w iron was about right for most of the stuff, but not hot enoufgh to do a clean job on the junction of the big power wires, for that I had to use leaded solder instead.
Edit: Wattage went from 0 and started climbing exponentially after plugging it in, so I yanked the cord, but not before smoke came out... Took maybe two secs?
Oil cooled is meant as per the description of the said mods, not any other way. Go read them, bottom of the first page.
Yes, you understood correctly that oil is to be used as a heat transfer medium, and that this use of oil as heat transfer medium is what makes the controller oil cooled. This is by opposition to keeping it air cooled such as it was in stock form.
I won't ask you again what makes you think that an active cooling system is required if oilcooled, when you suggest passive air cooling would otherwise be more then adequate, since I've asked twice allready and all I got was a nonsensical argument that does not adress the question at hand. You may as well say the computers are a bad example since they don't support your theory... Cooling a computer is not much different from a transformer, or a controller, it's just that cooling of computers is better documented on the net, and they mostly have built-in temp sensors to keep things accurate.
A computer in oil tank stays content beeing passively cooled, whereas removing the fans in an aircooled computer will make it malfunction within seconds and pop within minutes. Also, to get comapareable results to a passive oil system using air requires an active system (fans). This demonstrates that passive air cooling is quite drasticly less effective then passive oil cooling, more detail on this later on. But I like your idea of a fan in the case, that ought to work, though I don't see how it could be made to fit in there.
I do plan to cool the entire controller with oil, as per the mod's description, bottom of the first page. The buss bar is not melting it's the solder around it, look at the pics? And yeah, thicker wire I've allready thought of. So vegetable oil as a poor choice for cooling, well it was the first that came to mind in describing the mod, having seen a computer run fanless while dunked in the stuff. There's a crapload of other fluids that would be suitable, but at this point mineral, silicon and PAO are the top 3 oil candidates for various reasons. And though I've been reading up for days, I've not come across a description of this mineral oil used in X-rays and partical accelerators that you're talking about, may be interesting. If you were thinking about flourinert such as used in some larger computers though, I'm not considering it due to ridiculous evaporation rate and price.
The circuit board conducts heat (PCB .25 W/mK, Copper 400 W/mK). Air is an insulator (.025 W/mK) compared to the board, so heat builds up on the board and it's components. Heat is likely to be why the cap was cramming, but also the blackening of the PCB due to heat means carbon, this results in a conductive path that leads to erratic behaviour and premature failure. This is a bad thing. Now air is an insulator (0.025 W/mK) compared to oil (0.180 W/mK). And so, oil does a much better job then air at removing heat from components/circuit board, transfering it onto the case to be dissipated. So the entire case should feel warm and be a suitable place to put a heatsink.
In other words, having the controller filled with air keeps heat around the components more effectively then oil, this explains why mine stayed cool with just warm spots where the fets were clamped on, and where the pcb got near 200 degrees. Except of course for the time the solder finally went, then the entire case got hot...
Even having agreed that air is an insulator compared to oil, you still hold the idea that an oil cooled controller will keep the controller's components hotter then an aircooled equivalent, so I've thought of a mod for you to try: Line the inside of your controller real thick with pink fiberglass insulation, that'll optimise the cooling of the components so much that the entire case will stay cold, no more heat issues!
Air is better then oil kinda like the way fibreglass insulation is better then air...
-------------------------------------
Latest iteration of the controller mod has AWG12 coming in, splitting into AWG14 and the shortened thinner stock cables. In other words, the bussbar has been mostly circumvented, with the double effect of neutralising that heat source and removing this controller's amp limiting capabilities...
Used lead-free solder for pretty much everything, it does take much more heat to melt. My 40w iron was about right for most of the stuff, but not hot enoufgh to do a clean job on the junction of the big power wires, for that I had to use leaded solder instead.
Edit: Wattage went from 0 and started climbing exponentially after plugging it in, so I yanked the cord, but not before smoke came out... Took maybe two secs?
