ULA running hot
ULA running hot
I read on a couple of pages that the ULA would get quite warm, so I thought I would check that out for myself and I was in for quite a shock! I expected it to be warm but it was so hot that it is almost painful to touch! The 7805 heatsink was also warmer than I had expected. So I was in for an immediate modification.
I first took a TRACO TSR 1-2450 DC/DC converter to replace the 7805. This would then leave me with an obsolete heat sink. I then re-used this heat sink for the ULA, first applying some silver epoxy to the ULA package, then sticking the heat sink on top of it. Easy! Just be sure the heat sink is not slanted by the Z80.
I am sure the ULA should last for a bit longer now!
Michel
I first took a TRACO TSR 1-2450 DC/DC converter to replace the 7805. This would then leave me with an obsolete heat sink. I then re-used this heat sink for the ULA, first applying some silver epoxy to the ULA package, then sticking the heat sink on top of it. Easy! Just be sure the heat sink is not slanted by the Z80.
I am sure the ULA should last for a bit longer now!
Michel
Re: ULA running hot
That's really a funny idea. Most people use cooling parts in form of a DIL40 case.
Re: ULA running hot
Yes, why not. It will help. Next thing a cooling fan?
Re: ULA running hot
Yeah I had seen that, but I think the shape of the heat sink is not really important as you can hardly speak of air circulation inside the enclosure. It makes a big difference for sure, the DC/DC converter took nearly 1.5Watt heat dissipation away so the overall temperature inside the case is cooler, which also helps reducing the temperature of the ULA heat sink.PokeMon wrote:That's really a funny idea. Most people use cooling parts in form of a DIL40 case.
Re: ULA running hot
haha!! Now paint the heatsink matt black, this will help heat dissipation even more as black surfaces lose heat faster than 'silvery' shiny ones....will need all the help it can get as the thermal resistance from the ULA 'chip' itself to the heatsink will probably be quite high..
Re: ULA running hot
That's what I thought. It would actually be better to mount the ula on top of a heatsink rather than mounting a heatsink on top of a ula (or both ). I don't know about that DIL40 heatsink, is that something that clips onto the chip in a way the chip is sandwiched between a clasp/clamp system? Or is it just a piece of aluminium glued to the top (like my one)?gozzo wrote:the thermal resistance from the ULA 'chip' itself to the heatsink will probably be quite high..
Anyway, this seems to do quite a good job, the enclosure is a lot cooler so the inside temperature must be much less.
Re: ULA running hot
I too was shocked when I touched the ULA after the computer had been running for a time. It gets hotter than hot! Bad ULAs, burned out, are I think the 2nd leading cause of ZX81 demise, after keyboard ribbon breakdown. Unlike keyboard problems however, the only simple source of a ULA replacement is a donor unit. Pity.
I use a 50x25x10mm heatsink, eBay item 140940669626 (or for a better price on 10 see item 180947515009). I put thermal compound on the central 80% of the ULA, and epoxy on the two ends to hold it in place.
This arrangement seems to work, as the heatsink becomes quite uniformly warm (rather than the hot-hot of the bare chip), and does fit into a standard enclosure.
Ian
I use a 50x25x10mm heatsink, eBay item 140940669626 (or for a better price on 10 see item 180947515009). I put thermal compound on the central 80% of the ULA, and epoxy on the two ends to hold it in place.
This arrangement seems to work, as the heatsink becomes quite uniformly warm (rather than the hot-hot of the bare chip), and does fit into a standard enclosure.
Ian
Re: ULA running hot
"zx81user" - all the 40pinDIL heatsinks I've seen (not many, admittedly) have always been a 'stick on' type, no clamp...I've seen 14/16/18 pin types that do have a 'clamp' of some kind though, so maybe there are 40pin types out there that do...????
"TMA one" - thats probably overdoing it a bit !! As I said earlier, because of the thermal resistance of the plastic package, the ULA 'chip' will probably still be fairly 'warm' even if you held the centre point of the ULA at 25C or so..!!!! Only way to really cool it would be to rig up some sort of mini-refrigeration system!!
"TMA one" - thats probably overdoing it a bit !! As I said earlier, because of the thermal resistance of the plastic package, the ULA 'chip' will probably still be fairly 'warm' even if you held the centre point of the ULA at 25C or so..!!!! Only way to really cool it would be to rig up some sort of mini-refrigeration system!!
Re: ULA running hot
Maybe a Peltier-effect heat pump device between the heatsink and the ULA???? !!!!
Re: ULA running hot
It might not be as bad as you think, there is a lot of "metal" inside the DIP package, in fact, the center layer (where the die sits) is nearly an all metal layer due to the 40 pins that are attached to the die. These 40 pins take away quite a bit of heat from the die. Cooling down the top of the chip (although the thermal resistance would be quite high) will not only keep the die in a cooler environment but will also cool the 40 metal pins directly attached to the die. So I think it is quite effective. To give it the best shot, I also drilled five 10mm holes in the case directly underneath the ULA. I feel bad for modifying the case, but if it prolongs the zeddy's life, I think it is worth it. I didn't want to go as far as drilling holes on the top of the case though!! That would severely impact the sleek design!gozzo wrote:because of the thermal resistance of the plastic package, the ULA 'chip' will probably still be fairly 'warm' even if you held the centre point of the ULA at 25C or so..!!!!
For my application in development I need to run the zeddy 24/7 so it's essential that it will survive
Michel