Just a thought that in what is essentially a closed system what good do you think those heat sinks will serve anyway?
One of the reasons the reg heat-sink stays red hot is because there is no through put of air to cool it.
Once those sinks reach thermal equilibrium with the chip, without cooling air they wont radiate/convect heat efficiently if at all and become become a none event. If you don't believe that then remove the fan from the CPU on your desk/laptop, stop up the vents to a minimum like the speccy and see how you get on.
All the claims I've seen for how well they reduce temps etc have been on open to air systems ie board removed from the case with the ambient surrounding air doing the job of cooling the sinks as it should do. I've yet to see one where they drill a hole in the speccys case insert temp probe under the ULA say, reseal the hole, run it for half a day then tell us the result.
If you want to get rid of the biggest heat source and at the same time preserve the membrane, fit a switching regulator and let the chips breath, in what little air there is, without the reg heat-sink hogging the limelight.
Fitting heatsinks to chips that don't get very hot (Z80, ROM, RAM, 74 series logic etc) does not gain you anything.
I'm not sure the Spectrum ULA needs a heatsink, on account that back in the 1980's I left my issue 3 Spectrum on for days at a time and it was fine.
As Moggy says, replace the 7805 voltage regulator with a Recom or similar modern switch mode regulator. That will reduce the heat in the case, allowing the rest of the circuity to run in a cooler environment.
And the components that like heat the least, are electrolytic capacitors. Have you renewed them?
Semiconductors (chips) can run up to 125 degrees C (chip temperature not the temperature of the package / case / encapsulation).
! Standby alert !
“There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb!
Looking forward to summer in Somerset later in the year :)
Just a thought that in what is essentially a closed system what good do you think those heat sinks will serve anyway?
One of the reasons the reg heat-sink stays red hot is because there is no through put of air to cool it.
Once those sinks reach thermal equilibrium with the chip, without cooling air they wont radiate/convect heat efficiently if at all and become become a none event. If you don't believe that then remove the fan from the CPU on your desk/laptop, stop up the vents to a minimum like the speccy and see how you get on.
All the claims I've seen for how well they reduce temps etc have been on open to air systems ie board removed from the case with the ambient surrounding air doing the job of cooling the sinks as it should do. I've yet to see one where they drill a hole in the speccys case insert temp probe under the ULA say, reseal the hole, run it for half a day then tell us the result.
If you want to get rid of the biggest heat source and at the same time preserve the membrane, fit a switching regulator and let the chips breath, in what little air there is, without the reg heat-sink hogging the limelight.
Yes, I absolutely agree with you.
But in fact I am running an unconventional cooler to provide air inside the case of the spectrum + and I have already determined that after a while hot air starts to flow from the opening at the back.
So you see, for me the good old method of air cooling heatsinks still makes sense. ;)
Interestingly, it is the cool / heat cycle that is suspected to cause the most problems...
In my job, our equipment runs 24 hours a day, 365 days a year. Sometimes if something is powered down for more than a few hours say because of work on the power network, the problems occur when we power up again.
! Standby alert !
“There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb!
Looking forward to summer in Somerset later in the year :)
Interestingly, it is the cool / heat cycle that is suspected to cause the most problems...
Mark
Perhaps the hardware should be allowed to get hot up to a certain point first and only then start to being cooled.
Inversely, when shut down or interrupted, cooling should stop below a certain point.
That way, materials would have sufficcient time to safely go through changes caused by temperature slowly rising and droping ?
Just my guess, but I think it would be better to allow the spectrum heat up for a while (low system load?) and then start cooling it. I would apply the same principle and not kill cooling on shutdown, but let it stay on for a few minutes to avoid some sort of heat-soaking.
Hey, it sounds like I'm talking about the engine of my old car... :P
Interestingly, it is the cool / heat cycle that is suspected to cause the most problems...
In my job, our equipment runs 24 hours a day, 365 days a year. Sometimes if something is powered down for more than a few hours say because of work on the power network, the problems occur when we power up again.
Mark
Very true going back to the old thermionic valve days it was always more viable to keep them constantly switched on rather than go through the thermal stress of constant off/on.
Light-bulb constantly on 2000 hrs light bulb on and off 1000 hrs.
@Renegade
In fairness I did say that the sinks were useless without air through-put.
Having seen the turbo Coolmaster 3000 you're using I bow humbly in your direction, definitely a throughput of air there!! B-)
With that kind of cooling after switch off I reckon a few minutes would be sufficient, as Mark says these chips are not really hot runners..
Comments
One of the reasons the reg heat-sink stays red hot is because there is no through put of air to cool it.
Once those sinks reach thermal equilibrium with the chip, without cooling air they wont radiate/convect heat efficiently if at all and become become a none event. If you don't believe that then remove the fan from the CPU on your desk/laptop, stop up the vents to a minimum like the speccy and see how you get on.
All the claims I've seen for how well they reduce temps etc have been on open to air systems ie board removed from the case with the ambient surrounding air doing the job of cooling the sinks as it should do. I've yet to see one where they drill a hole in the speccys case insert temp probe under the ULA say, reseal the hole, run it for half a day then tell us the result.
If you want to get rid of the biggest heat source and at the same time preserve the membrane, fit a switching regulator and let the chips breath, in what little air there is, without the reg heat-sink hogging the limelight.
I'm not sure the Spectrum ULA needs a heatsink, on account that back in the 1980's I left my issue 3 Spectrum on for days at a time and it was fine.
As Moggy says, replace the 7805 voltage regulator with a Recom or similar modern switch mode regulator. That will reduce the heat in the case, allowing the rest of the circuity to run in a cooler environment.
And the components that like heat the least, are electrolytic capacitors. Have you renewed them?
Semiconductors (chips) can run up to 125 degrees C (chip temperature not the temperature of the package / case / encapsulation).
Mark
Repair Guides. Spanish Hardware site.
WoS - can't download? Info here...
former Meulie Spectrum Archive but no longer available :-(
Spectranet: the TNFS directory thread
! Standby alert !
“There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb!
Looking forward to summer in Somerset later in the year :)
This is the regulator I use in my ZX81's and speccies...
http://uk.rs-online.com/web/c/power-supplies-transformers/dc-dc-converters/switching-regulators/
Yes, I absolutely agree with you.
But in fact I am running an unconventional cooler to provide air inside the case of the spectrum + and I have already determined that after a while hot air starts to flow from the opening at the back.
So you see, for me the good old method of air cooling heatsinks still makes sense. ;)
Of course, Mark. :D
In my job, our equipment runs 24 hours a day, 365 days a year. Sometimes if something is powered down for more than a few hours say because of work on the power network, the problems occur when we power up again.
Mark
Repair Guides. Spanish Hardware site.
WoS - can't download? Info here...
former Meulie Spectrum Archive but no longer available :-(
Spectranet: the TNFS directory thread
! Standby alert !
“There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb!
Looking forward to summer in Somerset later in the year :)
Perhaps the hardware should be allowed to get hot up to a certain point first and only then start to being cooled.
Inversely, when shut down or interrupted, cooling should stop below a certain point.
That way, materials would have sufficcient time to safely go through changes caused by temperature slowly rising and droping ?
Just my guess, but I think it would be better to allow the spectrum heat up for a while (low system load?) and then start cooling it. I would apply the same principle and not kill cooling on shutdown, but let it stay on for a few minutes to avoid some sort of heat-soaking.
Hey, it sounds like I'm talking about the engine of my old car... :P
Very true going back to the old thermionic valve days it was always more viable to keep them constantly switched on rather than go through the thermal stress of constant off/on.
Light-bulb constantly on 2000 hrs light bulb on and off 1000 hrs.
@Renegade
In fairness I did say that the sinks were useless without air through-put.
Having seen the turbo Coolmaster 3000 you're using I bow humbly in your direction, definitely a throughput of air there!! B-)
With that kind of cooling after switch off I reckon a few minutes would be sufficient, as Mark says these chips are not really hot runners..