Home Theater Forum and Systems banner

21 - 34 of 34 Posts

·
Registered
Joined
·
50 Posts
Very clean....

...As I understand it, the temperature is not the temp of the actual heat sink or even relay. It's the temperature inside the box. So, for example, I was told to put a thermometer probe found in a kitchen inside the box. If it's under the ambient temp rating all is well.

Wiring up a box is one thing. Understanding it all is another. :rubeyes:
Thanks! Kinda simple in the end but did the trick nicely.

As to the temp; what paulcet said. The junction. Without any heatsink it would be the back of the relay.
 

·
Registered
Joined
·
281 Posts
400A for 10ms. The surge current rating is inversely proportional the surge duration, so given up to theoretical 20x for toroidal trannies and my 8A 'max' load I'm looking at a worst case inrush of 160A, which leaves plenty of margin. Which is good given it's a zero crossing SSR and theory could be "worst case" every time I power it on...
I won't ask you to explain this as I can easily Google it myself. Let's just say that as of this moment it's way over my head.

I don't think this potential issue was ever mentioned anywhere else.
 

·
Registered
Joined
·
281 Posts
The specs on the data sheet are for "ambient" temperatures, assuming the relay (and heatsink if applicable) are all in the same enclosure. In actuality, the critical point for temperature is at the "junction" of the relay which is right behind the metal plate. There is no practical way for you to measure that temperature, so it isn't published in many cases. In your case, the heatsink, and effectively even the plate is outside of the enclosure. So the enclosure temperature isn't going to matter much. Monitor the temperature near but not touching the heatsink. You could orient the fins vertical to get a little convective airflow.
So .... your post here launched a day long investigative task for me.

Turns out that vertical, as you suggest, would be parallel to gravity. Not facing towards the ceiling as I originally thought and have pictured in my thread.

This investigative task also led me to realize that the heat sinks I purchased and suggested, even though impedance rated the same as the Omron recommended heat sinks, have other variables that one needs to consider.

Most importantly they are rated for that resistance with a certain amount of air flow vs. a natural convection design. Also to note is the heat sinks I've been looking at are tested a limited size heat producing product in contact with the heat sink. This is OK, if you understand that the heat sink becomes more efficient the larger the surface area of the heat producing product becomes.

A lot to understand here. Nothing is every easy, but it always seems to be fun. :)
 

·
Registered
Joined
·
100 Posts
Yes, there is a lot of work in designing power systems. Here is one of mine. Banks of 125A SSR x 3 on each heat sink, fan cooled, in an enclosure containing 14 banks. 8 enclosures total. These get quite warm to the touch, and if a fan fails, the heat sink is quite hot to the touch, but the system is designed to tolerate that fault for up to 6 weeks, with the possibility of premature failure of the SSR.



Here is another, quite a bit smaller. Heat sink is on a TO-220 package 3.3volt regulator. Designed to operate on 5-24VDC. Enclosure size was constrained, so on 24VDC, this one gets pretty hot. I have 3 in operation at 24V for the past 6 months.




The guy at Power I/O may have found a good correlation between the inside-enclosure temperature and the suitability of the heatsink. Or maybe there is some misunderstanding somewere. I've been doing industrial electronics and engineering for 20 years, but I don't know everything either.
 

·
Registered
Joined
·
100 Posts
Retrofit of mercury relays in a glass tempering furnace. You wouldn't want to be looking into the cabinet when one of those failed. They go BOOM and mercury goes everywhere.
 

·
Registered
Joined
·
281 Posts
Yes, there is a lot of work in designing power systems. Here is one of mine. Banks of 125A SSR x 3 on each heat sink, fan cooled, in an enclosure containing 14 banks. 8 enclosures total. These get quite warm to the touch, and if a fan fails, the heat sink is quite hot to the touch, but the system is designed to tolerate that fault for up to 6 weeks, with the possibility of premature failure of the SSR.



Here is another, quite a bit smaller. Heat sink is on a TO-220 package 3.3volt regulator. Designed to operate on 5-24VDC. Enclosure size was constrained, so on 24VDC, this one gets pretty hot. I have 3 in operation at 24V for the past 6 months.




The guy at Power I/O may have found a good correlation between the inside-enclosure temperature and the suitability of the heatsink. Or maybe there is some misunderstanding somewere. I've been doing industrial electronics and engineering for 20 years, but I don't know everything either.

Those heat sinks look massive. What is the actual size of those things?

It appears from that pic the heat sinks are on their sides. That tells me the position of the heat sink might not be as detrimental as the actual design. Plus you have fans....

It's almost counter-intuitive to me. I tend to think heat rises so you would want the heat sink on top. But a little digging and things like natural convection comes up.
 

·
Registered
Joined
·
100 Posts
Those heat sinks look massive. What is the actual size of those things?

It appears from that pic the heat sinks are on their sides. That tells me the position of the heat sink might not be as detrimental as the actual design. Plus you have fans....

It's almost counter-intuitive to me. I tend to think heat rises so you would want the heat sink on top. But a little digging and things like natural convection comes up.

They are massive. Probably weigh more than 5 lb. They are from Wakefield Engineering, I think the SSR mounting surface is 5" wide by 7" tall. The extrusion I believe is 6" deep. These have been installed for a couple of years, so the memory is foggy! :eek:lddude:

The heat sink in the picture is not on its side, per se. It is arranged so that vertical air flow is not impeded by the fins. In other words, the fins are parallel to the flow of air. In this case the fan is mounted to the bottom of the heat sink, and blows upward. You can see the edge of a fan near the bottom of the pic.

The flow of heat by "natural convection" in air does go up due to differences in air density caused by temperature differentials. But the interface between the SSR and heat sink causes heat CONDUCTION. This is not affected by gravity. So the SSRs are just as happy to conduct heat to the heat sink 'sideways'.
 

·
Registered
Joined
·
100 Posts
Which relays? Mine? No, they are typical silicon solid state relays. Carlo Gavazzi. Very pleased with their performance.
 

·
Registered
Joined
·
281 Posts
They are massive. Probably weigh more than 5 lb. They are from Wakefield Engineering, I think the SSR mounting surface is 5" wide by 7" tall. The extrusion I believe is 6" deep. These have been installed for a couple of years, so the memory is foggy! :eek:lddude:

The heat sink in the picture is not on its side, per se. It is arranged so that vertical air flow is not impeded by the fins. In other words, the fins are parallel to the flow of air. In this case the fan is mounted to the bottom of the heat sink, and blows upward. You can see the edge of a fan near the bottom of the pic.

The flow of heat by "natural convection" in air does go up due to differences in air density caused by temperature differentials. But the interface between the SSR and heat sink causes heat CONDUCTION. This is not affected by gravity. So the SSRs are just as happy to conduct heat to the heat sink 'sideways'.

I don't know what it is, but just hearing 5lb heat sinks makes me feel all warm, fuzzy and silly inside. :neener: That's awesome to see.

The conduction interface you speak of. Is that just between SSR and heat sinks or is it the same for for CPUs or GPUs, or power mosfets or?

:gulp: In reality I could set my power boxes up so the relays are cooled by the amp fans :unbelievable::unbelievable:
 

·
Registered
Joined
·
100 Posts
I don't know what it is, but just hearing 5lb heat sinks makes me feel all warm, fuzzy and silly inside. :neener: That's awesome to see.
LOL! Nothing like thousands of watts to make a guy all goofy. :dumbcrazy: About 2 million when the furnace is at full power.

The conduction interface you speak of. Is that just between SSR and heat sinks or is it the same for for CPUs or GPUs, or power mosfets or?
Conduction works that way for anything. Including fingers/saucepan!

:gulp: In reality I could set my power boxes up so the relays are cooled by the amp fans :unbelievable::unbelievable:
Sure could, as long as the air temp coming out of the amps is under 40°C.
 

·
Registered
Joined
·
281 Posts
LOL! Nothing like thousands of watts to make a guy all goofy. :dumbcrazy: About 2 million when the furnace is at full power.


Conduction works that way for anything. Including fingers/saucepan!



Sure could, as long as the air temp coming out of the amps is under 40°C.
For sure!

What is this furnace heating?
 
21 - 34 of 34 Posts
Top