Transcription
6:18
Hello everybody, and thanks for attending our webinar today, we’re going to give it just a couple more minutes, for people are still filtering in. But will be with you shortly.
7:48
Hello everybody and welcome to Advanced Cooling Technologies, November webinar. We’ll be covering system level thermal solutions for military grid technologies. Our speakers today are Matt Keller, Devin Pellicone.
7:59
Matt is the general manager of our York Pennsylvania location where he oversees design, production and service operations.
8:05
Armed with over 17 years of engineering experience map, provides invaluable expertise and ensuring … customers get the best thermal solutions and technical service possible.
8:14
Devin is the lead engineer of ACT’s industrial products group at Lancaster Pennsylvania location.
8:20
With over 10 years of experience designing and building both passive and active two-phase cooling systems for a wide range of applications, including high performance computing in power electronics cooling.
8:31
Before we get started, I want to let everyone know that there will be a live question and answer portion at the conclusion of today’s presentation.
8:38
If you have any questions, feel free to submit them through the present throughout the presentation, by using the question function on the Dashboard.
8:45
Matt and Devin will be will be answering as many as time allows. We’ll follow up with an e-mail for anyone with questions or not answered live.
8:52
With all that said, I’ll turn things over to our speakers.
9:04
Thanks, Kevin.
9:06
Good afternoon, everyone. Thank you for joining our webinar today, Tell you a little bit about Advanced Cooling Technologies.
9:13
We are a thermal management company located in Lancaster Pennsylvania is our headquarters and we also have a branch in York, Pennsylvania.
9:20
We were founded in 2003 and we now have over 200 employees at both locations. We have approximately 140,000 square feet, facility space and growing every year, and we have an ISO 9001 and 9100 quality management system.
9:36
We’ve won a number of awards through the military and aerospace product innovation awards for the different technologies to be developed over the years.
9:43
Small Business Innovation Research Grant.
9:51
So we want a number of tidbits awards or that work and we have numerous patents and scientific publications related to thermal management technologies.
10:08
Today we’re going to be talking to you about some of our system level cooling technologies, particularly Environmental Control Units or ECUs … coolers and chillers on some of our pump to Face, or PTP cooling Systems. As well as, Phase Change materials PCM. And we’ll end with a little bit of discussion about some of our electronic systems, system controls, and packaging capabilities that a little bit more about our testing, an extra capability at the end. And by Kevin mentioned, we’ll be ending with some question-and-answer session to answer any questions you might have.
10:40
I’ll turn it over to Matt Keller to start things off.
10:48
When we talk about system level thermal solutions, we’re referring to equipment and accessories utilized to reject the heat generated by system components to the exterior environment.
10:58
That’s usually the ambient condition.
11:03
Ambient design conditions can be anywhere from -50 to 140 Fahrenheit and 0 to 10000 feet elevations for terrestrial systems.
11:13
In some cases there may be a chilled water infrastructure which would be the exterior environment we reject would reject T two. For example, shipboard chilled water or facility chilled water system.
11:28
So, the first thing we need to understand is the design criteria. What are the heat sources?
11:33
Electronics are going to be a big one in pretty much every system these days and they’re going to be sensible heat.
11:39
That may be a very continuous load, or it could be transient in the case of directed energy systems.
11:45
Occupants may be present in the system and they provide both sensible and latent heat loads that need to be addressed.
11:54
The envelope of this system allows conduction from outside temperatures to inside, as well as additional thermal loads that come through due to solar radiation on the exterior skin of a system.
12:07
Outdoor air may also be allowed in the system if we’re going to be providing fresh air for pressurization and making sure any leaks in the system leak out as well as ventilation for the occupants.
12:20
Once we know what the lows in the system are, we also have to understand what we’re trying to maintain in the system, so what’s what are the acceptable conditions?
12:29
For electronics, that’s usually an entering air temperature to the electronics.
12:34
With occupant’s, it’s more of the average space temperature that’s being controlled. 60 to 80 degrees Fahrenheit is kind of a reasonable range for that.
12:44
Relative humidity below 60% is also usually required.
12:48
Traditionally we don’t get into adding humidity because of the logistical issues.
12:53
Providing a water source and maintaining a water source to keep the humidity levels up.
13:01
So one of the solutions we offer are environmental control units.
13:06
an ECU as we call them is basically an air conditioner built for military environments.
13:11
They provide conditioned air to an enclosed space.
13:15
They can be mounted to hard wall shelters or connected to solve all shelters, or hard wall shelters via flex flexible. So you see some photos here. There’s a five time you see on the back of a trailer behind the humvee, We’ve got systems in the middle.
13:34
That’s, again, a five time, possibly anytime hooked up to a software shelter.
13:39
There’s A Radar hub, cooling unit in the upper right, as well as a dual vertical system.
13:47
That’s hard model to Hartwell Shelter.
13:58
Design considerations for ECUs, Um, you bet, skipped.
14:05
OK, additionally, is there a good way to introduce outdoor air to a system, that out to where it can be pulled in a mixed with the return air? So it can be condition prior to be introduced to the space.
14:18
Also, we can do 100% outdoor air in the case of flight wine cooler.
14:22
So the picture bottom left shows 100% Outdoor Air unit that maintains temperature of electronics while sitting on the tarmac, which is the systems on board that UAV aren’t suitable to maintain temperatures when you’re setting it up to 4260 degree F conditions.
14:42
Issue design considerations.
14:44
Air systems are a great four going redundant system.
14:49
Either if that’s for additional capacity, parler condition kind of capacity, or complete redundancy.
14:57
With air systems, you can use … and Duck’s back Jeff Dampers, to kind of manage those connections.
15:05
And redundant systems are great for unmanned or mission-critical systems.
15:10
Capacity ratings for ECUs is determined by the ambient condition. So if the ambient condition is higher temperature, that will decrease the capacity, as well as the return air conditions. So, temperature, humidity, airflow rate.
15:25
basically, the more energy you bring back to the unit, the more cooling it can do.
15:29
So when you, when, you think, you know what load, you need, understand return conditions, understanding, the ambient conditions will affect the, rating that equipment.
15:38
When you’re evaluating solutions, you really want to understand what the capacity you’re being told is, is based on.
15:46
Somebody may call something a five time unit. We would call a three time unit, depending on the rating conditions.
15:54
Airflow pattern, this space, is also very important. These drawings kind of show a couple of different ways. The Airflow could be managed.
16:01
On the left, you have short cycling. See of air coming out of the ceiling goes right into a return. An occupant and the loads don’t see all of that air.
16:11
On the other hand, the far-right eyes really well, how you’d want to do a data center electronics type environment where you’re providing cool air to unoccupied space. That’s then available for the electronics, and the heat is then directed directly back to the unit.
16:27
Because, as we just said, the hotter the air to the, the more capacity you get out of that, you.
16:31
So, the way a units connected to a space, the way the areas move through the space is critical in making sure that they’re defined requirements are actually provided to the equipment. The equipment can actually design into that.
16:48
Another solution that we have experience with are glycol coolers and chillers.
16:53
Glycol Coolers.
16:55
We referred to here is Liquid to Air Systems, um, basically provide cooling liquid at some temperature above ambient.
17:03
So, these are coil a fan, a pump, basically, a radiator, where we can provide electronics with a bit cooler than no, there.
17:16
We’re providing liquid cooling off to pull the heat away from them, and keep them happy, but we’re not providing coolant. That’s below the ambient temperatures.
17:23
Alternatively, we can do chillers. Chillers provide cooling liquid at a fixed supply temperature.
17:28
So we can provide, let’s say, 45 degrees Fahrenheit Glycol. Even when ambient temperatures are much higher than that.
17:37
Um, so, that is good for electronics, or require low ambient temperature cooling, as well as fixed temperature cooling. So, there may be something that’s calibrated.
17:48
It needs to have a constant temperature.
17:55
Design considerations for glycol coolers and chillers. You know, when you, when you go to a glycol system, you have to make sure that the components in the system are designed for that. right?
18:04
So you need heat sinks that except glycol, electronic equipment designed for chilled fluid.
18:12
Piping and fittings that don’t leak are obviously important.
18:16
If a system’s, modular, quick disconnects are needed so that you can disconnect system components, either for maintenance or replacement. Or, if you’re going to strike a system for relocation.
18:31
We also need to be considered when you’re adding fluid, you know, in our system. Obviously, there’s no additional weight penalty to the volume of the duct system. With a glycol system, any any storage, any piping, will be filled with glycol and, therefore, that weight will be a consideration.
18:49
In addition, glycol storage tanks or … phase change materials can be used for thermal storage.
18:55
This makes a lot of sense.
18:56
When you have a system that doesn’t have a continuous load, you can downsize the system to deal with the average load and use the storage volume to address intermittent loads. This is great when you’re looking at directed energy weapons where the peak load may only be present for a few minutes out of the tower.
19:14
Devon, I’ll tell you a little bit about a previous application like this.
19:24
Thanks, Matt. This is a case study for a direct energy weapon combined, vapor compression and chiller system.
19:30
So this is using the phase change materials. So this is the solid to liquid energy storage mechanisms, this latent energy storage as compared to using a sensible. That should be just a giant liquid tank.
19:43
This is for a direct energy weapon mounted on the back of a humvee, as seen in the bottom left there.
19:48
And really, what we’re showing is how tightly packaged a system like this can be. If you needed to size the vapor compression system for the peak load of the laser in this particular application, the entire system would have been twice the size.
20:01
So we were able to significantly reduce the size of the vapor compression system by coupling it with Apps, energy storage, media, like ECM, and then, also, including the chiller system inside of this box, This entire package, you can see, mounted on the left-hand side of the humvee here, includes all the pumps. All the fans, all, the electronics, all the vapor compression components, all of it within one package. So it makes it a complete thermal management solution for a highly transient pulse. Energy loads.
20:34
Next, we’re going to switch gears a little bit here and talk about a different type of cooling technology. This is similar to what Matt was describing as our glycol coolers. except this is what we call a pump to face going, so it’s an above ambient liquid cooling solution. But in this case, we’re allowing the working fluid to oil or changing from liquid to vapor as it flows across the heated components. So it includes a pump.
20:56
It includes a condenser and evaporator in the reservoir a lot of the same components that a glycol system would have, but the working fluids are typically refrigerants or dielectric fluids that boil at a relatively low temperature and pressure.
21:09
They are hermetically sealed systems.
21:11
So we can allow that boiling temperature to fluctuate depending on what pressure we impart on the system, which is really a function of how we cool the system.
21:19
So they can be paired with some of the chiller solutions that Matt was mentioning earlier so that we can do below ambient going, but the system by itself is not capable of doing any refrigeration. We really need that refrigeration cycle coupled to the condenser in order to get below ambient.
21:35
So, some of the benefits of two-phase over a guy called Show, or most of them revolve around energy efficiency. So. Because we’re using the latent heat of the working fluid instead of the sensible heat, like we wouldn’t glycol showers or coerce, we’re able to use a significantly lower flow rate of orders of magnitude lower.
21:54
And so, that allows us to really shrink down the pump That’s required for the system packaging reasons. It allows us to spend the pump at lower speeds, and it allows the system to run with our energy consumption.
22:07
So that’s a big benefit.
22:09
Another benefit is that, because, again, we’re changing phases, this all happens at a constant temperature.
22:14
As you boil the fluid and move, from a sub cooled liquid to a saturated liquid and eventually vapor, that all happens along a constant temperature amount.
22:23
And so as you’re flowing across the evaporator, the components mounted to that evaporator see a constant temperature, which is not the case in a glycol cooler. Where the fluid is heating up as it’s flowing from one side to the other. So if you have multiple components are mounted to the same evaporator, they can all be maintained within about five degrees C of each other.
22:44
And then we can maintain 10 degrees C temperature difference across the entire, whoops, so highly energy efficient method article.
22:53
There’s a lot of considerations that are different from a glycol system or a sensible cooling system to be discussed when talking about pump to base.
23:01
The first is the heat absorption devices, we call them evaporators, because we’re changing things.
23:06
And these are very similar to the Evaporators, you may see, in a chiller or vapor compression system, they could be conduction based, meaning you mount your heat generating components directly to those parts. Or they could be air to refrigerate heat exchangers much like an evaporator coil would be in a standard air conditioning system.
23:23
Or it could be some combination of those. And there can be many different evaporators all within a given system.
23:29
one challenge that comes up with two phase cooling, when you have multiple parallel evaporators, like we’re showing in the bottom right image here, is that you have some slow balancing issues to deal with. If you have one evaporator that’s dissipating more energy than the other than the fluid with oil more, and that evaporator compared to others, it will create more pressure drop. And then you have this sort of unbalance pressure balance in your manifold.
23:53
The way we deal with that is we add Flow restructures to the system so that we can run as many parallel evaporators as necessary.
24:00
All those different modes are no load at all and be able to provide constant flow rate to all.
24:07
Another consideration that’s unique to to face growing is refrigerant or working fluid selection.
24:13
In a pump two phase system, it’s a little counter-intuitive, but the pressure drop across the loop actually results in a thermal resistance change, as we drive around the loop, the pressure decreases, and then the temperature of that fluid also decreases. And so, that reduces the amount of temperature gradient, we have to get that energy out of the fluid and into the air.
24:32
So, that Comebacks, the, the thermal resistance of the system requires us to have bigger condensers, more airflow, things like that. So what we’re really trying to do is minimize the pressure gradients around our loop and a two phase system. In the top right, we’re showing an example comparing our 134 A to our 245 FA.
24:50
There are two different vapor pressure.
24:52
Refrigerants that could be used in a pump two phase system, are 134 A is significantly higher in vapor pressure, so at 50 degree C operating point.
25:01
The resulting pressure drop of 10 PSI around a loop only has a 1 or 2 degrees C temperature impact.
25:07
But if we were using a fluid, like our 245 FA at that same temperature.
25:11
A 10 PSI pressure difference across an evaporator could have up to an 8 or 9 degrees C impact on our system.
25:19
So, picking the right working fluid for the right application, and the right conditions is really critical, and it’s something we take a lot of care with when we’re designing these fancy-based systems.
25:28
Another instance that we need to keep in mind with these systems is that the refrigerants are ozone depleting. A lot of them.
25:35
There are new logo or potential replacements for a lot of fluids, for showing some of them here.
25:41
If that is of concern for your system, we have a number of fluids in our repertoire that have either zero or very low global warming potential that we can use.
25:50
The last note is related to pumps which comes up a lot when we’re talking about pump two phase systems, refrigerants tend to be very low viscosity fluids and we’re operating very close to the saturation point of those fluids. So pump cavitation in pumps performance is a real concern.
26:05
We try to select comps that have very minimal net positive suction head NPSH and we’re usually using pumps that have positive displacement to make sure that they can accommodate low viscosity fluids.
26:16
It’s always a consideration to make sure we’re getting the right sub cooling in the loop. So pump two phase system really are a system level design.
26:23
You can’t design one component without consideration.
26:29
Some applications are pumped to face. A lot of them are electronics related. What we’re showing in the larger image on the right is actually what we would call a coolant distribution unit or CPU.
26:39
For a very large data center application. This is a 200 plus kilowatt condensing unit, and in this case, it happens to be a liquid cooled condenser.
26:47
So we’re using facility water or whatever water sources available, possibly from a chiller, to cool the refrigerant down to the saturation temperature that we want to send off to the servers to do our cooling.
26:59
Right below the condenser, which is that large orange rectangle at the top is a reservoir, because we’re changing phases from liquid to vapor. There’s a volume change that needs to be accommodated. And so we need some volume in the system to account for that difference in density.
27:14
Not all the way at the bottom are our pumps. They tend to be N plus one redundant depending on the system we’re working on.
27:20
Positive displacement, pumps, distributing that liquid to all of the many servers that this system service, in this case, I believe it was servicing up to 40 individual server blades off from one central pumping.
27:31
So that’s what I mean when I say we can handle multiple parallel evaporators, as long as it’s designed.
27:37
Some other applications on the bottom there are utilizing either air or liquid, as their condenser in the middle is a large 30 kilowatt coolant distribution unit with a sort of residential air conditioner style condenser coil that’s you shapes that we’re using to dissipate the energy to the ambient air.
27:56
So these things come in all different shapes and sizes, and can be suited for any application. And there’s also the ability to combine pumped to phase with vapor compression.
28:06
Because they’re often using the same reference to be able to do some sort of hybrid energy efficiency refrigeration cycle.
28:13
So you can use eco mode and pump the refrigerant to the evaporator when the ambient conditions allow for it. And when the ambient gets too hot, you turn your compressor on to be able to do sub ambient going to keep your system at optimal temperature all year.
28:29
We talked about phase change materials a little bit in the direct energy weapon application, but there’s other applications for this. They can be used to supplement air or liquid systems, as well as electronics, that really, we would consider these thermal batteries or thermal capacity. So any system that has a transient thermal load can benefit from having a phase change material to damp out the peaks and valleys of that load.
28:53
In a glycol system, the PCM, like we talked about, can help reduce the swap of the system to damp out those pulse loads and minimize the size of the rest of your components.
29:03
A key thing when utilizing phase change materials is that they’re often fairly poor thermal conductors on their own, and so they require quite an infrastructure of thermal paths in order to distribute the heat into them evenly and utilize all the material.
29:17
We’re showing an example of something like that, in the bottom left, where we’re using sort of a folded structure in order to distribute the heat evenly throughout all of the phase change materials, that you’re really utilizing. All that extra mass you’re adding to the system.
29:30
If you don’t have something like this in place, then you wind up with clumps of solid and liquid material in there and you’re really not getting the weight optimization.
29:41
So phase change materials come in all different temperature ranges, and big benefit here is that, again, this solid to liquid phase transition happens at a constant temperature.
29:51
And so while you’re absorbing all of that energy energy of your transient mode, you’re not increasing the system’s temperature or components temperature, or showing that schematically on the plot on the right here. You increase in temperature all the way up to the point of that material. And then you flatline while you absorb all the way to heat. And then once you have notes it, all of the material, the temperature rises again, and you’re back to sensible heating now in Liquid Phase Incentives outfits. But that not sound is really where a lot of the energy is absorbed and where you get the biggest bang for your buck out of these materials.
30:22
Schematically, at the bottom, we’re showing another example of this sort of Pulse energy system where the peak load is shown by the red bars and this is the energy that would need to be dissipated by a vapor compression system or an equivalent chiller.
30:36
If you didn’t have some sort of thermal capacitor in the system. The blue line is showing the damped out well.
30:43
And you can see very clearly that having a load shaped thermal profile reduces significantly the load on the overall system, and so all those components gets smaller, they get more energy efficient, and your system becomes more optimized.
31:00
The last thing to note about Phase Change materials here, is that, They come in almost every degree C increment from Degree, C, up to Positive 400, and almost everything in between. Those are not all the same type of Material. So, Materials compatibility is a real consideration here, but there’s lots of options for how you implement these Materials into your system.
31:22
So I’ll hand it back to Matt to talk about our controls and electronics.
31:33
Um, so when we put together Control’s package, one of the things we have to start with is, what are we trying to control?
31:41
For air systems, If you’re doing an occupied space, you might do return air, might kind of function like a house residential style system, where you’re basically looking at the air coming back, And once it gets to warm, you start to cool.
31:55
Once you get to call, to turn it off so you can cycle on and off. We also have modulating systems either through the use of a digital compressor or an EPR valve.
32:05
And that gives you tighter temperature control.
32:09
Supply air, or supply glycol control can be used for electronics.
32:14
And if there, you really do want to have a managed flow path.
32:19
So if you know that the air you deliver to the space or the liquid you deliver to, the system will go directly to the loads that that temperature can be increased, which optimizes performance of the system.
32:34
When you have poor distribution, that’s when you have to supply colder air because it will get mixed before the electrons actually see it.
32:42
Um, sometimes there’s a critical location in the system, so we might have a attempts sensor directly add a piece of equipment and you might control the whole system just to keep that one piece of it.
32:55
You may, again, can be a control point. We may use hot gas reheat or electric reheat to over, cool the air and reheat the air to wring out moisture.
33:06
Power, Joel and Inrush events are also something to consider with a control strategy.
33:12
If, if there’s not v.f.d.s in the system, then you really don’t want to bring on a big load advocate at a time when a brown out to the power system could cause issues to other system components. So, we have systems that will operate continuous compressor. And that’s not energy efficient. But, that is good to keep the power system stable.
33:34
If you’re doing some sort of scanning or radar activity where you can’t deal with reduced voltage, um, fan in pump speed modulation also may be used in the system.
33:47
We do have systems that will. The ramp down the Airflow to the shelter, when the cooling load is not at its greatest.
33:55
A lot of these shelters are not very large.
33:58
They packed a lot of cooling load into a small space, and so it can almost be a wind tunnel in there at times.
34:04
So anytime you’re not in that worst-case condition to back off Airflow, does make the interior environment a little nicer.
34:14
We do have different types of controls we use here electromechanical controls digital, digital temperature controls as well as PLC, where we can basically come up with any custom control configuration that’s required.
34:29
Local control interfaces could be as simple as an on off switch or a red to blue dial to a touchscreen that allows you to change set points by tenths of a degree.
34:41
Remote connectivity is also something that we can provide can we can do Ethernet based communication, Modbus SNMP, TCP IP as examples and use discrete connectors and have analog signals to report temperature, discrete signals to report specific conditions are false.
35:07
So, up, to summarize kind of our capabilities.
35:13
No, we want to be involved in the project as soon as possible.
35:17
We want to help steer the system level design to make sure that all the components can be optimized.
35:24
So really getting involved at the requirements definition stage is ideal.
35:29
Um, but we go through the kind of normal military design cycle with PDR and CDR, CDR, phase critical design phase.
35:39
We’ll work on mechanical packaging for the environmental requirements. We’re very familiar with Mil Standard 8, 10 testing for shock and five.
35:49
Um, all kinds of environmentalist, blowing sand and dust.
35:54
snow, Wind, solid fog.
35:58
Um, we’ve gone through lots of EMI testing, not really is something that is unique every time we do design.
36:05
We always recommend, man, if there are strict EMI RFI requirements, that we do those by test.
36:11
We’ll also do reliability analyzes at the CDR phase.
36:16
Then, you know, we will prove, procure the materials based on any float down requirements and get into manufacturing.
36:27
We do test 100% of the equipment we make for thermal performance at increased ambiance. We do that in house.
36:34
And we will go outside for qualification type testing, based on the environmental requirements.
36:41
System support, then offered with manuals, spare parts, and replacement procedures.
36:48
And we deal with component obsolescence, as we help support these systems for up to 15 years 20 years.
36:54
We offer training, service, and support, and repair and research.
37:06
I hope everyone can ask questions out here.
37:11
I think that’s all that we’re ready to introduce.
37:16
Yeah. Thanks, Ben.
37:18
We’re going to take a couple minutes here to send out a poll survey here. You’ll see it pop up on your screen. Please feel free to answer that, And if you have any questions relating to our talk, or really anything, and put them into the chat box here, and Matt and I will be answering your questions live and a couple of minutes.
41:08
Thank you, Matt and Devon, for sharing this information. And thank you to everybody to participate in our poll.
41:13
This time, I want to transition to the question and answer portion, and thank you, again, to everybody who submitted questions throughout the, throughout the presentation. And feel free to continue asking questions, And we’ll get to as many as we can.
41:24
So our first question here, how do you recommend turning the actual electronics goods in the system?
41:33
That’s a great question.
41:35
You do want to consult any manufacturer’s data first. They’re going to be the experts on their equipment.
41:41
But anytime you can run a system in Power Draw, that’ll give you a good idea of what’s happening.
41:46
You don’t want to just add up worst-case loads and assume they all happen at once all the time. That’s a sure way to oversize your system.
41:55
And that causes all kinds of problems with control.
41:58
So we definitely try to guard against that.
42:00
Yeah. It’s also important if it’s an electrical cabinet to consider the loads from the ambient. This gets neglected.
42:07
Sometimes when we’re sizing components like enclosure course, things like that.
42:12
It’s important to consider the the solar loading on the cabinet and a natural convection loadings if you’re in a hot environment.
42:18
The electrical loads are definitely one of the main components, but those things can be neglected either, so make sure you’re considering all the potential loads like Matt went over in the slides earlier. All the words that could be in your system, to make sure you’re really getting a complete thermal solution.
42:35
How do you deal with? those are all connected via GCM resulting in a relatively long melting time.
42:43
Ah, we talked about this a little bit in the in the talk. So PCM materials, Most of them are pretty terrible semiconductors. And so, what you’re really trying to do is to short circuit the PCM as much as possible, thermally.
42:58
And so that typically requires having some enhanced surface area inside of the PCM material. You want to embed high surface area materials with phase change material. Try and minimize the conduction path as much as possible. That’ll give you the best result, in terms of how much PCM you need in your system and how efficiently Every X over time. And also minimize your, your mount times. If you have to conduct through a poor, thermal interface like Liquid layer of PCM, Before you get to the solid, that’s only going to slow things down, and it’s really going to increase the temperature gradients in your system.
43:33
So it’s really a system level solution, some people think that phase change materials are as easy as dumping some wax in a box, and that’s that’s not quite going to get you there. It’s really a highly engineered system.
43:43
So we make sure we take all of that into account.
43:48
Do you ever work with a Stirling Cycle course?
43:52
The short answer to that is not really. We have done some Cryer cores in our research and development group, so Stirling cycle using Cryer cores. But most of our refrigeration systems are the ones Matt was describing, where they’re vapor compression based.
44:09
Rarely, the Stirling coolers tend to be more for cryogenic applications and we don’t see a lot of those, so it’s not to say we wouldn’t do it.
44:17
We just haven’t done done a lot of it today.
44:21
Can you talk more about how clean data serves and helpful?
44:28
Oh, there’s a lot of ways. I’ll talk about the liquid going. Maybe you can start with air cooling.
44:32
So we did mentioned that pump to phase application, which was for data center cooling. That’s kind of the new wave of handling these is really high heat load, high performance computing type of data center applications.
44:45
And the reason pump two phase is really suitable for that application is that it’s capable of absorbing a lot of energy with a small amount of energy input.
44:53
So it’s above ambient cooling solutions. So a cooling tower or some ultimate rejection system is needed on the roof of your building.
45:01
But at the server level, you’re really picking up the most amount of energy or heat per unit of energy input into the system using pump to phase going or some sort of liquid to vapor cooling technique.
45:14
And so when energy efficiency is really the name of the game and data centers that that gets you where you need to be at the server level. Oh, yeah.
45:22
I mean, even if you have chilled water in the building, know, a lot of people are not going to feel comfortable putting chilled water into this into the server area. So heat exchanger, 2.
45:34
2 phase pumping system, one of the benefits of that refrigerant is if you have a leak, it’ll it’ll week as a gas. So you’re not going to have conductive liquid all over electronics if there is a leak. So I know that pump to phase, even with a chilled water system, is still, kind of preferred at the load.
45:51
And, like you said, with the continuous or the constant temperature across the heat sink, it really allows for optimized equipment.
45:58
From an air system side, …
46:01
called I’ll, maybe terminology, are familiar with, and this is really about making sure that the cold air from the crack units, or whatever the air handler is called in the system, it’s delivered to the inlet of the equipment.
46:14
And then hot Iowa beware the racks. Blow everything out, so. You know, one of the things I’ve seen in some kinda tactical systems, as you have a hodgepodge of equipment, some have fans in the front pulling in. Some have fans in the back pulling in, and so now you don’t have all the equipment even pulling from the same side. So, very quickly, you can have a piece of equipment that gets the heat from another piece of equipment, and isn’t happening even though, you know, the air handlers attached to the system is adequate capacity.
46:41
They can’t get to the electronics you have trouble. So definitely taking into account. The direction of Airflow management of the airflow for the electronics is a big thing.
46:51
And our last product, coming back to the energy efficiency partner data centers that we didn’t talk about today, is a product called a wraparound heat pipe heat exchanger.
47:00
And this is actually incorporated directly into the air handlers, for the correct units. And what it’s helping to do is to enhance the dehumidification of the system so, you can do some pre cooling some free reheat. On the system, we also have air to air energy exchange products, where you’re recycling that air through the data center and you tend to be throwing away cold air that’s already been conditioned. You can use that energy to pre cool, the hot air from the outside, without needing to input any more energy into the system.
47:27
So it can improve the overall energy efficiency of your data center, even if you’re still using air cooling, it can even work with the liquid cooling systems.
47:35
That’s quite a lot of technologies for, for data center applications.
47:41
What is near? Moved into space going from 20 mm with exposed.
47:49
That’s a great question. We have done some … going.
47:53
It’s I won’t talk specifically, I guess, to that die. But we have developed to face cooling solutions, for up to about 300 watts per centimeter squared, Cooling applications. It depends a lot on the type of application. It depends on what your maximum temperature can be and how exotic we get with. The evaporator we’ve done some true micro channel cooling evaporators for pump to phase where we can get even higher heat fluxes. It’s a little bit R&D.
48:21
If the question is referring to emersion going so direct cooling of the junction itself.
48:28
We have a little bit of experience again.
48:29
That’s more R&D, but you’re impinging the refrigerant directly onto your exposed chip, and you can do that, because they’re dielectric fluids, and they won’t short out your system.
48:39
The word of caution there is that surface area tends not to be sufficient when you’re doing Emersion cooling. You want a lot of surface area to distribute that heat flux to the fluid, or you start to create vapor bubbles very rapidly, on your surface, and then it creates this sort of critical heat flux. Situation where you’re impinging liquid onto a vapor bubble, and it’s not really hitting your heat source.
49:01
So there’s a lot of considerations there. That’s a very challenging problem. But we do have experience with very high heat flux. Microchip cooling applications.
49:12
What is a really big factor for Howard Ross?
49:19
That’s a tough one.
49:22
No. I know. Sometimes we like to use, like, whatever KW is being poured into the equipment is the KW of heat we have to reject. Obviously, theoretically, that doesn’t make sense because some meaningful work is happening.
49:34
Um, So we tend to use like, an 85% as a rule of thumb.
49:41
But really, it’s going to depend on the application and the actual equipment.
49:53
Time, so, as a reminder, if we didn’t have time to answer your question, we’ll be following up with an e-mail. Additionally, if you think of any more questions, or if you’d like to schedule a call with our engineering team to discuss thermal management needs for your ongoing projects, you can send an e-mail to solutions at one dash … dot com.
50:09
Thank you all again for attending today’s webinar, and we hope you’ll join us again soon. Have a great day.