Those are initially intended as rifle carrying cases. Of course, you can store whatever you want. Just saying what they are, so you can look for other options from other brands, not just from Pelican.
Putting this here…so you see what’s been done in this market sector:
https://cotscomputers.com/tag/rugged-workstation/
One other thing to note…since you’ve mentioned desert and heat…so I’ll assume outdoor is your operating environment. You’ll need a panel (or three) that can do 1000nit or more.
Wow, that is a heck of a laptop!
Yes finding display panels is going to be one of the trick parts, need to find 12-17" DP/eDP panels with high brightness at some sort of tolerable price. I think there are options but it isn’t trivial
i am not sure a pelican case with the lid open will still be considered as a laptop by that point.
I guess you still can put it on your lap and use it, so I guess we have that.
I can totally see it being useful, although maybe not for me atm. Maybe if I am a RC lunatic and I fly drones/drive cars around a park all day I might want one of those.
There are a few problems beforehand. First, cooling – if you want the case to be water-proof, you must need to find a way to cool the system without breaking the seal to the electronics. You can’t just stuff the board into a enclosed space and expect it to not cook-off.
You can get away by mounting the motherboard upside-down in the bottom of the case and make the entire case bottom out of aluminum and carve out channels for heat pipe and cooling fins and then just thermal pad/paste every inch of the board. I think that will be enough passive heat dississipation.
If not, you can make a water-tight “cooling chamber” where liquids (air or water) can be forced through cooling fins with heat pipes with maybe a water-proof fan.
Or alternatively, make an open-loop water cooling system with hose for radiators on the outside.
May be a bit extreme, but that’s a starting point I guess.
I think you can have a few “drawers”, if you would like, on the sides that will open up that will grant access to the ports and allow for equipment to be stored/strapped in to. Of course, they will have gaskets for waterproof. Should not be that hard to make.
Just don’t skip leg day.
Maybe some Raspberry Pi tomfoolery could help here? I haven’t had the chance to read it but Maybe this could add value to your project?
I look forward to reading more updates about your project!
Best Regards,
Varg
I’ve been working on a rough CAD model, and am planning a big update once that is in a state worth showing, but I’m realizing a lot has been refined since my last update so I should give a bit of an update now.
Case:
After looking at the number of mods that would need to be made to a pelican case, it isn’t worth using an off the shelf case if we are just going to swiss cheese it.
Instead, the laptop will be build into an aluminum and polycarbonate chassis. since we have good access to materials and machinery at our university this will also help keep costs down. This will also make the final product much nicer as it will be a lot more compact, or at least relatively speaking. The current dimensions when closed are about 900x250x75mm. It will also feature constant torque hinges for the display that means the lid will act like any other laptop.
Displays:
I hadn’t ever tested my framework outside, turns out the 450nit display at full brightness is actually probably close to the best you can get from a consumer laptop display. It’s not as ideal as a fancy 1000 nit display, but it’s a lot cheaper, more accessible, and we know it’s compatible, so the laptop will feature 3 framework display panels. (can directly use the work done here)
Router:
After some discussion about our mission control setup, we decided we’d probably prefer to keep this separate. as such it’s been removed from the scope of the laptop. We will probably take an off the shelf router and build a ruggedized enclosure for it, which will serve as a good test of construction techniques we plan to use on the laptop.
Battery:
It’s a complicated one, still weighing options to figure out what will be easiest to implement, while giving good performance.
Thermals:
@Xavier_Jiang you are correct, thermals have been a big concern that has been on my mind for a while now, since we are dealing with just about the worst possible scenario. passively mounting the mother board into the aluminum frame may work, but the issue is the laptop needs to operate in high temperatures (at least 40C), which means the cooling system needs to be very efficient as there isn’t a large temperature gradient to ambient.
I have been looking at heat pipes through a bulkhead wall to a weatherproof fan and heatsink, this is what is done on off the shelf ruggedized laptops, and may be the best option, but is more complex than I’d like. the MVP plan is probably to use powerful well filtered and shrouded fans to drive air through the case, and leave the fan and heatsink alone. Although this seriously effects it’s weather resistance, so I’d prefer to avoid it.
Payload bays:
They will essentially look like mini rack mounted devices that slides in the back of the laptop and have a gasketed seal with many fasteners to hold them in and sealed. this works well as all of our planned payloads will need external connectors for antennas.
This is a big concern in industrial situations where there’s lots of heat-producing equipment needing to be kept cool, packed into a NEMA 4/NEMA 4X (weatherproof) or even NEMA 7 (gas-tight, explosion-proof) sealed enclosure. Many solutions exist, from heatsink heat exchangers to heat pipe coolers to thermoelectric (TEC) air conditioners to vapour compression air conditioners. Unfortunately many of these solutions are expensive and most are very power-hungry and unsuitable to battery power. But just to get you thinking and pointed in some direction…
Probably the easiest and most effective would be fans, like https://noctua.at/en/products/fan/industrial with lots of filtering, downwards ducting or maybe even labyrinthine ducting to keep water out.
the problem with heatpipes through a wall is that, well, heatpipes might have a different shrink rate than the surrounding material, which might compromise the seal.
What I was thinking is similar to those cheap “laptop water cooling kit” where you would have hose on the wall instead of pipes, and (perhaps in conjunction with passive cooling through the chassis) you can pipe external radiators to it to help cool it. You can even spec the hose to work with some of those (really cheap and janky) aftermarket coolers. I think it is easier than with weather-proof coolers mounted internally.
There could be a fused/isolated power for the external pump, or the pump can be internal, but you still need fans on the external radiator. Unless you just opt to use a big clean water reservoir (like a pool?) to draw away the heat. The only problem with that is the plumbing might need cleaning afterwards, but should not be a big deal.
Or a lot of airflow.
My rather old Dell E6420 XFR (IP65) handles thermal like this in one of its modes: Fan off until 90c, then burst air to lower the temperature to around 50c, then fan off. It has this rollercoaster thermal profile. However, this only works because it has a fairly large mass to sink the heat into.
Some info here:
FYI, I’ve been running a 2011 i7 (another laptop) for an average of 95c for 36 hours each week, every week…and it’s still rocking. That’s to say, you don’t really need to worry about CPU thermal too much, not even if it runs into the 90c+. The processor will thermal protect itself.
Also, if I’m not mistaken, you didn’t mention any ingress protection requirement related to your anticipated operating environment.
Notice that IP65 is regarding jet spray…not submerge into water.
That is, it’s not about being water tight (in the case of IP65), but about making any seams not directly exposed to take on pressurized water ingress.
I don’t know of any IP67 laptops…only IP67 tablets.
@Second_Coming Thanks for the info about your laptop, it looks like it also uses the heat pipes through a sealed bulkhead into a separate unsealed thermal chamber.
I do trust the CPU to thermally regulate itself, and the desert gets hot enough that we may run into that regardless. This will essentially be an exercise in trying to cool it as effectively as possible to improve the performance.
As for Ingress protection, since we aren’t designing a product, complying with a specific standard isn’t as critical as ensuring it meets our ingress protection requirements.
Essentially it needs to be IP65, with the additional requirement that it should withstand being placed in very shallow puddles, so the bottom must be fully watertight.
In case you are curious this is the current requirements list:
| Number | Name | Description |
|---|---|---|
| 1. Environmental | ||
| 1.1 | Temperature | The laptop must be designed to handle from -30 degrees to +40 degrees, in direct sunlight. |
| 1.2 | Weather | The laptop must be able to handle heavy rain and snow, when either open or closed, and must withstand high winds while closed, or open with support. |
| 1.3 | Puddles | The laptop should withstand being placed upright in puddles up to 20mm in depth |
| 1.4 | Sunlight | The laptop must be reasonably usable in direct sunlight for short periods of time, and comfortable to use for extended periods when in light shade or better. |
| 1.5 | Dust | The laptop must be completely resistant to dust. Fans and filters may have only limited protection, but must be easily replaced in the field if prone to damage. |
| 2. Physical | ||
| 2.1 | Weight | The laptop should be light enough that it can comfortably be carried for at least 15 minutes when closed. |
| 2.2 | Size | The laptop should be small enough to be moved by a single person reasonably easily when either open or closed. |
| 2.3 | Durability | While conducting analysis or testing of this is out of scope, the laptop should be built to be very durable, think surviving being dropped out of the bed of a truck and then dragged 100ft across gravel. |
| 2.4 | Ergonomics | The laptop should be designed to be reasonably comfortable to use for two hours, such that it can also be used as a regular computer for programming, talking on slack, etc, as required. |
| 2.5 | Labelling | All Connectors and switches should be clearly and professionally labelled using a method that will withstand the required abuse and conditions. |
| 3. Performance | ||
| 3.1 | Computer Specs | The laptop must have plenty of processing power, RAM, and solid state storage, such that it can easily run as a fully capable windows machine. |
| (Recommended: i5 11th gen, 16GB of RAM, 500GB SSD or better) | ||
| 3.2 | Monitors | The laptop must have at least two, preferably three, built in displays, each with a resolution of at least 1080p. |
| 3.3 | Battery | The laptop should have a battery life of 12h, and be able to reach full charge in 5h when not running to allow short overnight charging. |
| 4. Connectivity | ||
| 4.1 | Networking | The laptop must have a standard wifi card, and an accessible ethernet port. |
| 4.2 | Human Interfaces | The laptop must feature a full sized keyboard, and some combination of a mouse, trackpad, drawing tablet, or similar, with consideration put into usage while wearing gloves. It must also feature an aux jack to allow online calls, speakers, a mic, and a webcam are desirable. |
| 4.3 | Payload Bays | The laptop should feature two modular payload bays that provide power and data. The first of these bays is currently planned to be used for live telemetry, with the second reserved for future expansion, potentially integrated . |
| 4.4 | External IO | In addition to previously listed connectivity, and any internal connections to payload bays, the laptop should feature one external thunderbolt port and two USB A ports, with consideration put towards the viability of a built in SD card reader. |
| 4.4 | Power | The laptop should be compatible with the standardised GSE connector for power / charging, as well as having an AC charger. |
Yeah. The heatsink fins act as deflectors to shield the water jets / droplet projectiles from reaching the seal at speed. By the time the water reaches the seal…those droplets would have their velocity minimized substantially.
You may not want to do that. Doing so would transfer any shock to the connectors. Unless you can ensure the entire structure has no give, no flex, a completely ridged body (and internal). Even shock with the tiniest displacement (e.g. 0.05mm) could break connectors / joints or cause it to fatigue over time.
Alternatively, you may want to think about a decoupled / roll cage type design, where the internal and external are not tightly coupled, and that material compliance is part of the design.
For thoughts (Zach likely knows this from engineering…so it’s for others really):
We prefer proper tightening and the use of thread locker, along with proper wire management over any use of epoxy. when given the chance we always use connectors with positive locking mechanisms, for things like USB-C into the main board where we don’t have a choice we may design retention clips. If we have learned anything from years of testing in all sorts of harsh environments it’s that Murphy’s law is real and it will get you, so much like the Framework, reparability is key. particularly, since things tend to go wrong at the worst possible time, being able to conduct repairs with simple hand tools is also important.
As for conformal coating, this is a standard practice for our flight hardware, and we will likely do it to any custom PCBs in the laptop. That being said, conformal coating can mess up hardware in a number of ways, particularly seeping into any and every connector, and reducing heat dissipation on components, as such in the past we have avoided applying it to off the shelf systems in the past since the risk of damaging the system seems to outweigh the potential benefits. If the main board is exposed to any dust or moisture, clearly something has gone wrong with the case design.
There is definitely some fun we could have with designing a ridiculously robust case, although, simply ensuring everything is very rigidly mounted and reasonably tough is probably sufficient. We don’t actually expect to be routinely dropping, kicking, or dragging it. It’s an expensive bit of kit and we plan to be very careful with it, the goal is just to make sure that if something does happen, it will hopefully be ok.
the problem is what if it is submerged into water
ATM I think the discussion is more on "what do you need to do if you want to make the device ip 67 proof)
Which frankly is overengineering. He probably can walk away with ip 65 definitely.
In that case simply get the fans some duct and draw/dump air from the bottom of the machine, and make light seals where the duct meet the current cooling stuff so in case water accidentally get in it will not fry the logic board. Add some simple grill/filter for protection against sand/rock particles.
Requirement 1.2 goes towards IP65.
Requirement 1.3 goes towards having 20mm tall laptop feet hahahaha 
Requirement 2.2 goes towards sending someone to the GoldGym.
I would imagine him putting it on like a collapsable stand/table/working surface since, well, if you are field testing something you probably have more than just a laptop (e.g. battery charger, monitor for fpv cameras, etc)
A few feet. not just a couple centimeters.
That’s just for when, for example, the wind blows. or when the rocket (or whatever he is testing) blasts some off.
Snow/rain should not be a big problem, since you can just make sure the vent/intake will not have water go back up in to them. which, because gravity, can be fairly difficult.
Resin on connectors? No. You need thick rubber pads on corners and plastic “exoekeleton”, or a internal chassis/shock absorbing frame (much like the ones in the Casio G-Shocks) to take on some of the impact.
Never use glue or epoxy on electrical connectors. The contacts can potentially be accidentally isolated, or bridged/shorted.
You can tape them down (with fabric tape), if you wish. Although I don’t think it will help significantly.
Soldering (directly to the board) will not help significantly, as repeated flex is known to cause BGAs to fail. Add additional support (plastic pillars) to the board if necessary.
Although, I don’t think said system is/will need to focus on shock absorption.
Haha, always good to question what the requirements dictate. If our our team wants to hit the gym and get jacked then yeah I guess I can make this thing pretty heavy.
As much as the feet where a joke, not a bad point, putting it on something like 10-15mm feet is a good way to give it a bit more protection. I was trying to figure out spacing of the connectors, which I only trust to IP65, without making the whole thing thicker, and this is a good solution.
@Xavier_Jiang you are correct, the nominal use case for this is being used on a folding table for mission control operations, well above the mud/snow/whatever. However it will also get used for field debugging of systems, which means it will get placed on just about any possible surface, including soft ones like wet grass and snow, so it does need to be sealed at the bottom, even with feet. (no bottom mounted fan vents 
, heck I nearly overheat my regular framework by accidently leaving it on soft surfaces all the time.)
aah. so …
yeah. not sure how the cooling will play out.
I think you will have to passive the sides (and bottom). Which isn’t a huge deal, except you also need to consider shock absorption. Although heat pipes are at least somewhat flexible, so you can get away with it so long as there is no rigid support for them near the CPU. You don’t want to crack the die.
I would imagine you will have long-boi pipes stretching from the CPU block to “transfer blocks” on the inside of the walls, so you can remove the pipes for service.
