Scratch Build – In Progress Logic - V 01 - The cook book!

Hi

Me and modding: I created my 1st mod and water cooling solution 26 years ago and this will be my 1st scratch build. Lets just say I'm motivated to get it right.

The Planning stage!

The cook book! My recipe for scratch building. Expectations, Purpose, Hardware, Design and a load of details!

Expectations: (Updated 16. maj 2021) Imagine a PC case made of carbon fiber, where the shell can be removed by pressing two buttons and then opened by pulling the shell and the skeleton apart, this without having to remove cables ensuring immediate access to the system. The Shell is a unibody configuration. The skeleton is a component with several parts, the entire water cooling can be extracted for upgrading and testing by extracting one part of the skeleton. The water block can be connected to a quick connection bracket if the water block does not have the quick connection function. The GPU can be removed in the normal way. Soft tubes an integer part to make this work. The shell as a single piece was chosen for rigidity and to reduce the weight and complexity of the building. A stand-alone system for monitoring and control will be mounted on the water cooling section. Not a theme build unless minimalism counts as a theme. Low weight so its easy to transport. Time until its ready 12 - 16 months depending on imagination, resources and materials.

Purpose: This enclosure will be made for a PC that will be moved relatively often. It’s not an open case design, because of the risks to the components. Its main purpose is to hold data that is in use on weekly basis, for long time data storage there is a NAS available. The design is chosen to be easy to upgrade, long term reusable (12-15 years thinking green maybe?) and for maximum flexibility in daily use. The PC’s hardware components are chosen to be all round capable as a gaming PC and a Work station for data recovery, rendering and video editing, flexibility being the key word.

Hardware: The enclosure has to accommodate an ATX sized motherboard, the reason being the flexibility on a board of this size when adding other components. For virtualization of several RTOS on this one system, there has to be room for more than 1 GPU and an additional USB card. The case will be large enough for two full sized GPU’s 340 (L) mm by 150 (W) mm by 60 mm (H). Available space for radiators is a minimum 2*240 mm (in reality more like 2*290mm). The PSU can also be 230 mm in length with space for cables.

NO RGB components unless my nephews force me with their built-in persistence.

I am still researching materials, production methods and tools for the job and since I have next to no experience with scratch builds, especially not at this level, I have read a lot and watched a lot of videos on YouTube - I will add a list of my favorites.

Design: Simplify design bring down complexity and then try to make it better. This is where I start.

 

My process: To be able to create the PC enclosure, I started with the purpose, hardware and design then I looked for a shape on www.pinterest.com that would fit my purpose. Some sketches where made of the cabinet with components. After that it was all about the details so a list of all the components of interest where created with their dimensions (largest available size was noted), to see what framework I had to work within. The access to materials and production methods is also being evaluated, since I am starting from scratch.

Paper cuts are used as a substitute for 3D software. The paper cuts are made with the dimensions of the components at a scale of 1:4 to see if they would fit within the sketch of the enclosure, also made to a scale of 1:4, and if the directions of air entering and leaving the case are according to plan. The cut outs are moved around to see where I could get the best fit.

Drawing the case from different perspectives, placing paper cuts of components from a specific perspective takes a few iterations, ok I’m lying it takes a lot of iterations. All of this is to make sure that components will fit and also to give me time to work with the pro and cons of my choices, before getting input aka constructive criticism.

Routing: After the placing of the hardware components on the sketches, the routing of the wires and the water tubes (including the cooling system) is added to the sketches.

To early for this part - The Modular Setup requires thoughtful planning on how and where to connect/disconnect cables with data, power and tubes with liquid.

 

Material: In the beginning my choice was aluminum for the shell, as it is the 3rd best heat-conducting metal after silver and copper with a lower price and weight than both, important when taking transport into consideration. Aluminum's ability to get rid of energy to a cooler environment without active cooling results in lower noise levels and power consumption, both low hanging fruits/benefits of aluminium. The system hardware has high power consumption by default, but that does not mean that it has to be increased unnecessarily.
* As an alternative, I started researching composite fiber materials for the parts to lose weight, but it will have a negative impact on the passive heat transfer. Might have a positive effect on the production method and the need for tools aka cost.

I am also researching additives that shift the material properties. Materials that are stronger then the original or have less friction.

Vibration dampening: The water pumps, hard drives and fans are mounted on rubber dampeners. The case has rubber feet. I might use silicone instead of rubber, silicone is easier to work with and has a higher resistance to heat.

tool-less drive trays - On the ToDo list

 

Thermal expansion is worth considering on a project if combining two or more materials to avoid skewing, cracks and fractures. I am using The Thermal Calculator to check the thermal expansion and to make sure I plan a gab that is wide enough to ensure functionality.

Compared thermal expansion: Comparison of different materials listed later.

Intelligent use of materials: Engineering add material where its needed and remove it where its not needed. Use the right material for the right job.

Structural forces: Stress, compression, internal stress, shear moment, bending tension, bottom, cross section, trusses, strong, effective, gusset plate, joints, tension, frame, base shape 3, Yea I'm still looking into this area.

Galvanic corrosion: When we pair 2 different conductors (conductive materials) we create a battery, the current is flowing from 1 conductor to the other and corrodes the metals (materials), we really want to avoid this. A volt meter will show you if it’s conductive. We can avoid this by using a insulating layer.

Anti-static: Selection of Antistatic Agents for Polymers

Recycling: Will depend on material choice and on location.

Planned obsolescence: In economics and industrial design, planned obsolescence is a policy of planning or designing a product with an artificially limited useful life or a purposely frail design, so that it becomes obsolete after a certain pre-determined period of time upon which it decrement ally functions or suddenly ceases to function, or might be perceived as unfashionable. Examples are clothing (fashion), mobile phones (iphones), cars (year) etc. The problem is us the consumers, we are imprinted with stupidity when it comes to these types of products! I'm trying to avoid this by a custom build and flexible case design.

Life cycle analysis - Life Cycle Assessment (LCA): Will depend on the of material choice.

 

Air flow: When looking at air flow I consider physics, thermals hot air rises (convection) and cold air falls down relative to the surrounding air temperature. The point where cooler outside air enters the case should be in the bottom or lower part of the case and hot air should be expelled in the top or on one side of the case. Trying to ensure that heated air is not expelled in a place where it will rise and be dragged into the case again.

Are fans able to overcome convection (rising air)? Yes, but at the cost of extra power consumption, more noise and a general less effective system.

This should help us define the placement of hardware and Air flow in a low noise system and to minimize the power consumption and to archive peak performance.

Air Cooling: We are in the design of the enclosure looking for positive air pressure in the hardware chamber to avoid dust and lint being sucked into USB and other connection ports in the PC. This means checking installation of dust filters, air flow and pressure.

Cooling: For the CPU and GPU cooling are a water based solution is chosen instead of pure air cooling the reason being ‘the weight of a premium air cooling unit mounted on the CPU can weigh 4-5 pounds, that is enough to break the CPU or motherboard during transport.

Liquid Cooling: The liquid cooling system is a dual loop. Each loop has a pump, a radiator and access to a reservoir. In both loops are mounted sensors that are checking flow and temperature. In and outside the reservoir we have temperature, level and overflow sensors. Mechanical flow sensors are not used as indicators because it’s a locked case with limited mounting area in a visual spot and because I want to use the real-time data to optimize the cooling.

The reservoir should be as small as possible, as the reason for the reservoir in the loop is to handle the expansion of the heated water, next to giving us a high point in the system that ensures that the loop works optimal without air inside and as a point for filling liquid on the loops. A filter for collecting loose components comes after the reservoir and before the pump.

(Updated 2nd of Juli 2021)
Cavitation is a phenomenon in which the static pressure of the liquid reduces to below the liquid's vapor pressure, leading to the formation of small vapor-filled cavities in the liquid - (Think of it this way at sea level water boils at 100 degrees Celsius, at a lower pressure the water can boil at lower temperature think room temperature. The pressure behind an impeller can become really low).
When subjected to higher pressure, these cavities, called "bubbles" or "voids", collapse and can generate shock waves that may damage machinery. This is part of the reason why I chose 2 pumps to be able to have a lower circulation speed even in a "cooling emergency. It is important because it keeps the cavities, called "bubbles" or "voids" from forming in the loop. Link

The tubes in this water cooling system will be soft flexible tubes with compression fittings, due to transport consideration and flexibility. The dimensions are 10mm (3/8in) Inner diameter (ID) and 13mm (1/2in) outer diameter (OD).

To minimize the heat radiated back inside the case the hot tubes should be isolated. Only 2 hot tubes atm.

A manifold aka distro plate was not chosen at this point in time because of weight, and IMHO the lack of flexibility and cost (high) vs benefit (mainly superficial IMHO). If you can prove me wrong please do! Then again who doesn't like a little bling

 

Materials: Sliding material for the surface between compartments

Polished polyester (polyethylene) has a low friction level. To be tested.

Alternative adding Boron nitrite cubic to the epoxy. Boron nitrite cubic is the 2nd hardest material after diamond similar properties to graphite and it’s a very lubricating aka a slippery material that can be added to the epoxy, it’s like a poor man’s Teflon just tougher.

Felt: Has its use at specific points inside the case.

Additives: Titanium coated diamonds an additive for the epoxy twice as strong as the epoxy, can be bought from china. Same goes for Boron nitrite cubic

Still putting it together if any of you have input I would love to hear it!

 

Standalone system, for control and monitoring:

Having a Standalone system, for control and monitoring has been a 20+ year old dream of mine and why I started on this part of the project. Monitoring because, I want to see how the loop is performing internally after that Control to be able to optimize system performance.

I chose an Arduino because of a project on REDDIT I read about 4-5 years ago with an Arduino Uno.

I chose an Arduino DUE because it has an upgraded processor (CPU) faster by a factor of 10 compared to an Arduino, and it holds more connections for sensors.

Arduino DUE has 54 digital input/output pins (of which 12 can be used as PWM outputs), 12 analog inputs, 4 UARTs (hardware serial ports), a 84 MHz clock, an USB OTG capable connection, 2 DAC (digital to analog), 2 TWI, a power jack, an SPI header, a JTAG header, a reset button and an erase button. With PWM it can control fan and pump speeds in the enclosure. No system that is commercially available has to my knowledge had all of these options. https://store.arduino.cc/arduino-due check documentation

The performance of the water cooling loop and the hardware meaning CPU, GPU, disk, memory, system temperature, etc are monitored and controlled by 1-3 standalone systems based on Arduino. Standalone meaning not online, this is to secure daily operations, data and privacy.

The 1- 3 Arduino's are needed because of the modular setup that I have chosen. The Arduino's are connected through the I2C bus 1 master and 2 slaves.

When I got a working prototype I will post the diagram, photos and a video. I still have components in back order.

 

I decided for a Carbon Fiber Composite material (It could as well have been glass fiber) as the main material for the case. This is after looking at the weight of the case (transport being important to me) and the cost for the tools that are needed when working with aluminium vs carbon fiber composite to get the result I aim for.

There are item that are common for both and thus not listed.
Aluminium: Metal CNC, Metal lathe, Column drill, Sheet Metal Brake are the larger items.
Carbon fiber composite: Vacuum pump, Carbon sissor, Vacuum clamp, Catch pump, are the main tools.

Then comes the space needed for a workshop when looking at both materials and composite gets a 2nd win.

Composite material is more expensive when looking at the cost per kilogram, since this is a single piece the cost of tools and space have a higher impact on the total cost in the short term.

I would like to throw 1 single you tube channel name in here clickspring he makes some really impressive metal work with hand tools and I will implement more of his techniques on my next build or my upgrades of this build, to be decided.

I have made calculations for the amount of material that will be used for the design:

The dimensions (amount of sheets) gives a total surface of : 42 cm * 46 cm * 7,5 sheets = 14490 cm² = 1,4490 m²

And the material depth for almost all plates is around 3 mm.

Aluminium 1 m² (equal to 10000 cm²) of 1mm depth weighs 2.7 kg.

Carbon (Prepreg) 1 m² (equal to 10000 cm²) of 1mm depth weighs 1 kg.

An aluminium case will weigh 1,449 m² * 2,7kg * 3 mm = 11,749 kg
A carbon fiber composite enclosures weight in at 1,449 m² * 1,0 kg * 3 mm= 4,347 kg

*This amount of material and weight is calculated on the basis of solid sheets and does not include intelligent use of materials at stress points and removal of material from other areas.

 

Materiale Heat conductivity

Carbon fiber-epoxy composite 5-7

Steel 50

Aluminium 210

Looking at the heat conductivity it’s obvious why aluminium would have been a good choice it’s about 40 times more efficient at transferring heat than a Carbon fiber-epoxy composite exterior. If in the future if I get access to a CNC I might retrofit the enclosure with an aluminium shell, this is also to give me a chance to compare temperature data, and to see what effect, it has on a system in total.

Depending on the ability of the active cooling system (the fans) to get heat out of the case, the built in insulation of carbon fiber composite might be an advantage, to be tested.

Thermal expansion is as mentioned worth considering on a project if combining two or more materials to avoid skewing, cracks and fractures.

Compared thermal expansion:

Aluminium 13

Steel 7

Carbon fiber-epoxy composite 2

Choosing carbon fiber composite i also have to change the size of the gaps I planned for in between the different modules. Check the 4th post with Thermal Expansion and The Thermal Calculator for more details.

 

Carbon fiber (maybe Epoxy) Composite: This part is actually a bit hard for me.

Carbon fiber material was chosen because:
it weighs as little as possible (check).
Is as strong as steel (check actually stronger).
Is relatively easy to start working with (check).
Is non conductive (unchecked) but with a little tweaking and a thin layer of glass fiber this is fixed (check).
Can be recycled (unchecked?) What yea the disadvantage of carbon fiber-epoxy composite is that the material until 5 years ago wasn't able to be recycled, this changed with the use of the resin (vitrimers). Link The right choice of resins is very important for heat resistance, for areas with food processing, racing cars and more!
Carbon and glass fiber are today mainly down-cycled and it's not a good solution when we are looking at the recycling of materials.
So why did I chose it? Where Carbon Fiber gets ahead is on CO2 / Pollution because of lower weight and thus saved fuel/energy during transport. Not really a main selling point when looking back into Covid19 time, 17 months and counting.

And yes it looks good if done right, even though I like minimalism I like a little bling too.

What makes carbon fiber products expensive are the manual work hours needed to get the right result.

I looked into wood and here we can talk about expensive options - Walnut Burl Wood Veneer 46 £ for a 63 * 25 cm piece, but its beautiful .



I made a to do list when working with Carbon Fiber (CF) from:


The Form – Pattern making, etc......

Might be easier to link to some you tube videos that inspired me - is that allowed? until I get my own case production on the move. I might actually be the only one reading this lol.

NoX out

 

Design details:

I will in the design of the case incorporate a docking bay for a minimum of 2 drives. Hidden connections (read USB and card readers) in the front of the case.

The Standalone system, for control and monitoring is fitted with a Touch screen display (not sure how much value this adds because of the dust filters and lack of access). I might have to find an alternative for the dust filters.

For the air to have less resistance when moving through the dust filters I have redesigned the area for air filters to cover 4 x the area of the Fans. With 2 cm of space in between fans and filters.

I'm still working out a few details regarding functionality and add-ons.

To Insert into the carbon fiber Female Thread Knurled Brass Threaded Insert Embedded Nuts.
Brass standoffs: Female Male Thread Brass Hex Standoff.

According to the ATX 2.1 standard, the standoff provided has to be a minimum of 6.5mm and the external cross section has to fit within a 10mm x 10mm area around the standoff hole.

I prefer a larger standoff from motherboard tray to motherboard it allows for better airflow behind the motherboard, in my opinion. To check and maybe help by shaping the carbon in the motherboard tray to guide air behind the motherboard.
Might also be a good idea with an exit for the air coming out from behind the motherboard in the back board of the skeleton (needs to be guided up) .

Power and reset are also hidden but easily accessible. No idea what buttons to use yet. Any suggestions?

Phono jack connections and RJ45 (network LAN) I will fit on the front. The RJ45 is extended from the motherboard in the back.

 

You'll have a RJ45 on the front? for attaching a laptop or something? interesting idea. Would it just be a pass-through to a second LAN cable in the back going to the router?

 

Hi good question! RJ45 on the front.

Short version is a 2 part explanation.

I have worked with rack servers for many years, where a RJ45 connection on the front is standard on a lot of systems, I like the convenience, and it inspired me to want more especially when a laptop isn't enough.

An example:

That leads me to my personal/work rig where I have added network cards for Virtual machines it's a security thing, fast total reset of OS and software when connecting to a compromised system even when the motherboards has 2 native network connections available. The 2nd network connection gives some extra speed on data transfers.

Back to your question it is a pass-through from a secondary network connection on the PC. I usually use this to setup network routers, firewalls or NAS units (with it own IP range) and it makes it easy to change the physical cables. You could also use it for a laptop to transfer data or network sharing. Remember its a tool and the only limit is our imagination. I miss LAN parties!

I hope this answered your question else feel free to ask again. I miss stuff since English isn't my native, 2nd or 3rd language.

NoX

 

Details - Add-ons

Built-in antenna and a hardware on/off button. This might only make sense with a firewall(FW) controlling the on/off function. I tend to lean towards not installing a wireless card for BT or WiFi. Alternatively running the FW as a virtual machine might be an option. Also to be decided. I could plan for an installation point just to be sure - flexibility.
The antenna could be a custom part . I will look unto it!

Antennas and SDR (Software Defined Radio) Why it's interesting because of garage openers, ships or plane tracking, satellite hacking and of cause digital TV, options are infinite. watch TV or do something stupid and you can go to prison as an adult. In my world its all about options so I will plan to incorporate 1 in my design. Hak5 has some interesting videos and for us working with Information Technology(IT) and thus also IT safety its sweet to be able to test for flaws and remove them if any are found.

A you tube channel called DIY perks has made a Concrete Volume Knob a project with some interesting upgrade option. I could imagine USB ports as an option for mouse, trackball, 3D tools, SD card and USB memory or a docking port for a mobile phone. An installed volume knob on the Case? To be decided.

NFC reader/writer will also be added internally.

I am venting the geek part of my brain ATM. I am trying to make the scratch build a functionalist expression of my needs or as a minimum a space with the right possibilities for the future.

To be continued.

NoX

 

I like the thorough planning of yours.

But when CPU and GPUs are watercooled and the "heat" from the radiators is properly vented out of the case...what source should heat up the case that its heat conductivity would matter at any point, or thermal expansion become an issue?
How hot do you expect the case parts to get?

All the cases in my build history were some kind of hybrids with a material mix out of steel, aluminium, various plastic materials, glas, etc. All with different expansion coefficients, all bolted, riveted or glued together.
None of them ever bend under heat, no matter how hard I pushed the hardware.

If you airflow is right, your case will not go much above ambient.
I have rarely seen loop temps above 40 dec Celsius. So this is the temp your rads will have and consequently the temp your ourgoing airflow will never exeed. Ever.
A 15 Kelvin temp difference to ambient will cause material expansion which you can simply neglect.


You want to minimize bolts and nuts. So we are talking snap fits?
Please explain a little bit more about the design concept.
Any sketches or pictures of the paper mock-ups?


Btw. carbon fiber is abrasive.
So your tools become dull rather quickly. Do not use regular drills if you do not want the 5th hole to fray.

And ALWAYS wear gloves AND MORE IMPORTANTELY a MASK.
Otherwise you have developed a nice little cancer by the time you consider recycling your case.
Carbon fiber is lightweight, strong, good looking and carcinogenic - not the finished product, but the nasty little fibers that pollute your shop air when you machine/drill/dremel the stuff....

 

Hi Dan297

Thanks for your input.

Part of the reason I looked at Thermal expansion is that I have planned a dual chamber skeleton, that is sliding into the shell (think motherboard tray). To make it fit as tight as possible the gap needs to be small thus Thermal expansion. There are more modules. I will publish my plans when I'm satisfied with the total concept I will document with hi resolution pictures and video. I thought it might be fun having a concept from idea to final product with the possibility of adding input for the final result.

Part 2 of why I focused on Thermal expansion is me moving around with this rig. We have temperatures in winter to summer from -20 degree to 35 degree Celsius equal to -4 to 95 degrees Fahrenheit, in Scandinavia. If it was made of aluminium the expansion could be 0.3 mm per module with the dimensions i am looking at. Expansion wouldn't matter as much if it wasn't modular.

Looking at heat in the case, my main concern is the temperature of the motherboard, NVME SSD and Memory since the rest is water cooled. The motherboard tray would be the part of the case experiencing the biggest temperature difference because of the mentioned components IMHO. Then again the expansion of carbon fiber is next to nothing and part of the reason i choose it.

The shell is sliding onto the skeleton this gives me the option of only using 1 button that is mounted on the skeleton and sliding out of a hole (in the top of skeleton) keeping the skeleton and shell locked together (I choose to use 2 buttons, safety).

Thanks for the tip I'll order extra drills, I'll use sandpaper to fix the edges, and saws to cut it to size. I will order carbon and glass fiber scissors to cut the fabric. I have to test a lot of material 1st. final thickness is important. Last but not least I will have to add a final layer of UV resin.

I agree safety 1st when it comes to working with these fibers, this is why I don't expect to be done within the next 11 - 13 months. Nobody should work with this inside a living area.

 

A mock-up of the on Screen data of the standalone system - System: Arduino Due and touchscreen (How I imagine it)

(Updated 16. may 2021)


It also shows the components in the H2O loops.

* PWM is Pulse Width Modulation and can be used to control speed of pumps, fans or color of RGB and more.

Water Cooling side (dual chamber):

RGB Diode (PWM) FFF ( I am spineless - RGB in the reservoir)

External temperature sensor 21 ℃ (outside the case air temp - taken where the air enters)

Fill port fitting - Pressure Valve * (keeping the pressure in loop low)

Reservoir - temperature 25 ℃ (Water temperature in reservoir)

In line filter (ILF) *


Loop 1:

Pump 1 (PWM) 800 rpm

ILFS 1 (In Line Flow Sensor) 12 lpm

CPU 42 ℃

ILTS 1.1 (In Line Temp Sensor) 29 ℃

Radiator 1 *

Fan (PWM) 1200 rpm

ILTS 1.2 24 ℃


Loop 2:

Pump 2 (PWM) 800

INFS 2 12 lpm

GPU 52℃

ILTS 2.1 32 ℃

Radiator 2 *

Fan (PWM) 1400 rpm

ILTS 2.2 26 ℃

ECS (Emergency Cooling System) (on/off manual) Work in progress add on part

ILTS 2.3 26 ℃


Drain port *

Hardware side:

Internal case air temperature XX C

Motherboard temperature XX C

Fan (PWM) x 2 In ( Air being dragged inside the case hardware side)

Air pressure? Difference ! It might be possible to use an Air pressure sensor to see if there is a positive pressure.


Sound level DB? Difference ! a Decibel(DB) sensor might give some interesting data on performance vs DB. Sound level matters.


Fan (PWM) Out (outgoing air)

Liquid/Moisture sensor: OK / Shut down (If it detects liquid).


//Note to self: Quick disconnect fittings CPU & GPU is likely a must as a minimum on CPU. Reason is the problems with CPU mounting accessibility.


I will add a picture of the drawings tomorrow.

 

Cool. I am looking forward to these pics.
Sounds like a similar concept than the Thorzone Mjolnir, just on a bigger ATX scale...

Wouldn't all modules expand equally? I mean it is not like your shell has -20 deg and your core has +35 deg...Or are you planning to use the thing outside?

Your total mobo to tray contact area is just about 171mm2. That's 13x13mm or the size of a fingernail. That's if you use all 9 standoffs.
These 9 heat influence zones are evenly spread over the tray.
When your airflow is right, thermal convection from the mobo to the tray over the air will not happen, especially since you even consider 10mm standoffs vs. the standard 6.5mm ones.
I still think no worries there.

I would be more concerned how to create an even (and controlled) thickness of those prepegs.
Do you plan to make the shell as negative build in a mold or as a positive build around a core?
Single piece shell or multi-piece design?

Using a mold has the beauty of a spotless surface finish. You can brush or spray in the UV resin first and than start your prepeg layers.
It is more work initially, but gives you repeatable results if you plan to do more than one case.
But the inside is kinda rough from the prepeg structure and the vacuum foil. Depending on the shape this might be difficult to sand later on and sanding will cause an uneven thickness anyway.
Also the prepegs will have thickness variation over the surface area. Your vacuum bag will have a fold here and there and that will already ruin the eveness.
And I believe the inside dimensions/surface evenness is key to your sliding design.

If you build positive around a core, your design determins the core material.
If it is a one piece design you need a lost core, so it got to be foam or light balsa, or something similar that you can easily destroy and pick out of the cured shell.
This core needs a glas fibre/epoxy surface and a spray painting finish for your release agent.
Sanding this kind of material to an absolute even surface within the tolerances you are concerned about will be at least a challenge...

Which build method have you planned for your skeleton and shell?


Concerning the inserts for the mobo standoffs look at Tappex Multisert vom KVT-Fastenings.
The M3 size comes in the right thickness and for this application you can just super glue them into a slightly smaller hole.

https://www.kvt-fastening.de/eshop/...d-inserts-for-plastic-materials/c/02.300.200/

 

Carbon: Abrasive wet saws and diamond drills are your best option. -All with water. With something this nasty, I'd suggest a dremel flex shaft and diamond cutters. You can isolate the dremel half while having water pouring on the bit.
Keeping it under constant water makes it hard to see, but it stops the deadly dust. Where the dangerous waste goes is more manageable too. You can have the water channeled to a hole in the ground or a sand filter, for example.
-Also, holy heck! I thought -I- was text-heavy.

 

Hi Cheapskate

And yes I am text very heavy I have filled 2 A4 checkered blocks with ideas of PC case details and test drawings then iterations, all of this during the last 2-3 years. Then I sat down and transformed them all into text. Before starting on this from scratch. I use it as a 4D puzzle.

Thanks for the advice my next problem is that I will have to cut the parts when they are 3D shapes, not sheets any advice?

Hi dan297

I do agree with you on the heat on the motherboard tray it should be solved with the increased distance MB to MB tray and air flow. Funny option I looked into Thermochromatic Pigment reversible - pigment that reacts within a certain temperature range, and can show hot spots. Link

Regarding core material I have looked into:
Nomex aramid honeycomb Thickness 1.5 mm Cell size 3.2 mm Nomex provides galvanic corrosion resistance.It might come in handy when forming the spine aka the Motherboard tray.

I plan to do it like a puzzle, like a maze from the inside and out. I am also concerned about how to get this even and with a controlled thickness.I am still looking for alternatives so please, keep the ideas coming.

I might have cracked this but with some heavy risk taking, and since I'm new to this I am willing to, in Danish we say pay to learn (make sh*t happen and then pay when it doesn't work)

The plan is to make a solid form shaped like the skeletons (external surface) that is longer then the final product to create some work surface, where material can be glued/taped.

Then use an offset wax method in my case 1+3 mm wax on the form (The gap 1 mm and the shell 3mm thick) I will create a 2 or 3 part mold for the Shell The thing is to get the initial form perfect since its the basis for both the shell and skeleton.

I will face some challenges when placing the fibers that's why I made the form longer then the shell will end up being.

I'll infuse first, if/when i f#ck this up I will use prepreg , its what the pros do! I am in the progress of making a longer description! Yea you know me by now

After this I will make a 2nd mold for the skeleton, I think, I am still working on ideas to save on material and to get a tough enough skeleton for some of the other details I haven't mentioned yet.

i plan on using some of the techniques from the videos in the following links:
-Using Sheet Wax to Create an Offset Mould for a Carbon Fibre Cover Component
- Composite (Fibreglass) Pattern Making Tutorial a 3 part worth watching.

- Stabilized Carbon Fiber Comparison (Pro Finish vs Web-Lock vs Spray Adhesive)
Last but no least and I would have cried like a little girl if i had missed this! It would have messed up the quality of the finish. What difference looking at it from a tailor's perspective can have when working with carbon fabric cloth.
More to be added later.