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

Adapting the skeleton approach to the Spine Core, has been the plan for some time. I got curious how well it might do with the wiring so I started to implement it ahead of some other improvements.



I spent more time on this then planned, then again it points to other parts that can/should be improved. This is the 1st iteration.





I like the look of this part but with Aluminium the weight will be close to 3 kg, in Carbon Fiber the weight should be below 1.7 kg. To create this part in Aluminium I need a CNC or access to the same, this is not in the cards atm.

The problem with heat dissipation inside the Spine if made of carbon will have to be addressed later. I am looking into a few few different options.

This approach with leading the wires inside the Spine Core gives a lot less clutter and air resistance inside the case.

Rear


Front


These are all the main motherboard Connectors attached to cables (almost) not including Sata, Fans and RGB (if its on the board).



At present the cables to the SAS4MC have been planned for but are not included on the CAD drawings yet. I will add them when time allows.

Added panel and improved on the next pictures:





An overview and a look of the Wires, LAN - RJ45 (the large RED cables) moving data from the Rear to the Front when needed.



Thats it for now

Nox Out

 

Hi all

Latest progress update.

Still rapid prototyping ideas for inserts, shape, forms and placement of components to be sure it all fits and is functional. I went back and took a look at the Clips that are used for bringing the Power Wires through the Spine Core (read Motherboard Tray). They will be made of aluminium to make sure heat is not building up to an unwanted level inside the composite spine core, with a deeper core it seems to be necessary. As a design element the aluminium could be anodized into a color that works for this project.

A few pictures before more rambling ;D










Some parts still need to be added to the Clips, will follow later.

Placement of the Clips are next to the areas on the motherboard where needed.






The Clips added into the Spine;






Inserted Clips can be seen next to the cut out for the Core Cover with raised area for wires comings out of the spine core.






The core cover is added here in yellow.


This is still far from done, the wires need to be closer to the core cover (read closer to the surface). Why? To ease access (serviceability) of wires, and to dissipate heat created in the wires. The cover could be made of paper, composite (the most likely candidate) or a thin sheet of Aluminium.

A design element not added yet is using strain reliefs and cable organizers to prevent sharp bends, loads on the wires and add ease of access, will be added soon.

The wires are now laid out to avoid sharp bends because the rules that are true for resistance in a water tubes also work in a data or electrical wire. Lets take a USB Cable sharp bends heighten the likelihood of data loss. How? Imagine a race track and a sharp corner data running wild and not lowering the speed and some of it will not make the corner.

USB cables are designed to be flexible, but excessive bending or frequent bending at sharp angles can lead to physical damage that may result in data loss or intermittent connections.

Why am i planning this to this detail level?

It started with a lack of knowledge on small electronics, wire resistance, power drop etc. Now it has transformed to looking at the limitations on wire length versus resistance, wire diameter, voltage drops etc BUT most importantly finding a way around these limitations!

Example: A normal recommended limitation when working with the Arduino 3.3V and the distance of wires to sensors with data and power cables is 30 cm (read 1 foot) resistance and voltage drop seems to be a big part of it but so is Signal Interference because of a lack of shielding depending on the purpose of the wire. Wire diameter vs the resistance is also a part of the equation. I might in the near future go all out detail geek on how these parts are working TBD.

A big part of planning this Spine is imagining how to build it and let me be honest this is by a large margin the most complex part to date. Can't help being a bit exited about the process.

Well that's it for now

Nox Out

 

Hi all

The Process of building the case as I see it at the moment.

My initial idea and long term solution is 3D printing the entire case.

This does present some difficulties:

1. The case is 450 x 450 x 650 mm try finding a 3D printer within a budget that can handle this size short term. This is not even including the cost for 3D print failures, in both time and materials.

2. Outsourcing would be an option looking at the price, the cost for outsourcing is close to https://creality3dofficial.com/products/creality-cr-10-s5-3d-printer

This means we hit the thinking mans option, work the problem. Get maximum effect of the resources available. So a smaller printer, gluing the 3D prints together and optimize?!

For the Shell this is surprisingly simple. instead of 3D printing the sides of the shell I use glass/mirrors cut to the size of the sides. This cuts down the 3D prints to the corners of the Shell with some support for the glass. I will only be using this 1st edition Shell to make a form for the carbon fiber Shell.

The advantage of this process is the flat surface and cost of glass that's is actually cheaper than if it was 3D printed and the time saved is massive.

So printing 4 corners in 2 parts each with support for the glass on the inside, check.

Creating a form likely of fiberglass on the outside in 2 parts splitting it on the corners. There is a lot of grinding and polishing in between prepping the form for the Carbon-fiber fabric, lets be honest that not all.

Thinking about it all, RESET: Creating a form with the glass and corners for the carbon-fiber is the premium choice, discarding the glass fiber form! So1 step back and less work for me and the option of looking through the glass and seeing if the carbon-fiber is mounted where it is supposed to be. Might have to create a Top flange to mount the Carbon-fiber on.





10 minutes exercising the grey matter, money and time saved. Problem is I still, that I would like a large ass 3D printer, well now I might have time to build one TBD.

The main Challenge is mounting the carbon fiber in a 3 mm deep layer, not using prepreg. I did some time ago choose to use a wet layup/vacuum process because prepreg has a limited lifespan in a freezer, and needs a autoclave (oven) to harden, The oven part is not a deal breaker it can be build relatively cheap using IR lamps and insulation in a metal box.

The same process is initially used for the Frame, it will need some more preparation and 3D printed inserts (for the Cut outs) before mounting the Carbon fiber in the form. I will at a later date go into detail on this part.



The pièce de résistance is at present the Spine/core (The motherboard Tray) it has some complexity that will likely go up before I find the design fitting my wish for functionality.







I would go into detail about the wires and the wire mounts but I am out of time.

Added a day late

Glass for the bottom a tat longer to be able to add a MNT for the Fiber fabric:









Fabric inserted from the Top







Nox Out for now

 

Hi all

A short note not intended to blame or point fingers, just a FYI.

Some days ago i mentioned that I really wanted a large printer on 1 of my posts in here and today YouTube pushed 2 Large printer projects on my account, If you are interested it puts it all into perspective, in my opinion that's a healthy point to start from. I still want a large 4ss 3D printer.

The links are:

 

Hi all

As a part of the normal process I am looking for parts of the project to improve. So looking at my last additions being critical and constructive is as always important.
The next picture of the wires it's obvious that there is no support for the wiring in some spots. Especially at the rounded corners.






So 5 unsupported corners om this part.

In the picture above even the support is wrong when checking a wire route. The support angle is to steep for the it to look healthy structural. This will be corrected with changes and added support in the Spine Core. This can also be seen below (check the yellow line) Next to it there are very LoOONG holes for the wires. Not to mention the problem trying to recreate holes of this type.




The next part is looking at elements that can/will be changed like the motherboard. Yes I found a mistake made by me, damn. Well better now then later.



Changing the size of motherboard. Next a E-ATX size:



Next up Server board dimensions.





When looking at the picture above the wire access for the USB C and USB 3 are disappearing behind the motherboard when its a server sized sample. Will be fixed, To be honest its very unlikely that a server board has this type of connections in the same area.



So enough to do atm.

The most of the wires above are permanent wires that is the reason they are embedded into the Spine Core. I did consider hiding the wires. I might be complicated but I can appreciate structured wires that can be seen and accessed when it's needed. Function first then esthetics,

See you all later.

NoX out

 

Hallo again.

An Update on the progress. Sensors and Wiring a view from the front of the case, the components that are connected to the 1st Tech Chamber (on the low right front):





Arduino® Due - Processing and output.

RTC Clock Module DS3231 AT24C32 I2C - keeping time.
IIC I2C SPI MPU-6500 6-Axis Gyroscope Accelerometer Sensor Module GY-6500BDS*- impact and theft system, it needs a battery pack to be online 24/7.
BME280 High Precision Temperature Humidity Pressure Sensor - One in each enclosed volume.
TCA9548 TCA9548A 1-to-8 I2C 8 -way multi-channel expansion board IIC - 3 to 4 to connect all the sensors, iy might be 5 MUX .
Waveshare 128×128 General 1.5inch OLED display Module - Display x 12 on the curved front
MQ-7 Carbon Monoxide Sensor Module CO - Nice to know when your productivity is going down
Micro Switch with lever - Status of the modules and are they in place?
Output Current Power Monitor Module INA322 - Power use in specific areas (more to come).
MAX30102 Pulse Oximeter Heart Rate Sensor Module - Status of me/you.
APDS-9960 Sensor Spacing Gestures Color Brightness I2Ca - Touch less control of data on screens and more.
TOF050C - Is the air intake clear? Distance to objects.
Luftfeuchtigkeit Regentropfen Erkennung Sensor Modul - Rain Sensor for Arduino- Water on the bottom of the case alarm!
AMS1117 Multi DC-DC 12V-3,3V/5V/12V Step-Down - Power supply.

Why the CO sensor: Impact of Carbon Monoxide (CO) on the Human Body Carbon monoxide (CO) is a colorless, odorless, and tasteless gas that is highly toxic to humans. Sitting in an office or a class room this is important. We produce it when using tobacco, fires to heat the living area.

CO Concentration (ppm) Symptoms & Effects
0-9 ppm No harmful effects for healthy individuals.
10-50 ppm Mild headache, dizziness, and fatigue after prolonged exposure.
50-200 ppm Headache, nausea, dizziness, and confusion within a few hours.
200-400 ppm Severe headache, disorientation, and weakness after 1-2 hours.
400-800 ppm Life-threatening effects, unconsciousness, brain damage within 2-3 hours.
800+ ppm Death within minutes to an hour, depending on exposure.

Why the MAX30102 Pulse Oximeter Heart Rate Sensor Module: Age and fitness data not on the phone (Privacy matters)

Next Picture has the Panel open on the Tech Chamber, I will add a better read more informative picture on this later.



---

The planning of the wiring is important for when the system needs to be taken apart for maintenance or upgrades. So the Design is being improved to make this easier. Modular and accessible mounting of components.

A Wire Relief MNT is implemented between the Air Intake Tunnels





This support is for the wires to the secret sensor. No I will not tell its my ace in the hole.



The Pulse Oximeter Heart Rate Sensor Module Far left top "floating" on the next picture




The next picture has the QUAD its to the left of the Reservoir. The Quad is a sensor platform/MNT that holds (Top center of the front):

BME280 High Precision Sensor - to be able to collect Outside (the case) data on Temperature, Humidity, Pressure and height above sea level.
APDS-9960 Sensor Spacing Gestures Color Brightness - Touch less control (Gestures), Brightness to control light levels in and on the case, color hmm.
TOF050C x 4 to make sure the Air intake is not blocked (Laser measured distance to any object)




A little Purple Plate is visible top left on the Spine its used for a BME280 High Precision Sensor and has a twin on the other side of the Spine.



The next slide has the Base for the Triple Button Lock(3BL) Top center and blue:



The Back Panel for the CPU MNT area (in RED) has been added with some improvements. I would like to have it made in Aluminium (or as our US friends say Aluminum.



The locking mechanism is good (Tool less) but I got an idea or two to improve it



Some springs will hold it in place, not yet added to the drawings.



The bolts have rubber spacers to keep the sound level low or more precise to avoid clicking during transport.Some levers/handles will be added to the bolts.



I decided to cut the hard drive docking bay since NVME M.2 drives are now the same price per TB as the old SSD's, with a significant speed difference. There will still be an internal MNT for 2 x 3½ or 2½ drives.

Bonus the area meant for the Docking Bay can now be used to cover up the Wires in the front that are bending around the rear of the front panel.





At this moment the majority (95%) of the wire routing has (likely) been planned for, I still need to work through the power wires, dimensions and resistance.
The code used for the sensors 1-1 with the Arduino Due has been tested on a virtual setup and works. An added layer of complexity when combing them all is headed my way, I am sure it will be fun.

I still got some work with the wiring and code on the Water Cooling Rig (WCR) why, well its a combination of analog and digital sensors.Over all the timing of the sensor read outs is a B14tch.
A sensor like the MQ-7 Carbon Monoxide Sensor Module depending on the time the sensor has been inactive , the sensor needs to be heated to age for it to deliver correct data, We are talking days. There is more but if you would like to create a environmental sensor suite its a good choice and not as complex as it sounds.

I'm out of time for now more has been prepared and will follow ASAP.

Nox Out

 

Hi everyone, it's been a while.

With everything happening around us — both in the US and in Ukraine — it's been challenging to stay focused on this project. I have made some progress, but there are also areas where I see room for improvement. That brings me to what I’m currently exploring:

The main focus is making the data easier to check on the fly, while keeping the setup standalone. The data I’m referring to includes system and cooling stats, which are currently visible through a filter mesh/fabric. I really like the Multi-OLED setup, but for a few years now, I’ve been considering making the information accessible via a tablet instead. Recently, I started working on integrating my current solution with a tablet mounted at the front.

This is still a work in progress, both in terms of design and functionality, but I’m actively exploring the options. I’m leaning toward this approach over built-in OLED displays because it’s easier to update and can also serve a dual purpose.

I am still testing dimensions and will adapt the present design if I can make it work.