The following article shows the second version of the Astrohub that I presented about one year ago. The previous version, which article can be found here, has been a completely success for me, in terms of setup time and reliability, and after a 1 year testing, I felt the need to go a step further so I planned to include some features that wasn’t included in this first version, to achieve a full automated setup.
The index above has some direct links to the most important aspects of this development, in order to allow a better navigation through this long article.
- THE ASTROHUB AT A GLANCE
- DEVELOPMENT OF AN ASTROHUB
- CABLE HUB
- HOW TO DETERMINE CABLE LENGHTS
- CABLE WRAPPING
- CLAMPING POINTS
- CABLE HUB
- CONSTRUCTION AND ASSEMBLY
- TOOLS AND MATERIALS
- POWER HUB
- DATA HUB
- CABLE HUB
- PURCHASE LIST
THE ASTROHUB AT A GLANCE
The Astrohub is a suitcase that contains all the connections involved in an full automated astrophotography equipment. It manages a monochrome camera and its filterwheel, the focuser, rotator, and dew heaters of the telescope, a guide camera, and a mount. All the cables that are necessary to power, and allow the communication between the equipment and the computer, are also stored inside. The cables are grouped and wrapped togheter, to allow a quick and reliable setup.
The internal space of the suitcase is divided into separated spaces including:
- A Power HUB to power the equipment:
The Power Hub has 3 DC independent circuits:
- L1 – PC Circuit: With a power supply that gives 100W (12-24V/4A) for a laptop. It can be powered with 220V AC or 12V DC.
- L2-DATA Circuit: That gives up to 120W (12V/10A) for the electronic devices that transer data (in my case, 2 cameras and a mount).
- L3 – MOTORS Circuit: Powered with a 120W (12V/10A) power supply for motors and other noisy electronics. I chose to separate data lines from motors lines to avoid the introduction of noise in the images.
It includes also a Control Panel to manage the circuits, and the devices connected to them.
It has a computer fan with a controller that allows it to automatically cool the electronics.
It can be powered with the included AC plug at 220V AC, or connecting 12V batteries to the DC IN connectors.
- A Data HUB to control the equipment:
It connects the equipment with the computer through a powered StarTech USB 3.0 Hub with 7 data ports. The hub is rated as 12V/3A Max, and has led indicators to check the status of the connections. The device has been oriented upwards for better accesibility.
- A Cable HUB
Consists in only 3 “master cables” to connect all the setup, one for the astronomical equipment, one for the laptop, and one for plugging the AstroHub to AC.
The Equipment Master Cable includes all the interfaces to connect to the equipment. This master cable has been marked to know were to be clamped to the mount quickly and safely. Data cables are shielded to protect data from electronic interferences. It is 3m long.
The PC Master Cable has a USB 3.0 interface coming from the USB Hub and a power connector to power a laptop without the need of an additional power supply. It is 2m long.
The AC Plug is an standard one, 1.5m long.
- Some Accesories are organized in the lid.
There is an organized attached to the lid that I use to store Bathinov masks, caps, and other accesories…
…and a DIY Flat Panel made with a iluminated panel for telescopes up to 8″.
HOW TO USE THE ASTROHUB
To use this Astrohub is quite simple:
Put the Astrohub on the ground, below the tripod.
Get the AC Cable, plug it.
Get the Equipment Master Cable
The Equipment Master Cable has 3 clamp points
Use the RA Clamp to attach it to the polascope cover holder.
Use the DEC Clamp to attach it to the dovetail knobs
The loop created between the RA Clamp and the DEC Clamp has the correct distance to allow the free movement of the mount around the whole sky without snags.
Attach the Cam Clamp to the imaging camera and connect the equipment
Connect the mount
Power the equipment
Get the PC Cable and connect it to the Laptop with the help of Yoda.
And it’s done, the equipment is prepared to be controled by the corresponding software and start imaging.
If you are interested in how this device has been developed, you can find more information below.
DEVELOPMENT OF AN ASTROHUB
This section explains all the planification, designs and calculations needed to develop an AstroHub 2.0. I suggest to first read all the article, as this is a development on which many of the calculations and assumptions are interconnected, while the write up is linear.
Almost every imaging rig is different, so the solutions and results presented her are only valid to the specific equipment described in this article. Nevertheless, it is easy to redo this calculations to be adapted to other similar imaging rigs.
This is a DIY project that involves manipulation of dangerous and/or expensive equipment. If you decide to make it, it has to be at your own risk.
The AstroHub is an all-in-one suitcase, designed to allow a quick, easy and reliable setup of the equipment involved in a full automated astrophotography session.
In addition, the Astrohub has to be suitable to rest otudoors in case of multiple days sessions, and transportable to be able to imaging from different locations.
And as always, the cheapest the better!
This is the layout of the complete system.
The equipment consists in a minimal full-automated imaging rig, composed by a monochrome camera with filterwheel, autofocus, autorotator, guide camera, mount, dew heaters and a flat panel. A computer is needed also to control the devices.
All this equipment has to be powered, and controlled connecting the devices to power supplies, controllers, hubs… All this connections have to be wired with data cables, and power cables.
This imaging rig needs 5 data cables and 4 power cables, so a total of 9 cables that have to go from the AstroHub to the equipment without snagging, or disconnecting.
The devices of my equipment that must be powered/controlled are:
- Imaging Camera: QHYCCD 163M
- Filterwheel: QHYCCD CFW2M-US
- Guide Camera: QHYCCD 178M
- Mount: SKYWATCHER HEQ5
- Focuser/Rotator: SELETEK ARMADILLO 2
- USB Hub: STARTECH ST7300U3M
- Laptop: ASUS ZENBOOK UX-32VD
- Dew Heaters: PWM CONTROLLER (Generic)
- Flat Panel: DYI FLAT PANEL (Eventually)
The following table summarizes the connections and power needs of this devices:
Some notes about this table:
- The requirements have been divided into Power, Data and Cables needs.
- The lenght of the cables should be calculated prior to the start of the development. You can use the methodoloy proposed on the Cable Hub section to do it.
- The power circuit will be splitted into 3 subcircuits:
LINE 1 PC: My computer operates at 19V, so cannot be powered with an standard 12V circuit. In addition, the laptops/computers tend to be the most power demanding devices, and in case of powering the equipment by batteries, can be usefull to power the laptop with an independent battery.
LINE 2 DATA: I chose to isolate the components that transfer the images data (Imaging camera and Usb Hub) from the circuit that powers other components that can introduce noise on the circuit, so this is a “clean” circuit.
LINE 3 MOTORS: This circuit includes all other equipment that isn’t involved in image transference.
- The filterwheel is directly connected to the imaging camera, so it can be excluded from this analisys.
- The power needs in this table are the rated (claimed by the manufacturer specifications) needs, so they are not the real needs. To determine this real needs, one has to measure every component load while performing a standard astrophotography session. It is important also to take into account a high load situation.
- It’s very important to take into account the voltage drops that can happen when running long power cables and/or when power supplies work under high loads. For example, I use an old Skywatcher HEQ5 mount, that is very sensitive to voltage drops, and doesn’t work properly if the voltage goes below 11.6V, so I decided to use a buck-boost device in order to keep the voltage stable at 13V.
ASTROHUB MAIN LAYOUT
The main layout keeps the same distribution of the previous Astrohub as I found it to be very simple and effective. The main container of the suitcase is divided into 3 spaces:
- POWER HUB
- DATA HUB
- CABLE HUB
The accesories are now planned to be placed on the lid. In the previous version I realized that to have a bag on top of the cable hub makes the operation of putting off the cables a little more difficult.
That means that the lid with include:
-Caps and filters
(This development is Work in Progress. The configuration of the lid is the same as the previous version)
The cable hub is a space in the suitcase to simply store the cables of the system. Seems simple but it is not. If the cables are left in space without any order, they will become entangled and at the time of using them we will need a lot of time to untangle them. My proposal for the cable management is to group the cables positioned as «ready to use» and attach them to the mount in specific points. This points will be calculated to avoid the cables to be hooked.
HOW TO DETERMINE CABLE LENGHTS
The cables must have the appropiated lenght to reach the devices and to allow the mount to move the imaging rig safely. To calculate this lenght accurately, I propose the followind metodologhy:
Note: If you are attempting to determine wich cables you should buy, you can proceed with old cables, or even with cord or rope of a similar diameter. Simply take into account the extra lenght for the USB connectors dimensions.
- Connect all the cables to the equipment as for a usual astrophotography session.
- Group the optical train cables (we will name it as DEC Cables) with removable fasteners (like velcro straps or small pieces of wiring spiral) and choose a DEC Clamp. The general rule to determine a Clamp point is to search for a point as near to the geometric axis as possible. If the cable passes through the axis it will only rotate, and his position relative to the RA Clamp will remain constant. This will allow to shorten the cable lenght. Use a removable fastener for fine tunning the lenght.
- Rotate the camera to +90º and -90º. In my opinion more than that has no sense , because in that case I will choose the oposite position, but for other setups, maybe other angles should be considered.
- Shorten the cables as much as possible. In my case, I have to add some extra lenght for my other optical train, that is slightly longer.
- Make sure that the devices can be desconnected if necessary.
- Join the DEC Cables with tape in the DEC Clamp position. This will keep the relative position of each connector and will allow to remember where the cables must be attached to the DEC Axis when the cables are unclampled.
Now the DEC Cables lenght and the DEC Clamp can be considered fixed. We can continue with the RA Cables and RA Clamp.
- Connect the Mount Cables.
- Rotate the DEC axis 180º from Home Position.
- Rotate the RA Axis to +120º and -120º (or more) form the home position and clamp the cables to the RA Clamp in the position where the lenght is maximum.
- Double check that all the cables have the same distance between the clamp points and group the DEC Cables and the Mount Cables with tape in the RA Clamp position.
- Return the mount to Home Position and check that the Mount Cable doesn’t hit the DEC Cable.
- Let the cables to hang to the floor and add the necessary lenght to reach the USB Hub in the expected position of the Astrohub Case. Put a mark on this point.
- Now you can unclamp the cables from the equipment and measure them.
After doing this procedure, this is the result for my setup:
Note that the distance between the RA Clamp and the floor is between 1m and 1.2m. That’s because I have to manage two observatories, one with a tripod and one with a pier and they have different heights.
As a conclusion, I will need 3m cables for the imaging rig and 2m cables for the mount.
In the previous Astrohub I used spiral wrap to group the cables. They worked quite well but it was very slow to unwrap in case of making changes. I searched easier ways to wrap the cables and found this:
It has similar characteristics as the spiral wrap and includes a tool that makes the wrapping process extremely quick.
The general rule to determine a Clamp point is to search for a point as near to the geometric axis as possible. If the cable passes through the axis it will only rotate. This will allow to shorten the cable lenght.
I chose 2 clamping points, one for each mount’s axis:
For the first point I chose the polarscope cover. This point passes through the RA geometric axis. In addition, (in the HEQ5) this cover rotates in solidarity with the RA axis, so if the cables are fixed, they will keep the alignment with the connectors of the mount.
I designed a fast-introduction/fast-release clamp that fixes also the orientation of the cables.
The cable holder is attached to the cables and it has a dovetail to be introduced in its housing. The release ensures the cable holder to not to fall when the RA axis rotates more than 90º.
This clamp design is still in development, so at the moment I continue using an elastic cord attached to the saddle plate knobs.
The data hub is the responsible to manage the data flow between the equipment and the computer. It is located on the top side of the suitcase, next to the power hub and the cable hub.
The schematic of the data hub is quite simple. The idea is to use a USB Hub between the PC and the equipment, going one cable from the USB Hub to the PC, and one cable for each device from the USB Hub to the equipment.
Both cameras use USB 3.0 and the mount and the Seletek use USB 2.0 connections. The connection between the USB Hub and the PC will be USB 3.0 also.
As explained, the Seletek will be placed inside the Power Hub, so there are an additional cable that must go from the seletek to the motors that control the focuser and rotator (represented by dashed lines)
limitations in data cables
The lenght of the data cables is a discussion that is not always clear. The general rule is, the shorter, the better (of course!), but in my opinion it makes not sense to take this as mandatory. For passive USB cables, it is more or less accepted the limitation of 3m for USB 3.0 and 5m for USB 2.0, but as far as I know, the specifications of the USB 3.0 interface doesn’t states a determined maximum lenght.
Anyway, there are chances to break this limitations, as the active repeaters and powered USB Hubs, but not all the extenders or hubs work as expected. I had bad experiences with active repeaters, so I decided to go to the powered USB Hubs option.
The theory says that with a USB Hub, one can extend the cables to the «acceptable» limits in both ways. This is true in this case for the cables that go to the equipment (3m), but for the cable from the USB Hub to the PC I would prefer to go to the shortest possible, but comfortable to operate. That’s because each equipment data cable transfers its own data, but the cable from the hub to the PC has to manage all the combined data at the same time, so is a cable with much more stress.
I didn’t really planned any change for this new version of the AstroHub, apart of the obvious (the old guide camera was USB 2.0 and the new one is 3.0), but I had communication problems with the QHY178M and in the debug process, I adquired new cables and hub. It is not really bad (apart than for the budget) but I wanted to clarify that the components selected in the AstroHub 1.0 worked flawesly.
I chose a 7-port StarTech ST7300U3M powered USB Hub. StarTech is a well reputated brand for USB Hubs. They are not cheap at all (this costs about 65€), but in general should provide a realiable data transfer without conflicts. It is powered and rated to 12V-3A and has led indicators for power and connection in each port. In addition is can be wall mounted, so is easly attached to the suitcase.
cam 1 & cam 2 & Usb hub: delock a-b premium USB 3.0 (3M)
This is the same cable that I succesfully used to operate the QHY163M, so I bought another two, one for the guide camera and one for the USB hub. It’s reliability has been tested with the QHY163M for 1 year without issues. It is double-shielded and the internal wires are 28AWG for data and 24AWG for power trasmission. The connectors are gold plated.
mount: lynx eqdir USB 2.0 (2m)
The previous cable for the mount worked reasonably well, but it had some unexpected disconnections. I didn’t found the reason for this, but I suspected form the quality of the internal chip (Prolific). For this reason, I searched for a better cable with FTDI chip, to replace it. The Lynx EQDIR cable is FTDI and has good reports for users that have used it without issues. In addition I liked to know that they are sold with lenghts up to 5m, so this tells something about its reliability for long cables. I have tested it and works without issues for now.
seletek focuser/rotator: generic rj45 24awg
I searched for the thicker AWG RJ45 cable, and this was the best I could find (at a reasonable price).
The Power Hub is a box embebed in the left side of the suitcase that provides power to all the equipment. It is capable to accept AC 220V or DC 12V and can deliver DC current up to 8 independent circuits. It has a total power of 340W (or 28A at 12V).
From outside, the powerhub has a control panel to control and monitor every device of the equipment.
Inside the enclosure, there is an AC/DC Hub, a fan and a temperature controller, a Fuse Hub and a Seletek Armadillo 2 (Focuser Controller).
The diagram above shows the schematic of the power circuit:
As an overview, AC power enters to an AC/DC Hub to convert and split it into 3 DC Lines. This DC lines are connected to a control panel where the lines are again splitted creating independent subcircuits for every device. Then one cable for each device exits the control panel and passes through a Fuse Hub to the Cable Hub, where the cables are grouped and routed to every device.
More in detail:
ac in + ac/dc hub
As I usually imaging from my backyard, I use a AC plug to power the equipment. The plug is controlled by a Main Switch that allows a quick power up/down of the whole system. When switched on, the AC power enters the AC/DC Hub, is converted and divided into 3 DC circuits as commented before (L1-PC, L2-DATA and L3-MOTORS). Each circuit goes then to the control panel.
The control panel includes a sequence of Switch-Voltmeter-Fuse for each incoming DC circuit (L1/L2/L3) and then is divided again into subcircuits for each device. Each device subcircuit is switched, and depending on how I want to monitor it, it can have additional monitors or controls. In total, there are 12 controls:
- L1-PC (19V)
- L2-DATA (12V)
- L2-CAM1 (12V)
- L2-HUB (12V)
- L3-MOTORS (12V)
- L3-MOUNT (13V)
- L3-FOC/ROT (12V)
- L3-DH1 (12V)
- L3-DH2 (12V)
- L3-AUX1 (12V)
- L3-AUX2 (12V)
- L3-AUX3 (12V)
A DC IN has been included to allow the power of the whole system with batteries. For L2 and L3, it is connected after the PSUs, and for the L1, it is connected to the laptop PSU to be converted from 12 to 19V (this is feasible because of the special features of the PSU used. It will be covered more in detail in the Power Hub – Component section).
A fan, controlled by a Temperature Controller, will be used to cool the Power Hub enclosure. It is wired to the L3-MOTORS line, but connected before the switch. This allows the cooling system to be operative even if the devices are powered off. This is important because once the Main Switch is switched on, the PSUs are powered (and producing heat), even when the L1/L2/L3 switches are switched off.
Some of the devices are very sensitive to power drops and can produce failures or even got damaged so it’s important to calculate the power needs. It is helpful also to choose the suitable PSUs, wire gauges, buttons, fuses…
I created 2 tables, to show the measured values (or expected or calculated, depending on the situation) of my equipment.
The tables have 3 different sections:
- Under the LOAD section, there are the values of voltage and amperage measured, and the resulting power.
- The section 12V EQUIVALENT LOAD, calculates the equivalent amperage for 12V voltage. No losses have been considered, so it is important to take this numbers as a (good) reference only. Apart of the amperage, the amperage per line, as well as the power per line is calculated. The Power/Line value in the one to take into account when choosing the PSUs.
- The following section, MINIMUM BATTERY, takes the pow/line value to calculate the minimum capacity of the battery needed. The following constrains have been taking into account to determine the capacity:
- Battery Voltage: 12.8V (LiFePo4)
- Max Discharge: 10%
- Required Autonomy: 16h (2 nights)
This table was populated by taking measurements during an standard astrophotography session. I noticed that some of this values where more or less stable, while other had a wide range of values, normally on peak situations. The laptop is a good example as it has a very different consumptions depending on the CPU/GPU load. In any case, it is suggested to be conservative and to choose a value representative of the avegare load plus a bit more.
To take this consumptions as a reference to choose the corresponding PSUs would be incorrect. It would give no margin for future expansions, and we could (and will) face some situations that could result in higher loads. That means that is necessary to consider the worst scenario that could happen.
Anyway, this table is useful to know the usual consumption of the system, and can be considered to chose the appropiate battery. In this case seem that a 120Ah LiFePo4 battery would be enough to power the equipment a 2 nights imaging session without problems.
The next table summarizes the maximum loads expected for this equipment. It is important to note that the values are, in most of the cases, choosen arbitrarily, taking into account the information found. For example, the HEQ5 is rated by the manufacturer as 12V/2A, but I have never seen consumptions like this on this device.
I chose to took the pow/line column values and use a safety factor of 2. As a result, we can consider the following requirements for the PSUs:
- L1-PC = 94W (100W)
- L2-DATA = 84W (80W)
- L3-MOTORS = 102W (100W)
I chose slightly different values to meet the usual rating of the commercial power supplies. It is acceptable to go 5W up or down the recommended, as they are widely overdimensioned.
Once we know the loads involved in the circuit, we can calculate wire gauges that will be necessary to avoid excessive voltage drops.
The first thing we have to do is to make a schematic, and identify the different sections. The following diagram shows it. It is important to note that it is not a very in deep analisys, but a rough one. For the internal wiring on the control panel, I assumed an approximated lenght and considered the maximum amperage for all sections. I think that this approximation is enough as in the end we will try to choose gauges slightly overdimensioned.
I created this spreadsheet to automate the calculations:
The main difficulty here is which current draw to consider as well as how much voltage drop is acceptable. Going to far can lead to massive wire gauges, that would be very difficult to manage and will provide not much (or any) benefit.
I haven’t used the maximum load for all the devices and I have permitted some voltage drops in devices that I know that they can manage it.
- The PC has a total voltage drop of 0.25V, but I’m powering it at 19.8V, while the manufacturer specifications suggests 19.0V. That means that I don’t have to worry too much about voltage drops (I measured 19.6V on the input connector of the PC).
- The CAM1 has a voltage drop of 0.25V too. I didn’t measured the real voltage at the input connector, but the camera works properly. In fact, it is supposed that the power input of the QHY163M is used only to power the TEC cooling, so it doesn’t affect to the communications with the PC. The calculated AWG value for L2-CAM1 on the table is 15, but if the corresponding voltage drop is changed to 0.16V (+0.1V) the AWG would be 16. I considered that 16AWG would be enough for this device. In the previous version of the AstroHub I used a massive wire of 12AWG to power the camera, with almost no benefit, but that made the grouped cables in the cable hub to be too much stiff.
- Both L2-IN and CP, and L3-IN and CP current draws have been chosen arbitrarily, instead the maximum rate of their corresponding PSUs. I’ve made stress tests to the AstroHub and I haven’t seen more than 2.5A on any of this lines, so the chosen values are good enough (a very conservative value, in fact).
- My only concern about the gauges was the cable for the Focuser/Rotator. It will be covered on the proper section, but I can advance that I made some tests and it works correctly with a 24AWG cable (and even with a 26AWG).
This is the list of components selected for the AstroHub 2.0. They have been selected to meet the requirements calculated above, but I also looked for some other features, like to be the smallest possible, and to be cheap.
As stated at the begining of the article, I chose to use 3 different circuits, so 3 PSUs are needed.
L1 – PC: TACENS ORIS DUAL II 100W – 12-24V 4-5A
To power a laptop is not as easy as the other components because most of the laptops work at different voltage than 12V. Mine is rated at 19V, and thats a problem in terms of power flexibility. The solution that I found is an universal adaptor from the brand Tacens. It has a wide range of voltage outputs (12 – 24V) so can be used with future laptops if necessary. It’s cheap, about 28€, and the most important, can be powered with either 12 DC or 220V AC.
It’s realibility is very good with no power failures in 2 years. The frame is from aluminium and the heat dissipates correctly so is always reasonably cool.
L2 – DATA & L3 – MOTORS: MEANWELL EPP-120s-12 – 12V 10A
This is a high quality power supply from the brand Meanwell. It is extremely compact (76x50x28mm) and very powerful (12V/10A), perfect for an application like this. It doesn’t has enclosure, so it has to be well isolated from other components to avoid issues.
Full specifications can be found here.
This device is used to stabilize the voltage of the mount. It can manage an input from 1.25V to 26V DC and gives a constant output voltage (adjustable) with a maximum load of 3A. I noticed that this device gets hot during the stress tests so It will be necessary to put an eye on how to cool it.
I was looking for very small illuminated button switches and finally found the brand LANBOO, that produces high quality switches with small dimensions.
I choose 16mm diameter buttons for the main switches and 8mm buttons for the individual equipment switches. All of them are latching operation and the actuator remains sunken when pressed. This makes really difficult to push it accidentally, avoiding unwanted disconnections. They have a led ring in a color that can be choosen.
I choose the following colors:
– AC Main Switch: RED
This makes very easy to identify to which line belongs every button.
The led is powered by separated pins, so one can choose the operating mode. I wired them to turn on the led when the button is also turned on, except the AC switch, that I wired to turn on the led when the AC connector is plugged. That gives me the feedback that there’s power coming into the AstroHub (important to be aware that theres AC power flowing inside).
On the previous version of the AstoHub the multimeters where simple Voltmeter-Amperimeters designed to be wall mounted. They did a very good job and gave me enough information to know what was going on, but there are other multimeters on the market that give much more useful information. The problem is that the ones I found where too big to be implemented on this project. Finally I found a very nice model of multimeter that gives a lot of information in a really small package. It measures only 48x29mm (44x21mm when disassembled, I will explain the reason of this statement in the construction section), it has a alarm fuction for high-low voltages (still investigating this feature), the backlight can be dimmed, and on top of this, the wiring is very easy… It’s really marvelous.
FUSE and fuse holder
I chose to use 5x20mm fast blow fuses to protect the equipment. In addition, I ordered a very small fuse holders, that are designed to be soldered to PCB, but can be used alone too.
In a inicial state of the project I was completely sure that the way to go was the DewHeater Controller developed by Robert Brown. But then I realized that I don’t imaging in locations where dew is a problem. Maybe I have had 1-2 foggy nights in the last 2 years. Taking that into account, I planned a very simple dew heating system consisting in little PWM controllers to regulate the load on the dew heaters.
This controllers are very small and they are supposed to be able to handle up to 10A (not sure of this capability). I modified them to be fitted in the control panel. This modification will be covered on the Construction section.
During the stress tests I noticed some malfuctions on this devices. I have to investigate more to be sure of it’s reliability and to know if they are introducing noise on the circuits. If they fianlly don’t pass the stress tests, I have considered the option of bypass them and use the dew heaters without regulation. As my dewheaters are self-made, they will be designed to keep a safe temperature even at full load 100% of the time.
This device controls the temperature inside the Power Hub enclosure and gives power to the fan when the temperature reaches the limit temperature. It includes a temperature probe.
This is a feature that I wanted to include on this new design to reduce the audible noise of the fan when powered, but it is not mandatory. In fact, the fan that I chose for this AstroHub 2.0 is very quiet, so it doesn’t disturbs so much, but I wanted to improve the noise that the previous version emitted. The previous fan was very big (and not specially quiet) and was always on, so always making noise.
I didn’t want to make experiments with this component, so I went directly to a fan by Noctua, a well known brand for its quiet fans. I chose the 60mm model, that fits perfectly on the side panel of the Control Panel. This fan rotates at 3000rpm but includes 2 special cables that reduce the speed of the fan, making it even quieter. It normally operates at 12V/0.08A, a very low power need.
I have been looking for shielded power cables, to have better protection against EMI, for a long time and they are very difficult to find (at least for me). Finally I discovered the specification UL1185 for shielded power cables. Probably there are othe specifications like this, but I didn’t found them.
They are stiffer than parallel style cables, and even when I had no problems with parallel cables when used in the previous version, I preferred to go to this type of cables, just in case…
Once the components are ordered and received I started the 3D development. This is not mandatory, but very recommendable in order to fit all the components accurately.
I measured every component as accurately as I could and made a 3D of each, to create an assembly of the whole enclosure. The most accurate the 3D is, the less modifications will be necessary in the construction.
The AC/DC hub is mainly composed by the 3 PSUs described on the components section. The goal was to find the distibution that needed the minimum volume. I wanted also the PSUs to be grouped and near to the AC IN plug.
The internal depht of the suitcase is really small, about 110mm, and the control panel’s thickness is 35mm, so the avaliable height inside the encosure is only 75mm. The PSU L1-PC fill this space completely.
The control panel measures 313x145x35mm, a very limited volume to fit 12 buttons, 5 multimeters, 3 fuses, 2 PWM controllers, a buck-boost regulator and the corresponding cables.
The outer face of the control panel was designed to be as flat as possible, mainly to facilitate the manufactutring in 3D printing. It has been designed to allow a an easy control and monitoring of every device, using a simple but efective distribution of the controls, that follow the natural wiring routing.
Inside the control panel is where the components has to be attached. One of my concerns when designing it was to achieve a reliable and accurate assembly of the components. This is why there are many features (housings) designed to meet this requirements.
One example of the housing development is the multimeters housing. As I explained in the Component section, the multimeters have been disassembled from his original frame, resulting in a smal LCD screen attached to a PCB. The multimeter windows have been reduced to the minimum, so the centering and the rotation alignment must be very accurate. To achieve this, I designed a frame that fit directly to the screen. This frame only touches the screen by 8 tangency points, avoiding to touch the inner vertex of the frame. This is important because all the printers make a small radius on this inner vertex, making necessary additional machining processes to adjust the fitting.
Another example can be the PWM knob housing. The PWM potentiometers are wired as a separated component to fit into the panel. This potentiometers are too long to be directly screwed to the surface of the panel, as it would result in very high knobs outside the control panel. The solution has been to make «bridges» that allow the potentiometers to be mounted more deeply, and that can be operated by a self designed knob. As a result, the knobs have almost the same height as the buttons.
While I was constructing the control panel I realized that the AC switch would have the terminals uncovered. This is a really dangerous situation in case of having to manipulate the inside of the control panel while powered (for example, in case of having to replace a fuse). Because of this, I designed a cover for this switch.
FOCUSER CONTROLLER (SELETEK)
The Seletek Armadillo 2 needed a deep modification to fit inside the power box. The original design had a very big box, and the connections where too big also (1 USB + 1 DC IN + 2 DB9). Even discarding the original enclosure the connections didn’t reach the desired volume, so the next step was to «bypass» some of the connections and replace them by smaller ones. The procedure to change the connections will be covered on the construction section, so in this section we will only see the improvements on the enclosure.
The following image shows the Seletek on his new enclosure. Apart of being big, the original enclusure didn’t allow access to the additional USB port (seen on the upper side of the image) as well as to the internal fuse (not seen in the image, but that is inside the big hole next to the additional USB.
The enclosure has been designed with a hole pattern that allows a good cooling of the device, but also to attach other custom-designed devices (like the Fuse hub) to it. It has also feautres to be attached to the bottom of the suitcase.
The height of the new enclosure has been optimized to be as small as possible, but giving enough space under the PCB, to allow the wiring of the new connectors.
The following image is a comparison between the original enclosure and the new one.
The fuse hub is a device that I designed from the begining, but that I was (and currently am) considering to remove. I have read statements from people with much more knowledge than me about electronics, pointing out that the fuses are not intended to protect the equipment, but to protect the wires, so maybe is a no sense to have more than 1 fuse for each line. In any case this device works also as a connection hub, allowing to securely connect all the cables that come from the control panel and have to go to the equipment (please note that has no sense to try to solder directly cables from the equipment to the control panel. It would lead to big difficulties in case of having to remove or replace any power cable, or in case of having to repair the control panel if necessary). By now, it has ben designed and implemented, but it is very easy to remove it if necessary.
The Fuse hub is composed by a PCB, some fuse holders and its corresponding connectors. I used KF-301 connectors, that are extremely compact and reliable.
KF-301 connectors are, in my opinion, better than the popular Anderson Powerpoles for this application. They are smaller and the cables can’t be disconnected because they are screwed.
As a comparison, a simple 2 pole connection can be done by putting two KF301 connectors face to face, measuring 15.2x10x14mm or using Powerpole, measuring 16x8x42mm. This is a huge difference in a development like this.
As pointed in the introduction of this section, the devices generate heat, and some of them need to be cooled. In the first version of the AstroHub I went for the simplest solution, and I putted a fan that I already had, and placed where it was easiest to install. This resulted in a poor efficiency cooling. This time I wanted to make something better. One of the key inefficiencies was that there wasn’t a airflow. There wasn’t an intake on one side and an exhaust on the opposite. The flow was created only by the insanely big fan that was sukken air from everywhere. In addition, the previous panel was opened right aside of the fan, creating a very short airflow, that let some areas of the power hub without flow at all.
In this new design the fan was planned to be placed on the side of the power hub, giving the opportunity to make something on the opposite side. The first idea was to make intake hollows on the fornt surface of the control panel, but this idea didn’t liked me so much as I was concerned about the possibility to the dew to «fall» inside the power hub. Then I was inspired by the laptops intakes and realized that could be implemented in this design.
I chose the left side of the control panel to include the intake
This would allow a correct airflow even with the lid of the suitcase almost completely closed. In addition, the fan exhausts the hot air inside the cable hub, creating a «free flow chamber» as the cables are not into the cable hub when the power hub (and the cooling system) is working. The buck-boost was placed next to the intake to take profit of the fresh air coming from the ambient.
CONSTRUCTION AND ASSEMBLY
The construction of the Astrohub is a long task that took me more than a month to finish, but is not so difficult. In fact, the steps are quite simple and the difficulty is that there are so many steps (specially in the control panel) and is important to execute every one with precision and patience.
This section is intended to be a visual guide, with many images and less text than in the development section.
TOOLS AND MATERIALS
Most of the tools needed to build this Astrohub are common ones (screwdrivers, pliers, cutter, multimeter…). Another story are the 3D printers. They are not so common but they are becoming cheaper and cheaper, so is an investment that I recommend for sure.
This is the execution of the power hub.
From outside, is a clean and flat design, with all the controls well distributed, lateral connections and an automatic cooling system…
…but inside, it contains a real cable mess. This the goal, to keep the cable mess far from the equipment.
I began the construction by the control panel, as is the most complex part.
This is an overview:
I suggest to execute this construction with patient. Check every connection twice, route the cables carefully to avoid messes, and solder the components in a proper way.
The first step is to print the parts. I used black ABS for all the pieces. As explained in the 3D development section of the power hub, this parts are bigger than my 3D printer, so the 3D had to be splitted by the central rib into two parts. The following image shows the upper part of the control panel.
The paths of the printing are very noticeable and will not disappear once painted. It is only an aesthetical issue, but I wanted to improve it, so I made some additional processes to achieve a flat and smooth appearance.
Once printed, the 2 parts of the control panel were glued using cyanocrilate on the central rib. In addition I applied this glue to the areas where the filaments seemed to be prone to detach. This is specially true on the edges of the piece (outer perimeter, multimeter windows, button holes…).
Then I sanded the piece putting a sandpaper on the ground (or the flattest surface available), facing down the piece and gently making pressure on the whole piece. This would take care of deformations caused by inaccurate gluing and/or differential shrinkage of the parts, improving the overall flatness.
To sand the piece will make the surface flat and smooth, but it probably will not remove the path pattern. To totally hide it I decided to put plaster on the front surface and sand it again.
Now the front surface has become perfectly flat and smooth, and the path patterns created by the 3D printing have almost disappear.
I used a cheap matt black painting spray to paint the control panel (and all the other pieces). I tried first high quality spay painting (Tamiya) but the matt was too smooth and the fingerprints were very noticeable. Conversely this inexpensive paint had a very rough finish that made the control panel totally fingerproof.
Next step is to attach the components.
First, the buttons. Make sure to keep the same orientation of the pins on both of them. This will facilitate the wiring.
The next step is to install the multimeters.
I used hot-melt glue to attach the multimeters because is sticky but removable.
Next step would be to prepare the PWM controllers. I desoldered the potentiometer and extended it. The first model that I tryied failed in the tests, so ordered another model. It didn’t fit with the original housing that I already printed, so I had to stick it with hot-melt glue. Not so elegant, but does the trick.
The fuses where installed while wiring. I bended the terminals intended to be soldered to a PCB and connected the wires to them, finally protecting them with shrink tube. Then I glued the fuse holder to the control panel with cyanocrilate. I changed their distibution from the 3D to allow to operate them better.
This is the execution of the DC-IN RCA connectors. Again, terminals protected with shrink tubes.
The buck-boost is placed in his housing with screws. The nylon tyes hold a temperature sensor. This device gets hot when in use, so I wanted to monitor its temperature. The missing ribs on the intake were made ad hoc, to increase the incoming airflow in this area.
I extended the status led of the Seletek (in the center on the image) to be visible from the control panel, making a hole next to its switch and gluing the led with hot melt glue.
Once the components are attached and internally wired, it is time to wire the output cables. It is important to label them to avoid incorrect connections. It is very important also to avoid pulling the components to which they are connected. For this reason, I printed cable holders attached to the frame that keep the components totally isolated from the outer movements of the cables.
Now the control panel is prepared to be assembled into the suitcase:
Note that the last 3 switches on the L3-MOTORS circuit hasn’t been wired. They are reserved just in case of future need.
Note also that there are screw holders on the walls of the control panel. They will be used to hold a cover to protect the electronics, that is still in development stage.
Before attaching the control panel, the internal components of the Power Hub have to be installed.
I used nylon ties to fix the PC PSU (Tacens) to the internal shell of the suitcase by drilling holes on it. Apart of being fast, it saves much more height than by designing 3D printed holders. The PSUs for the lines L2 and L3 (Meanwell) were attached with PCB supports that I screwed to the suitcase.
I placed the temperature sensor of the temperature controller that controls the cooling fan between the 3 PSUs. The PSUs don’t become as hot as the buck boost but I liked this placement for the sensor because it is just above the buck boost when the power hub is closed, so in the end the sensor is in a «hot area» of the power hub, where any of the components involved can raise the measured temperature.
The modifications on this device were a litte bit scary. I have to admit that in the beginning I considered to unsolder all the interfaces and start from zero, and finally I only bypassed them, but anyway it was scary. Obviously, after making this modification, I have lost any kind of waranty from the manufacturer, but I took the risk. Also obvious that I decline any responsability about a user that could attempt to make this modification after reading this article. This is a DIY project that can only be done at your own risk.
So, as stated, the connections of the Seletek has been bypassed allowing the use of a smaller enclosure. The image above shows the resulting PCB.
The Seletek controller has 18 avaliable pins, for multiple purposes, but I only need 8, to control the motors of the focuser and the rotator, So I let the rest untouched.
As a result, a redesigned device, with extended connections, accesible fuse, and a coupling to be directly screwed to the internal shell of the suitcase.
The assembly of the Power Hub side panel can be done lately, as it is easier to manipulate the inner components without it. In any case, the execution is quite simple.
I haven’t repiclated the finish of the control panel, as it is not as visible, so I simply glued the two parts of the wall (again, it is a piece bigger than my 3D printer) with cyanocrilate and painted it in matt black.
The temperature controller had a housing like the used for the multimeters, so it was placed and fixed with glue. For the fan, I first attached a light net (previously I purchased a computer fan dust filter and it was too heavy, not allowing enough airflow) to the fan and then glued the fan to the side panel.
The fuse hub is the last component to install in the power hub. It is the moment where all the cables coming from the control panel are connected to its corresponding equipment cable, alllowing to power them. Again, very important to label the cables to avoid mistakes.
Its construction it’s as simple as to use a PCB board, solder the fuse holders, the KF301 connectors, and wire them. Then put the result in the enclosure, and connect the corresponding cables.
The Data Hub is a space where the cables come from the cable hub and go to the power hub or the USB Hub, depending on the case.
The USB hub it is simply screwed to the suitcase, as other components. This model includes a plate for this purpose. It has a power led and status leds on each port to monitor the connections.
It is important to choose the correct situation into the avaliable space to allow the eventual conection/disconnection of the cables, but without adding too much lenght to them. As analized in the development section 3m is the limit for USB 3.0cables, but the shorter, the better.
Here again is important to isolate the movements of the cables inside the power hub and also the data hub, so I installed a cable clamp, attached to the suitcase.
The side panel of the data hub is a simple wall of plastic, splitted as other pieces to be 3D printed, and then glued and painted. It has housings to screw it to the suitcase and the power hub, but I finally left it unscrewed. This allows a very quick release.
The Cable Hub is the space where all the complexity has to become into simplicity. Only 3 grouped cables, one for the incoming power, one for power and data of the computer, and one for power and data of the astrophotography equipment. All stored as a spiral, ready to pull and connect.
The spiral shape of the cables is key for this project. Grouped cables always tend to become stiff if tighly grouped. My solution has been to respect the spiral shape and manipulate them as a spiral cable, so I take them keeping the natural twist while attaching it to the mount. This facilitates the manipulation and the storage.
The construction of this cable hub begins using the calculated relative positions in the development section.
For the execution I left the cables on the floor keeping an eye on to release any excesive bending or twisting.
Then I placed every cable in the calculated position and grouped them as parallel as I could.
I fixed them in the RA Clamp point with a nylon tie looking for the way to group them to achieve the minimum diameter. I wrapped the section between the RA Clamp and the DEC Clamp. Note that the tube wrap has been modified by removing some of the sections. This reduces the stiffness a lot and allows to reduce the minimum radius that they can be bended.
As explained in the development section, the DEC Clamp hasn’t been developed yet, so I used the previous solution, a simple elastic cord that is clamped to the DEC dovetail screws. From the DEC Clamp, the cables are splitted in 3 groups. One goes to the guide camera (CAM2-USB 3.0), two of them go to a connection rig hub installed in the telescope and to the focuser and rotator motors (DH1+DH2-GX16 12V and FOC/ROT-RJ45 12V), and two go the imaging camera (CAM1-USB3.0 and CAM1-12V). The cables that go to the imaging camera have an additional elastic cord, installed to be attached to the camera body, to let the cables near to the camera and avoid touching the mount cables.
In the following image the cables are connected to the imaging rig, for clarification.
Once the cables are correctly routed, it’s time to wrap the remaining cables from the RA Clamp to the AstroHub. Then they can be fixed to the clamp located in the Data Hub. The shortest cable should be connected and it will be the reference for all the other cables, that should be shortened or looped in order to arrive to their corresponding ports.
For the cables that go to the PC the procedure is similar. First, fix their relative position.
I didn’t wrapped them, as they are only 2 cables, intead of this, I used nylon ties. The other end has to be clamped, and connected.
Put all the cables into the cable hub, trying to use all the avaliable space to describe the maximum diameter possible, this decreases the stiffness and excesive bending, and will let an empty space that can be used to store other stuff.
And finally, the result.
This is the purchase list of this project:
Prices are in euros (€) and are the prices when I made the order. The Keter suitcase, for example, was bought one year ago and the current price is 37€.
It can be done cheaper, of course. There are other alternatives for the switches, multimeters, cables, Usb HUB and fan. Not much saving can be achieved with the rest of the components (I was looking always for the cheapest components)
The list doesn’t include the Seletek controller (I have considered as part of the equipment, not a component of the AstroHub). Its price is 220€.
WORK IN PROGRESS
This section is intended to share some of the documents used to develop this project.