×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Astrohat - An open hardware RPi Hat for astronomy equipment

  • Posts: 1208
  • Thank you received: 559
If the goal is to power everything through this, I'd worry about the current draw of a mount while slewing. I'm pretty sure I've seen my Atlas Pro draw 4 or 5A, but just for a short while.

USB outputs might be nice, but I suppose one could just connect a powered USB hub to this hat.
4 years 4 months ago #46343

Please Log in or Create an account to join the conversation.

  • Posts: 39
  • Thank you received: 38
That sounds a lot, my EQ6 fully loaded barely passes 2A while slewing. In any case 6A per channel should be OK as long as the neighboring channel (with which they share the load) is on less than 0.5A (they share 6.5A rating)

I plan to do some extensive testing on the loads to confirm the design assumptions.

There is no way to get USBs through the GPIO header of the RPi, so that leaves you with the option to sacrifice one rpi port and put a hub. It makes little sense to integrate functionality in this hat unless it can be supplied through the internal GPIO header.
4 years 4 months ago #46344

Please Log in or Create an account to join the conversation.

  • Posts: 39
  • Thank you received: 38

Since we are going for an aluminum case, the antenna needs to be external. It also allows for flexibility of placement for the whole system. Having an external thing though got me thinking.... why not get the GPS completely external (IC + Antenna) and thus save on internal space (since we are already packed) and also allow for a possible usage of the TTL (serial TX/RX, VCC, GND) port for something else (in case someone does not want the GPS). Such GPS modules are cheap and look like this . Makes sense?
4 years 4 months ago #46345

Please Log in or Create an account to join the conversation.

  • Posts: 39
  • Thank you received: 38
I pushed some changes on the repo and concatenated a bit the BOM in order to do some price calculations. Here is what we are looking at:

All the components I selected are high-grade ones. Automotive and Industrial grade capable of performing to extended temperature ranges (since it is highly important imho for our setups) and less prone to EMI. I did not "cut-corners" at any point during the design and component selection process. All in all if someone was to buy today all components for the board you would be looking at around 45 USD. PCB production would be around 5-10 USD depending on your source. (w/o shipping obviously).

Now those are prices for a *single* board, and without trying to make everything less expensive (e.g. searching for cheaper connectors, cheaper RTC, replacing some Caps with cheaper options etc).

Just to get an idea about how prices differ on 100x scale, if you were to produce 100 boards like this you would be at around 35 USD including assembly! (with a possibility to get a bit lower while still keeping quality). (still w/o shipping ;) )
Last edit: 4 years 4 months ago by Pierros Papadeas.
4 years 4 months ago #46347

Please Log in or Create an account to join the conversation.

  • Posts: 1067
  • Thank you received: 140

Well as far as I am concerned that is an absolute bargain, and I would pay double for it...mind you I am in the UK so I have all the import duties and taxes too.... :)
4 years 4 months ago #46350

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182


This cooler is definitely cutting it! Keeps my Pi4 at 45 C and below!

smile.amazon.com/gp/product/B07V7SS5BZ/r..._title?ie=UTF8&psc=1

Highly recommend, as long as you can 3D print your own case for it.



BTW, I would be happy to share the FreeCAD and STL files, if anyone wants them (and hopefully improve on them). They are not very elegant, but functional for my purposes.
Last edit: 4 years 4 months ago by Jose Corazon.
4 years 4 months ago #46351
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 460
  • Thank you received: 69
Anyone able to guess if the hat alone would fit onto a Odroid-N2 without modification?

www.hardkernel.com/blog-2/odroid-n2/http...om/blog-2/odroid-n2/
Last edit: 4 years 4 months ago by Jerry Black.
4 years 4 months ago #46352

Please Log in or Create an account to join the conversation.

  • Posts: 1067
  • Thank you received: 140

All well and good but would not work on what this thread is about...... :)
4 years 4 months ago #46354

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
"All well and good but would not work on what this thread is about"


I'm not so sure, it just might be possible. That's why I showed the image with my current cooling solution.

I agree that with the conventional HAT design, which sits directly on top of the Pi board this would not work. That would fatally interfere with the cooling, whichever way you try to cut it.

However, this tower is hands down THE BEST cooling solution for the Pi4 (check out Christopher Barnatt's video on this cooler and how it compares to other solutions:
)

So I thought it might be worth exploring at this early stage, whether the AstroHAT can be designed in a way that allows incorporating this top of the line cooler for the RPi4 into the assembly.

I just think that might be possible, all that is required is to add some GPIO connection extension adapters that allow it to be moved up above the cooler. The 3D designs can easily be modified to make a case that incorporates the HAT with the cooler sandwiched in between the HAT and the Pi4 board and forming essentially a wind tunnel the cooling air is being sucked through, thus cooling the Pi4 and the HAT at the same time.

Worst case scenario, one can just use a cable to connect the HAT to the PI4 board ( core-electronics.com.au/raspberry-pi-b-g...-cable-40-pin-6.html ), but having a solid GPIO extender of the correct length would be preferable. That must exists somewhere.

It may be sufficient to stack 2 of these connectors on top of each other to clear the tower: cnclablb.com/raspberry-pi-gpio-tall-header-2x20.html

I would LOVE to have that construction on my rig. At the moment, I have 3 power lines running up the rig, one for the 12V needed to run the mount and the dew strips, one to power the USB hub and one to power the RPi4. If I could replace that with a single power line running to the HAT/RPi4 and everything else being powered off that, that would be ideal.
Last edit: 4 years 4 months ago by Jose Corazon.
4 years 4 months ago #46355

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
Adding to my last post, Pierros, if you think that would be feasible, then I would suggest including a ground and a +5V pin on the BOTTOM of the board, to allow for the direct connection of the tower cooling fan (without having to run the wires around the HAT board and connect them to the pins on top).
4 years 4 months ago #46356

Please Log in or Create an account to join the conversation.

  • Posts: 27
  • Thank you received: 9
Active cooling is probably not required.
The RPI foundation have recently released a new firmware for RPI4 with better thermal management.
There are passive cooled cases which rival active cooling now available.
Ambient temperature outside at night works to reduce heating.
Finally, active cooling adds bulk and complexity.

I'd like to add a vote for focus motor driver be added
locking or latched DC connectors would be icing on the cake
Last edit: 4 years 4 months ago by Rob.
4 years 4 months ago #46360

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182

Maybe not required, but for most of the time that Pi4 runs at 30 degrees C at RT. That can only be good. It crashes far less than my also actively cooled Pi4, but that one only has a fan blowing down on the heatsink. That goes up to 70 degrees C, which is when it is being throttled. I suspect that's one reason why it crashes more often.

No matter, with the design Pierros has posted so far, I can make this work with my large active cooling fan.

Really looking forward to implementing this!

Best

Jo
4 years 4 months ago #46366

Please Log in or Create an account to join the conversation.

Time to create page: 0.593 seconds