05-10-2016, 08:58 AM
I had initially thought to leave space for the headers but it caused the lid to be overly complicated, that was when had the ribbon cable through the side idea. Also, it seems that powering the fan with the pi header may be too much current for the board. Currently I have the board being powered by a supply capable of 2.4a power port, with two purrs available. These are very high quality heavy supplies. I had the system lock up under load several times while the fan was connected, I am not sure if it's a problem at the board out with the cheap amp/volt meter I am using to monitor the power usage. With the fan connected and the system at idle it's using about 400ma disconnected the system uses around 200ma. When the fan is disconnected I have seen the power usage spike to around 800ma, so combined it should be right at or just over an amp which I don't think should be too much as I understand it should be able to handle that (both the supply and board) but it locks up. I will have to take the meter out of the loop to check to see if it's the culprit. I did notice that pin 2 on the pi header is always hot, even when the board is "powered down" which leads me to believe that the 5v pins pi header are being fed directly from the USB power port, but I could be wrong I have not looked at a schematic for the pine yet. I think one is available on the wiki so I may search around to find out later. It certainly is not switching off when the board is shut down and the power LED is off. It locked up both under Ubuntu and Android, which I run on separate cards, so it's not that.
Anyway speaking from a design standpoint the fan I am using would sit nicely and probably a little snug between two walls two studs apart. A side blowing fan might work well but the case would need to be much higher, what about building a duct instead. This would leave fan horizontal and would blow/suck air across the top and bottom and also look neat. I think you would need to raise the board to a full three or four single heights to do and would need added under board support if you plan to use the headers I think. I am currently supporting the center section and directly under the headers with plates (the smooth ones) this way the board doesn't flex under Purdue when connecting to the headers. That flex strains the solder and I don't like it so I fixed it. I have done a lot with this board that makes my phone almost to hot to hold and never has the memory gotten much higher than ambient temperature. The main processor gets warm but with the sink on it isn't bad at all... And if I figure out the lockup issue, then I shouldn't need to worry with that fan on at all.
If the fan is drawing too much current for the board then maybe a power port, that just provides power to the headers would be prudent, maybe I will get one of the stackable project boards and rig it for an external 5v supply if not already capable.... I would have to cut out space for the chip to get airflow probably but in that situation a side on airflow may be better.... Maybe I will post in another area of the forum and ask if anyone knows the amount of current that the headers are able to utilize safely...
Anyway speaking from a design standpoint the fan I am using would sit nicely and probably a little snug between two walls two studs apart. A side blowing fan might work well but the case would need to be much higher, what about building a duct instead. This would leave fan horizontal and would blow/suck air across the top and bottom and also look neat. I think you would need to raise the board to a full three or four single heights to do and would need added under board support if you plan to use the headers I think. I am currently supporting the center section and directly under the headers with plates (the smooth ones) this way the board doesn't flex under Purdue when connecting to the headers. That flex strains the solder and I don't like it so I fixed it. I have done a lot with this board that makes my phone almost to hot to hold and never has the memory gotten much higher than ambient temperature. The main processor gets warm but with the sink on it isn't bad at all... And if I figure out the lockup issue, then I shouldn't need to worry with that fan on at all.
If the fan is drawing too much current for the board then maybe a power port, that just provides power to the headers would be prudent, maybe I will get one of the stackable project boards and rig it for an external 5v supply if not already capable.... I would have to cut out space for the chip to get airflow probably but in that situation a side on airflow may be better.... Maybe I will post in another area of the forum and ask if anyone knows the amount of current that the headers are able to utilize safely...
