BQ24297 Thermistor (NTC) resistors - Hot fault


So your fix is to remove R710 and use your modified thermal script, and it improves charging performance? Is that right?
Yes.

...but, I also have an air gap between CPU and battery, and some heatsink upgrades.
...and, I'm surprised that removing R710 worked. I expected to need to replace it, so be warned.
 
2A_charge.png
The battery discharged 65% in under half an hour?
 
The battery discharged 65% in under half an hour?
Yes, at 1.5Ghz the processors are power hungry. I think the original pyra-thermal.sh script limits 1.5GHz to short bursts by not letting the battery management chip warm up. That way you get better battery life but you don't have the option to run at 1.5GHz continous with a charger plugged in.
 
With dual 1.5GHz I see 8W power usage (~2A current) . The battery is 6Ah but we don't get all of that to protect the battery and not overvolt the modem. I think 1 to 2hours from full to flat at flat out dual CPU, depending on when the CPU gets hot.

Dual 1GHz is 4.5W (~1.1A) so more like 4hours.

Normal browsing and editing is 2.5W so 6 to 7 hours.
Post automatically merged:

Power.png
 
Last edited:
Any thoughts for a future battery upgrade?
It depends on your use case. Do the emulators need dual core full throttle? Compiling stuff uses a lot of CPU. The battery is plenty for browsing or playing music.
I've got a 15Ah Anker battery pack that fast charges the pyra fine at high load.
 
An external battery could have any form or dimensions - it then could dock to the Pyra as the battery door does. Think brief case with a Pyra attached to its side. I wonder, if that would still go in matzesu's pockets.
 
What of you build in a Pandora Battery into the Pyra and use the MorePlace in the in the Case Battery Place for more cooling Solutions?

EvilDragon Said that the Pandora Battery should fit with the Connections but you must anything for make the Battery Fix in this Place.

The Pandora Battery should have 4000mah and so do you maybe more Place for Cooling Solutions/Experiments


Just an Idea.
 
An external battery could have any form or dimensions - it then could dock to the Pyra as the battery door does. Think brief case with a Pyra attached to its side. I wonder, if that would still go in matzesu's pockets.
I'm not sure I want to have to pick up a briefcase every time I want to play on my Pyra to be honest. Or you do say dock, so it should be removable, but if you remove it from a big battery pack, you've now got a pyra without a battery, no? Or are you imagining some sort of dock that extends the battery contacts out of the case? Seems simpler just to plug a battery pack into the charge port and use that to be honest.
 
I'm not sure I want to have to pick up a briefcase every time I want to play on my Pyra to be honest. Or you do say dock, so it should be removable, but if you remove it from a big battery pack, you've now got a pyra without a battery, no? Or are you imagining some sort of dock that extends the battery contacts out of the case? Seems simpler just to plug a battery pack into the charge port and use that to be honest.
Not sure it was entirely serious \o/
 
Or you do say dock, so it should be removable, but if you remove it from a big battery pack, you've now got a pyra without a battery, no?
Si. Said "brief case" would have have some protruding form on the side that's in the shape of the normal battery door and battery. And indeed, I wasn't too serious. Though, if that 'brief case" would have the same footprint as the Pyra and was only a few cm thick. ?
 
OOohh... that's a very interesting thread!
I already know that the air gap between the plastic and the CPU board happened because the old case company didn't give us one part of the mould which is needed to make the plastic in that area thinner.
This is why it worked with the prototype case but now causes issues with the final case. We're working on improving that and if all goes well, it should be possible sending everything who already received a Pyra a new backside which can easily be replaced :)

So... if you remove the R710... it still shuts off at some emergency temperature?
Or was that simply done by my thermal script?

That would be an easy fix for the PCBs that have been produced already before I ship them out.
However, we need to make sure it cannot cause any burns, etc. even with the thermal throttling script disabled.
 
So... if you remove the R710... it still shuts off at some emergency temperature?
Or was that simply done by my thermal script?

That would be an easy fix for the PCBs that have been produced already before I ship them out.
However, we need to make sure it cannot cause any burns, etc. even with the thermal throttling script disabled.
Yes, I removed R710 and I now see the _hardware_ BQ24297 thermal cut-off detect the battery overheat at 39C and cold at -15C.

With R710 in place, the BQ24297 NTC hot fault triggers before the battery is hot (I guess ~30C) This is why the power light flashes during charging and the charging stops early.

Also, your pyra-thermal script uses the BQ27421 temperature reading but this is NOT connected to the battery. It is reading the motherboard temperature. The motherboard gets hot because it is attached to the CPU so we also stop charging because the CPU is busy. We can change this back to "bq24297" at the start of pyra-thermal.

We need to confirm all this on another board. If you remove R710, what resistance do you measure on the pads? Is the problem just my board or all production boards?
 
Last edited:
Back
Top