User avatar
Imperf3kt
Posts: 2865
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Brief voltage drop on UPS like system. would a capacitor fix it?

Wed Oct 10, 2018 11:19 pm

Hi mosespi, thank you for the info!

A sudden thought struck me, do you think it would be possible to replace the resistor with say, a 200k trimpot or rheostat?
I'd like to test what value works best once I manage to get it working again, but dislike the idea of repeatedly soldering and desoldering resistors.
55:55:44:44:4C
52:4C:52:42:41

mosespi
Posts: 508
Joined: Mon May 12, 2014 3:35 pm
Location: 34,-118
Contact: Website

Re: Brief voltage drop on UPS like system. would a capacitor fix it?

Thu Oct 11, 2018 8:57 pm

Imperf3kt wrote:
Wed Oct 10, 2018 11:19 pm
Hi mosespi, thank you for the info!

A sudden thought struck me, do you think it would be possible to replace the resistor with say, a 200k trimpot or rheostat?
I'd like to test what value works best once I manage to get it working again, but dislike the idea of repeatedly soldering and desoldering resistors.

Shouldn't be a problem. Maybe put a resistor in series to give you a minimum resistance (50K maybe?), because if you crank the trimpot one way you essentially have a short.

Regards,
-Moses
Power problems? MoPower UPS for the Pi
http://www.allspectrum.com/mopower/

User avatar
Imperf3kt
Posts: 2865
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Brief voltage drop on UPS like system. would a capacitor fix it?

Sat Oct 13, 2018 8:59 am

Well... I think I'm going to throw this thing in the can. I removed all other variables that may affect power, and tried to power my Pi as directly as possible from the MP2636.
However, it appears not all MP2636 are created equally, the new one I just modified tonight behaves radically different from my last module.
I am seriously disappointed in this unit and very much disheartened by months of failure.

I no longer think ANY portable Pi is possible. The Pi simply uses too much power. On the bright side, powering via battery works amazingly well, even manages to run cpuburn on all four cores no sweat (until it gets over 82°C at which point it thermally throttles on me) as long as you don't use the official raspberry pi power supply - plugging that in to charge the battery causes a high pitched whine, the screen starts distorting and the red LED goes haywire.

Time to find a use for all this junk now :(
55:55:44:44:4C
52:4C:52:42:41

User avatar
Imperf3kt
Posts: 2865
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Brief voltage drop on UPS like system. would a capacitor fix it?

Tue Jan 01, 2019 9:06 am

Okay, I understand it has been some time since my last post, but I never actually binned anything xD

I'm posting today to update for everyone a couple of things.

First, I have no idea how to use an oscilloscope, being unable to even calibrate mine.

Second, I've succeeded!
Well, mostly. I'm happy enough with the results.


Effectively, I ended up placing a 100uf capacitor across the 5v and GND out from the MP2636, and used a 5A power supply (modifying the Pi to accept power via PP6 / GND and the MP2636 to accept a different power source than the micro USB port was necessary).
Now when powering via either source, the Pi shows no sign of undervolt even under high load.
Switching still shows a few seconds of undervolt, but all other instability previously experienced has disappeared. I'm thinking ~220uf might be a better capacitor size and may eliminate even this issue.

The boost light on the MP2636 seems to not be solid, (when not turned on) but this is of no concern (it just means its using less power anyway)
My major headaches with my last attempts came from a faulty MP2636 unit. Using a different unit, everything is working as expected.
55:55:44:44:4C
52:4C:52:42:41

Return to “Advanced users”