[QC102]

Thanks Cliff - I will  try that.
I added a screen grab of the other Revo (no pun intended) - I thought I had  included in post but seems not!
Nice to get some technology that  works well from BG...

[anthropo]

hello,

I have this results (altitude, in meters)  with this board



https://www.banggood.com/OpenPilot-CC3D-Revolution-Revo-10DOF-STM32F4-Flight-Controller-Staight-Pin-p-1000068.html

I suppose I have to remove the C9 capacitor but I don't know where the capacitor C9 is on this board

Can you help me please ?

[anthropo]

still searching. Seems that this regulator isn't the same ?
it's labelled R13c
https://www.google.fr/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwi7hbuy7OjVAhXDtRoKHeLwDOMQFggnMAA&url=http%3A%2F%2Fwww.ti.com%2Flit%2Fpdf%2FSBVS031&usg=AFQjCNFOIAxeSwhojRhFbUOYeDxwCTX9Ow

[Mateusz]

Hi anthropo

You should plot BaroSens.Altitude [m] where 'm' stands for units in meters, or kPh. Without units, not much can be said about plot.
Just compare your plot, with my earliest post https://forum.librepilot.org/index.php?topic=311.msg2157#msg2157

I only looked at BaroSensor.Altitude [m] on the left part of scope you can see Bad baro on the right part same scale you can see good baro after switching boards.
Bad baro (lowest 173.5, highest 182.5, difference 9 meters!)
Good baro (lowest 175.5, highest 177, difference 1.5 meter)

So bad baro is when you have more than 1.5 meter from most lowest peak to most highest peak. That is like +-0.75 (half of 1.5) from the average. This is in meters [m].

I cant help with capacitor, I identified it was faulty and send back Revo to seller immedietly, using my bayers protection from PayPal. If you don't send faulty product back, they might think it's ok to sell clones like this because no one complains.

[anthropo]

Thank you. I have a bad baro.  5m at least. Nothing to do but change board ?

[TheOtherCliff]

The problem is that manufacturer of this batch of Revo used cheaper, different spec part for voltage regulator.  Original regulator specified that it needs a large capacitor (large value, tiny physical size) for circuit #1, but replacement needs a smaller capacitor.  Replacement oscillates slowly if the capacitor is too large.

Solution is to completely remove large capacitor.  I have also heard that adding capacitance at the baro chip on circuit #2 (+3.3 to ground) will fix the issue.

#1 and #2 are just my titles to make it clear there are two different circuits, one where you remove capacitance and another where you add it.

There is a post in a thread that shows which to remove.  Some talk about C8 on Sparky2 board.
https://librepilot.atlassian.net/wiki/spaces/LPDOC/pages/12058671/Altitude+Hold#AltitudeHold-Hardwareissues
https://forum.librepilot.org/index.php?topic=311.msg2246;topicseen#msg2246
https://forum.librepilot.org/index.php?topic=311.msg3705;topicseen#msg3705
https://forum.librepilot.org/index.php?topic=311.msg13515;topicseen#msg13515

[anthropo]

I finally searched, found and removed the C9 capacitor on my chinese clone labelled Rev. C

for those with the same board the culprit is cercled in red

Much more better, but still sometimes peaks about 4 meters

[TheOtherCliff]

Those spikes look like something else.

I would install the desired 490pf at C9 if I had one and/or add some capacitance at the baro chip from ground to V+.

[anthropo]

I decided to use the 10nF capacitor to put over the C29 like described in this thread. Then it works !