Why does an IMU need to warm up?

[jimheid]

This is a "I'm curious to know what's going on under the hood" question, not a problem report.

I'm writing some stuff about the Mavic Pro and would love to include some technical details on what's happening during IMU warmup. Does the IMU actually need to be at a prescribed temperature to work reliably? Does it have a tiny heater inside? A set of vacuum tubes, perhaps? (Kidding about that last part.)

Thanks in advance to anyone willing to share some geeky details.

 

[HowSR]

Who says an IMU has to warm up?

 

[jimheid]

Who says an IMU has to warm up?

The DJI GO software -- it displays "IMU Warming up" for a few seconds very shortly after you power-up the drone.

 

[Robert Mitchell]

I'm not an electronics engineer but yes, I believe that IMUs do have a temperature range within which they function most accurately.

From Wikipedia, Inertial measurement unit - Wikipedia "- environment sensitivity: mainly sensitivity to thermal gradients and accelerations."

There wouldn't need to be a heater of any kind though, as even solid state electronics warm naturally when functioning. Thus, the need for a fan on the Mavic to carry away the excess heat.

Others can give more precise details, or correct me if I am wrong, but this is the way I understand it.

 

[BorisTheSpider]

My understanding from looking at csview is that, looking at the IMU temperature parameter which is recorded, it's very tightly controlled during operation so the output must be quite highly temperature dependent. I also don't know the physical reason for the temperature dependency and would be interested to know.

 

[AlanTheBeast]

This is a "I'm curious to know what's going on under the hood" question, not a problem report.

I'm writing some stuff about the Mavic Pro and would love to include some technical details on what's happening during IMU warmup. Does the IMU actually need to be at a prescribed temperature to work reliably? Does it have a tiny heater inside? A set of vacuum tubes, perhaps? (Kidding about that last part.)

Thanks in advance to anyone willing to share some geeky details.

Some accelerometers are based on a heated gas in a chamber resulting in the warmest gas being at the top (v. gravity). Change in speed causes the mass of heated air to deflect and that is measured. (I don't know if that is what is in the Mavic or if such is practical in an airborne vehicle)

Many (most) accelerometers are a micromachined arm that deflects during acceleration and since the electronics will warm up at some point, the measurement is calibrated to work best at their "warm" point or higher. Newer MEMS do not need to be pre-heated. Not sure if gyro sensors need to be pre-heated.

(If you have an iPhone there is an app called Sensor Log where you can observe the accelerometers and gyros in the iPhone. You can even output them via WiFi IP to a local network - telnet them and watch; record, etc.)

 

[jimheid]

Thanks VERY much! These are exactly the kind of details I was hoping for. So fascinating that so much precision can be built into such a tiny package. (And I'm referring to the MEMS as well as the Mavic as a whole here!)

 

[AlanTheBeast]

This is a "I'm curious to know what's going on under the hood" question, not a problem report.

I'm writing some stuff about the Mavic Pro and would love to include some technical details on what's happening during IMU warmup. Does the IMU actually need to be at a prescribed temperature to work reliably? Does it have a tiny heater inside? A set of vacuum tubes, perhaps? (Kidding about that last part.)

Thanks in advance to anyone willing to share some geeky details.

In any case, to me what is more important than waiting for the IMU is waiting for GPS to have a good position before starting the motors. By the time that happens the IMU is long past warmed up.

 

[Big Fil]

In any case, to me what is more important than waiting for the IMU is waiting for GPS to have a good position before starting the motors. By the time that happens the IMU is long past warmed up.

I'd disagree. The aircraft can flown without a GPS lock. The aircraft will basically not fly without the IMU working.

 

[AlanTheBeast]

I'd disagree. The aircraft can flown without a GPS lock. The aircraft will basically not fly without the IMU working.

It's not about what can be done but about what should be done.

 

[JHSlayer]

For what it's worth, if you start up everything and do your preflight check list, the Mavic will be ready. 3 to 5 minutes to make sure you are ready shouldn't hurt you.

 

[AlanTheBeast]

For what it's worth, if you start up everything and do your preflight check list, the Mavic will be ready. 3 to 5 minutes to make sure you are ready shouldn't hurt you.

And will likely result in a high intact drone over time ratio as well.

 

[JHSlayer]

I think that's a compliment? Thanks if so. [emoji6]

 

[JHSlayer]

This can be a hard crowd to read...

 

[JHSlayer]

This can be a hard crowd to read...

I'm still getting over someone's April Fools joke. His name was...
Oh nevermind.[emoji12]

 

[AlanTheBeast]

I think that's a compliment? Thanks if so. [emoji6]

Threw a like at your post to erase all doubt! No fooling.

 

[JHSlayer]

Threw a like at your post to erase all doubt! No fooling.

[emoji6]

 

[jimheid]

For what it's worth, if you start up everything and do your preflight check list, the Mavic will be ready. 3 to 5 minutes to make sure you are ready shouldn't hurt you.

Very much agreed! I like to say that when I'm on the ground, I've got gravity exactly where I want it. But as soon as I take off, that equation flips. I'm very willing to take a few minutes to make sure I'm ready to go up against gravity.

 

[dewster]

Who says an IMU has to warm up?

The IMU. lol

 

[SRJ_Mavic_PRO]

I'm surprised nobody answered.... Because it was cold.....