High altitude - near space use of mosfets and IGBTs. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

High altitude - near space use of mosfets and IGBTs.

1 2 3 next

Move Thread

LAN_403

Patrick

Sun Sept 18 2016, 06:30AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

I volunteer for a company that does high altitude rocket and balloon research. being on the electrical / avionics / ground station side, i was not surprised that there would be all kinds of valuable experience they have gained across 30 years. But, they use almost entirely mechanical relays for everything of medium power.

A typical example would be a hot nichrome wire that we use to burst the Balloon or stage other actions. The micro-controller turns a pin high, which drives a TO-92 NPN transistor which drives the coil of a relay, which then the relay transmits full battery power to the wire.

but this seems convoluted. I suggested using a single transistor (avoiding moving parts, which I despise), but they claim to have used them, finding a 25% reliability rate above 30,000 - 100,000 feet. Temperature was my first thought, but they seem to think its the near vacuum. yet IGBTs and mosfets should be fully void-less plastic or similar material.

Are there any ideas we've missed ?

As for temperature, Gopros seem to work, as do basic stamps and other common hobby electronics at 105,000 feet. Its interesting. but odd.

Carbon_Rod

Sun Sept 18 2016, 07:27AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Most near earth orbital electronics have relatively high reliability, and often include regular consumer grade rated parts.
Note a relay can still work at >250'C or <-20'C... your go-pro will not...

Mission critical systems are a different subject area, and there are satellite circuit designs which use fault tolerant topologies.
These designs may no longer be popular, but describe how to get something to last 20+ years in space.

Mechanical linkages act differently under high impulse...
Many engineers do not make a distinction between stiffness and strength of parts.

Mads Barnkob

Sun Sept 18 2016, 09:03AM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

EEVblog did two videos with a design engineer doing space electronics, there might be some valuable information regarding temperature and radiation in these two videos:

and

and there is also a third bit with some he forgot in the edit of the first two:

Patrick

Sun Sept 18 2016, 06:54PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Its shocking that for a gentle balloon ride to 105,000 feet in 45 minute return to earth cycle ,that a mosfet or igbt fails !?

Yet all of our go pros, basic stamps and other electronics work almost all the time.

johnf

Sun Sept 18 2016, 07:47PM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

With a temperature drop of three degrees per 1000 feet of altitude you will now see why your igbt / transistors fail. I presume the failures are soft ie the fault dissapears when back at ground level. It is quite tricky designing for a wide temperature range as bias conditions change quite quickly ie as it gets colder you need to increase your gate voltage above the datasheet max so as to keep it hard switching. As for the gopro it keeps itself warm and if its in its outdoor enclosure it is well protected

Patrick

Sun Sept 18 2016, 08:05PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

we do keep things insulated, the npn transistors which drive the relays work just fine. And temp was my first guess, but they all think its vaccuum.

Eventually for their more advanced flights relays wont work, so im thinking of an insulated styrofoam "enviromental box" with heat sinks and PIDs adding heat / losing heat. the Aluminum heatsinks would be silver epoxied to each IC and transistor body and all the aluminum connected together kept at constant temp via the PID/resistor.. with only 1% or less air at our altitude, convection aint gonna work.

Carbon_Rod

Sun Sept 18 2016, 08:37PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Indeed, many satellites use a pressurized bay with a circulatory fan, and a near-deflated balloon on a switch to detect leaks.

A simple thermal cut-out switch with a Polyimide film insulated flexible heater pad is far more reliable.

Mass is the number one minimization problem...

Patrick

Sun Sept 18 2016, 08:49PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Carbon_Rod wrote ...

A simple thermal cut-out switch with a Polyimide film insulated flexible heater pad is far more reliable.

Elaborate please i dont understand.

But heres some pics:

Nik

Sun Sept 18 2016, 08:52PM

Registered Member #53 Joined: Thu Feb 09 2006, 04:31AM
Location: Ontario, Canada
Posts: 638

You might try sending up an experiment in the future that tests a bunch of transistors with various kinds of protection (hot glue, silicone, epoxy, etc). Have them switch every couple of seconds and see at what altitude they fail and see if you can make a correlation between the conditions and the mode of failure.

If nothing else it means you get more data to play with!

Patrick

Sun Sept 18 2016, 10:22PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Nik wrote ...

You might try sending up an experiment in the future that tests a bunch of transistors with various kinds of protection (hot glue, silicone, epoxy, etc). Have them switch every couple of seconds and see at what altitude they fail and see if you can make a correlation between the conditions and the mode of failure.

If nothing else it means you get more data to play with!

this was exactly my plan, :) i just need to beg for about 250 grams on the "high-rack flights" they go the highest for the longest.

I was on the left side fill team, we went through $5,000 of helium. You can see me on the cylinder waiting for my Dr. Strangelove moment. it took about 2 hours to fill.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob