Parallax Forums

Parallax Forums > Public Forums > Propeller Chip > Extreme cold operations

Forum Quick Jump

[ << Previous Thread | Next Thread >> ] | Show Newest Post First ]

Earl Foster
Registered Member

Date Joined Nov 2004
Total Posts : 171

Posted 3/10/2009 6:59 PM (GMT -8)

I am designing a new long duration high altitude payload that will be powered by the Prop. All the components MUST be rated for -40C (40F), -55C (-67F) where I can get them.

So far I have been able to find all components within my temperature range except one. I am having a problem finding a 5MHz 20pF crystal rated for that low of a temperature. I have found a 5MHz 18pF crystal rated at -40C in a metal can SMD (Mouser:695-HCM49-5-U) but I do not know how critical the 20pF is for Prop. Plus … I would like to stay away from the SMD devices for this payload if possible.

So it boils down to 2 questions:

Does anyone know where I can get a NON SMD 5MHz 20pF crystal rated for an operating temperature of -40C?
If I use the SMD I found rated @ 18pF will this have an impact on the Prop performance?

Thanks

WWW.HAPB.NET

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

Tracy Allen
Registered Member

Date Joined Jul 2004
Total Posts : 2847

Posted 3/10/2009 8:21 PM (GMT -8)

The Citizen CMR309T5.000MABJ (Digikey has stock) is rated for operation from -40 to +85, and while it is ostensibly also surface mount, it is a cylindrical package with wire leads that are simply bent out to sit on SMT pads. So you could conceivably bend those down into holes.

I don't understand however, why basically the same package can have two different operating temperature ranges. For example, the HCM49 is rated for -40 degrees in surface mount, but the same package in through hole, the HC49, is rated only to -20 degrees. AFAIK, the surface mount HCM49 is made by bending out the wire leads of the HC49 along a pad a the bottom, so they come out at the ends for attachment to SMT pads. Otherwise the same crystal, same can. Maybe it has to do with better control over the stresses of mounting. Huh?

Tracy Allen
www.emesystems.com

SRLM
Registered Member

Date Joined Jul 2008
Total Posts : 2720

Posted 3/10/2009 9:25 PM (GMT -8)

You could probably just use a regular crystal with a small heater and temperature sensor. That way, it will keep the crystal temperature constant (PID loop in a prop cog), giving you better timekeeping accuracy. It shouldn't be too difficult to surround the crystal with a couple of heating elements, the temperature sensor, and some thermally conductive epoxy.

Jay Kickliter
Registered Member

Date Joined Aug 2008
Total Posts : 242

Posted 3/11/2009 7:07 AM (GMT -8)

Aren't those temperature ratings at sea level? At the altitude you'd be using these components, would they even loose heat fast enough to be a concern? I've been wondering about this for a while. On your past flights, did you ever take board level temperature measurments?

Earl Foster
Registered Member

Date Joined Nov 2004
Total Posts : 171

Posted 3/11/2009 7:28 AM (GMT -8)

I found it interesting that the HCM49 in surface mount is rated lower then the non-SMD version as well. I would agree with you and say it is because of the leads /no leads can package. I imagine that the 2pF will not have an impact on the performance and it is standard for SMD due to the lack of leads.

Using temperature controlled heaters is a good idea but since the crystal is the only component I am having a problem with I would prefer to stay away from that solution right now.

Failure due to under rated components is a major concern. For my normal payloads I do not worry about the rated temperatures since I know my housing will stay warm during the 4 hour mission and I record temperature, however, this is not a normal or amateur mission. The prop will be controlling balloon float at altitude, along with other critical functions for a period of days. The temperature at 110,000 is -55C so the payload will get cold soaked over time.

WWW.HAPB.NET

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

Post Edited (Earl Foster) : 3/11/2009 3:45:01 PM GMT

Phil Pilgrim (PhiPi)
Registered Member

Date Joined Feb 2006
Total Posts : 6000

Posted 3/11/2009 8:18 AM (GMT -8)

You might also consider driving the XTI pin with an external oscillator. This one, for example, is through-hole, runs at 5MHz on 3.3V, and operates down to -40C. Here's another.

If you decide that SMD is okay, here's one that runs at 80MHz on 3.3V (precluding necessity for the PLL), and operates down to -55C. Here's another with the same specs.

BTW, is your concern about SMD due to a difference in tempco between package and PCB, or because of other factors?

-Phil

Post Edited (Phil Pilgrim (PhiPi)) : 3/11/2009 4:25:18 PM GMT

Tracy Allen
Registered Member

Date Joined Jul 2004
Total Posts : 2847

Posted 3/11/2009 8:18 AM (GMT -8)

Jay, I don't think the temperature ratings have to do with altitude or pressure, but more with frequency drift. The graph here from Citizen shows the temperature effect on an AT cut crystal from -80 to +120 degrees Celsius.
http://www.citizen.co.jp/english/crystal/aspect/index.html
The spec sheets say that their crystals (in certain SMT packages) will stay within 50ppm of the calibration frequency over a range of -10 to +60 C, but the operating temperature range is -40 to +85, and storage from -55 to +125. When I first got a Propeller, I stuck it in an environmental chamber and had it operating down to -65 degrees C using the standard HC49 crystal. Another thing, I don't ever recall seeing a graph of crystal impedance versus temperature, in terms of Q factor that would need to be accounted for in the oscillation circuit. It may not be a factor.

Maybe there is some limitation due to the packaging. In a project for a deep ocean probe that used one of our data loggers, they insisted that we take out the resistance welded metal resonator package and use an epoxy package instead, due to the high pressures that would be encountered.

Earl, the 18pF vs 20pF shouldn't make a significant difference.

Tracy Allen
www.emesystems.com

Earl Foster
Registered Member

Date Joined Nov 2004
Total Posts : 171

Posted 3/11/2009 12:29 PM (GMT -8)

I really have not gotten into soldering SM devices at this point so it is just a comfort factor why I prefer through hole devices. Guess its time to start. Thanks

WWW.HAPB.NET

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

mctrivia
Registered Member

Date Joined Jan 2009
Total Posts : 2540

Posted 3/11/2009 12:36 PM (GMT -8)

surface mount can be really easy if you have steady hands, and good iron. I just picked up a toaster oven to play more with bga

Need to make your prop design easier or secure? Get a PropMod has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Phil Pilgrim (PhiPi)
Registered Member

Date Joined Feb 2006
Total Posts : 6000

Posted 3/11/2009 1:15 PM (GMT -8)

You do have to mind the temperature coefficient issues at low temperatures, though. Packages with compliant pins (e.g. SOIC) are not a problem, since the pins can flex when the package expands/contracts differentially with the PCB. But packages with no pins (i.e. just pads on the bottom) could be a problem under those conditions, since solder joints could crack. This is exacerbated at low temps by the fact that, under reflow conditions, the solder solidified when package and board were both hot.

I honestly don't know whether the tempcos between crystals/oscillators and fiberglass PCB materials are different enough to be an issue, though. Maybe they're not.

-Phil

Earl Foster
Registered Member

Date Joined Nov 2004
Total Posts : 171

Posted 3/11/2009 1:25 PM (GMT -8)

Thanks Phil, thats something I will keep in mind.

WWW.HAPB.NET

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

Tracy Allen
Registered Member

Date Joined Jul 2004
Total Posts : 2847

Posted 3/12/2009 9:47 AM (GMT -8)

As SMT goes, the CMR309 tubular package and the HCM49 have long compliant leads as Phil described and should be no problem at all to solder to a through hole board.

Since this is a critical mission, it seems that pre-flight testing of the whole Propeller system with crystal and circuit board is in order. Even when individual components are cold rated, the whole is more than the sum of the parts. If you don't have an environmental chamber that can reach down to -40 or below, a cooler and dry ice can be a go-no go test chamber, albeit harsh and uncontrolled. For reasons of supply and cost the space program sometimes "uprates" commercial components for use way beyond their specified temperature/humidity/radiation ratings, but they have a strict regimen for testing and handling and packaging them. Harsh testing is a double edged sword, as it can exacerbate the micro-cracks etc. that lead to ultimate device failure.

I'm curious, at that 20 mile altitude, do you expect high temperatures also due to solar radiation?

Tracy Allen
www.emesystems.com

CannibalRobotics
Registered Member

Date Joined Feb 2008
Total Posts : 235

Posted 3/12/2009 10:07 AM (GMT -8)

I would think that at altitude you get into atmospheric density issues and cooling becomes more difficult. Heat dissipation may become the problem to your solution.
What if you put your crystal right next to your voltage regulator - maybe even solder the heat sync tab to the can of the crystal. Epoxy a thermistor to the crystal can and see what happens.
Jim-

Signature space for rent, only $1.
Send cash and signature to CannibalRobotics.

Earl Foster
Registered Member

Date Joined Nov 2004
Total Posts : 171

Posted 3/12/2009 11:26 AM (GMT -8)

Tracy ... The whole system will be tested in an environmental chamber that is designed to go below -55C. Part of testing and integration will involve multiple small payload tests performed for several hours and then days at altitude before being incorporated into the real payloads.

The payload will be covered with the same material that is used on satellites to reflect solar radiation. Since I have never done a long duration flight before I do not know how much additional heating, if any, is expected. I was given the temperature rating specifications that I must follow in order to comply with mission parameters.

Jim ... that is a possibility. Since I have never been involved in a long duration mission this is new territory for me.

WWW.HAPB.NET

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

Jetfire
Registered Member

Date Joined Dec 2008
Total Posts : 34

Posted 3/12/2009 11:41 AM (GMT -8)

Mouser has this 4MHz crystal

Through hole, 20pf, and a frequency stability of +- 100 over the operating range.

Also this 6MHz crystal.

virtuPIC
Registered Member

Date Joined Jan 2009
Total Posts : 189

Posted 3/12/2009 12:21 PM (GMT -8)

High altitude also means high radiation. Ionization might cause shortcuts in capacitors and MOSFET. I don't know current status but in former times avionics had no DRAM and programs have been stored in fused PROM. Atmel produces special hardware for extraterrestrial use.

In brief: Also check radiation tolerance of the prop and associated components. Especially Flash memory, RAM, and CMOS circuits.

Airspace V - international hangar flying!
www.airspace-v.com/ggadgets for tools & toys

[ << Previous Thread | Next Thread >> ]

Forum Information

Currently it is Saturday, November 21, 2009 11:43 AM (GMT -8)
There are a total of 393,863 posts in 55,536 threads.
In the last 3 days there were 84 new threads and 713 reply posts. View Active Threads