Hmmm..that is exraordinary.. I've had batteries go open and short, but not both...then again I've never kept a shorted one (that I could not recover) for 20 years either.
With solar panels, supercapacitors and a bit of intelligent circuitry, it should be possible to give dodgy NiCds a "zap" of high current charge if it looks bad when the sat comes out of eclipse.
Would it be possible to work with supercaps alone - I would emagine the eclipse periods are not very long? I'd think supercaps have a better lifespan than the best batteries?
I would imagine that Li-ion was out because the satellite had to be stored inside the ISS alongside the astronauts for several months. So any battery technology capable of leaking, outgassing or bursting into flames is certainly a no no. Although I do see lots of Thinkpads in photos of the ISS... Do they have Li-ion battery packs?
This particular battery was designed with high reliability in mind, not longevity. The number of charges and discharges are carefully monitored. After the specified number are met, the battery is disposed of.
Hi Sanjib, it is true that FPGA manufactures make radiation hardened components for the aerospace industry. They are very expensive for an amateur project. Thus we chose to go the cheap and easy route. I do believe that we shall see the proof of the pudding (as it were) with ARISSat-1. Letís see how long it operates with industrial temperature grade components.
On the question of temperature specification, simulations were telling us the range of -20 to +40 degrees C. However, we are seeing in the telemetry higher temperatures. We are still investigating why that is.
Hi David, I only mention NiCad because past amateur satellites have flown them with good results. In other words, itís something we know. But we would like to explore other chemistries and charging management methods to both learn and expand our level of knowledge Ė within the safety boundaries we are given.
If you look at past amateur satellites and their battery track records, the life of each most often is dictated by the battery. As you mentioned, they like to short which renders the satellite power system inoperable. If we could have a smart system that could remove the short, at least we can continue to operate in the Sun; my like ARISSat-1 is doing now. We were lucky; it appears that the battery failed open.
For an extraordinary story, take a look at the life of OSCAR 7. It was rendered dead for several years because of a shorted battery. Then one day it was discovered operating in sunlight. http://en.wikipedia.org/wiki/Oscar_7
Hi Steve, thanks a lot for the information. While browsing through the parts e.g FPGAs, controller, I've seen some manufacturer sells "radiation hardened" chips. Do you really need them? They are usually sold at a premium. Also, what is the environmental spec. (temperature etc.) for which you design your electronics?
Thanks Steve. Interesting you mention NiCds, I would have thought you'd go for something like Li-ion? I'm quite a fan of NiCds,but it's mainly as I have access to lots of them. If you look after them they are good batteries...but they do have an annoying habit of going permanently short circuit if they are left discharged for any time....which, as with your current battery's problems, would be disastrous. Any reason for choosing NiCd?
Bummer about the battery!! It seems that (in hindsight) a backup/redundant storage system would have been in order. Is there any consideration for a low power down standby/hibernate mode for the next satellite? This coupled with a small redundant power backup could be used to enable the electronics to not get reset during short periods of eclipse. Thanks for the details! A classic case of "this is what we have to work with" and making due. Keep up the good work.
My Mom the Radio Star Max MaxfieldPost a comment I've said it before and I'll say it again -- it's a funny old world when you come to think about it. Last Friday lunchtime, for example, I received an email from Tim Levell, the editor for ...
A Book For All Reasons Bernard Cole2 comments Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...