Inductorless DC-DC converters for space application

[sophiecentaur]

It comes from the fact that current flowed around the circuit, through the inductor. In this case, the 'charge conservation' law doesn't apply - it doesn't need to because it's not a total charge involved but an 'imbalance' of charge. The only respect in which the conservation of charge applied is in the fact that the whole circuit starts off neutral and ends neutral.
(I must say, you had me there for a moment )
What is conserved, however, is Energy - because (ignoring resistance) none is dissipated.

 

[sophiecentaur]

And why is everyone so inordinately chuffed about super capacitors? They have their uses and are eminently suited to many applications but their one huge disadvantage is surely that their voltage is so variable. Not so much of a problem is the circuits they feed have been designed to cope with it but it makes a lot of designs a lot more complex. It seems to me that batteries are doing pretty well at the moment - with increasing capacities and fast charging times - for most applications.

 

[skeptic2]

I'm sorry but your explanation is a little vague. I wonder if you could either go into more detail or point me to a reference. I have searched for and looked at references but haven't found anything that addresses this point.

It seems to me that it must be the charge that is conserved as it is in the traditional two capacitor problem. That problem is a paradox only when zero resistance is assumed. When resistance is added, the lost energy is clearly dissipated in the resistance. What happens in your example when a small amount of resistance is added? On the other hand, charge represents a quantity of electrons, something that to me would be a great deal harder to create or lose.

 

[sophiecentaur]

VAGUE ?!
The cheek of it. (I thought it was more than adequate and not too bad as I was making it up as I went along)
Nevertheless, it seems right enough - helped by the fact that the answer produced the conservation of energy - quite a good case for the prosecution M'lud.

In the 'two capacitor problem' there is no 'alternative' current path to the wires joining the two capacitors. What is being shared is the imbalanced charge as no charge can flow anywhere else. I would agree that, as no 'discharging' can take place then the imbalanced charge must not change: a kind of charge conservation, if you like but not in the 'particle Physics' sense of the phrase.
Connecting an inductor across the terminals of C1 allows the charge on C1 to be any value from +CV to -CV and back up to +CV, depending on which part of the oscillation cycle you disconnect the L. You would agree that is what happens in an LC circuit? Where is your requirement for 'conservation of charge' during that process? Energy is not lost, of course, because you have a current flowing in the inductor so the Electrical Potential Energy is transferred to Magnetic Energy.
I repeat - no electrons are ever lost or gained by the circuit; they are just re-arranged about it according to the permitted rules. There may be more or fewer excess electrons on one plate but that number is totally balanced by a change in the number of electrons on the other plate.

 

[mheslep]

A problem with using any kind of a switching circuit as a solution here is that for a given inductance the B field is proportional to the current I, and current pulses/spikes intrinsic to a switcher are many multiples of the steady state current through a linear converter. And, it is not just the switching magnetics that have inductance: capacitor-diode-resistor leads, circuit traces, all of it does; therefore everything that conducts switching currents may be radiating. The switching currents may also couple into everything else connected to the battery power supply - an undesirable effect that can be attenuated many orders of magnitude by prudent design but perhaps not sufficiently in this case.

Therefore I'd be disinclined to use a switcher unless the application measurements can be made relatively quickly while the switcher is actively disabled/blanked as berkeman suggested earlier:

Do you briefly shut off the DC-DC to make the measurement quickly, while capacitors hold the 3V3 and other rails to minimum droop? If so, do you even need a shield at all?

I've seen disable / blanking circuits like that used before; you can generally oversize your output capacitor to holdup for whatever time you require. I suspect it will cost you little to design in blanking, and if later during integration you find your magnetometer is not impacted by your converter running continuously, you can just forget about syncing up the blanking circuit.

 

[sophiecentaur]

Doesn't this all boil down to the range of signal frequencies in which you are interested and the frequency range of interfering signals? If you are looking for a low frequency / DC signal, then you would filter accordingly and I can't see how the HF switching waveform could not be filtered out (I am, of course, referring to analogue filtering - which is not always how it's done these days). Turning your charging circuit on and off could actually be introducing components into your (now modulated) signal which could be as much of a worry as the presence of high frequency switching components, which may well be a long way out of band and, hence, easily filtered.
It would have to depend on the actual requirements.
I just had another thought. What is the likely level of other interference sources on the satellite? There will be a specified maximum and you can be sure that one of the passengers will be producing something near that value. That could be far more of worry than you own PSU.

 

[mheslep]

It's not clear that anything is 'easily filtered' with the usual discrete component filters relative to the sensitivity of these instruments - 0.015 uT.

 

[sophiecentaur]

"Easily filtered". OK - a bit glib there, perhaps.

But your situation is by no means unique and there are many good engineering solutions to the problem of producing suitable 'clean' supplies of power. I can see how a chemical cell, unconnected to any charging circuit, has its attractions. However, if your signal are really as embarrassingly small as you think, then you are in the realms of SNR - or S/Interference Ratio. You quote sensitivity in T but, also the frequency response is highly relevant here. If the magnetometers are low frequency devices then why would you expect significant response at 100kHz or more (you could choose a much higher switching frequency if you wanted)?
If your numbers regarding bandwidths, linearity, sensitivity etc. all lead you to conclude that a high frequency DC - DC converter is not viable then fair enough. However, the low frequency ('on/off') solution may involve other problems. What other sources of RFI are likely to be on the satellite or is it all your own territory, after it's been launched?

This autonomous satellite system needs a lot of expertise poured into it if it is to be successful and, presumably, you are getting ideas from other directions than this forum. I have been involved with loads of projects in which there have been aspects that weren't adequately addressed, initially. That's why I am making these comments. I hope you have a lot more to draw from than the few interested contributors to this thread - who's provenance (including mine) is unknown. It does bother me that so many posts refer to Wikipedia as the ultimate fount of knowledge. It's a great start but, for a very expensive project like this one, you shouldn't risk relying on any of it, on its own. This may be at the cutting edge of Engineering and you can't leave anything to chance.

 

[p1ayaone1]

This autonomous satellite system needs a lot of expertise poured into it if it is to be successful and, presumably, you are getting ideas from other directions than this forum. I have been involved with loads of projects in which there have been aspects that weren't adequately addressed, initially. That's why I am making these comments.

...

This may be at the cutting edge of Engineering and you can't leave anything to chance.

You are correct to the nth degree. This is a pilot project at my school, 6 students plus an experienced space engineer as a supervisor. If we can come up with a viable design, it will be passed to the next generation who will use it as a baseline for Concordia's entry in the Canadian Satellite Design Challenge (http://www.geocentrix.ca/index.php?option=com_content&view=article&id=2&Itemid=2 ).

So far we've had a little something for all walks of engineering. Mechanical will need to come up with a viable chassis that meets tight specification (0.1mm), thermal performance and that will not fall apart under intense vibration. (They call the environmental test "Shake 'n Bake").

Parts of the design include H-Bridges implemented with BJTs (for all the analog junkies out there) controlled by a combinational logic circuit (ibid. for digital).

Of course the communications system is its own little bag of monkeys - how much BW can we get, what modulation is most efficient and most reliable, how are the data packets constructed, how many channels, what bands give enough range but don't sap up the entire power budget, and so on.

Programming? That'll be fun. Ground station application with 2D and 3D visualization? Absolutely! A CMOS camera connected to an Atmega? All in a day's work.

And as all readers of this thread will know, there is the famous power system. The design is yet to be approved, but it looks like we'll go with the BQ2057 as a battery charger, the TPS6050 for the 3V3 bus, and the TPS6013 for the 5V bus (all from the wonderful folks at Texas Instruments' free samples program). The MO has been simply to find inductorless circuits whose input and output voltages are within margin, and whose current throughputs are high enough.

Will the design be perfect? No. But we will be within scope (and budget hopefully). And it turns out that this type of design is totally ADDICTIVE! I'd love to share more of this experience with you PFers over the next 5~6 months if you are interested.

Next up is sourcing parts, breadboarding, and code-writing, and after that is PCB and mechanical assembly.

Attached to this post is the power subsystem schematic, which in a way you have all designed.

 

[sophiecentaur]

Sounds a fantastic project to be involved in. I wish you the very best of luck.

 

[skeptic2]

Will the design be perfect? No. But we will be within scope (and budget hopefully). And it turns out that this type of design is totally ADDICTIVE! I'd love to share more of this experience with you PFers over the next 5~6 months if you are interested.

Are you kidding? We're here because we're addicted too. Of course we'd like to hear about your design as it is developed and particularly the reasons behind the decisions you make. For instance, what were your reasons behind deciding to go with the BQ2057 charger? It is a linear instead of a switched charger isn't it? The temperature sensor was a particularly good decision for something designed to operate in space.

 

[p1ayaone1]

The temperature sensor was a particularly good decision for something designed to operate in space.

It does have that nice feature, but it's disabled (the TS (temp-sensing) pin has been set to Vcc/2).

The reason is this: Ok, the battery is hot. Now what? Switch it off, and the whole satellite dies. Leave it on, and the battery burns out, and the whole satellite dies.

There is also a possibility that things get really hot because of the sun, not the battery. I wouldn't want a false-positive to switch off the battery charger.

There is a safety feature not shown on this schematic: there are several analog temperature sensors strategically located throughout the satellite. There is one on the battery, one each on the Tx and Rx, and the rest on the chassis walls. The MCU will poll those sensors, and if, say, the battery gets hot, the entire satellite will go into "sleep" mode (stop drawing current) while things even out.

 

[gnurf]

You said something about a Cubesat in an earlier post. Assuming the five rows of four 6.7V sources represent the solar cells on five sides of a cube, what happens when one or more of those rows (i.e. sides of a cube) generate no power because they receive no sunlight while others do?

Solar cell equivalent circuit: no irradiance ==> no current from the source, and you're left with a small resistance in series with a diode to ground (ignoring the shunt resistor).

Also, with your current solar cell configuration, if a single cell fails short (I'm not sure if a short-circuit is a common failure mode for a solar cell, but still..) your entire solar array dies.

The solution to both problems is to use protection diodes, e.g. one for each side.

 

[p1ayaone1]

Your assumption (5 sides x 4 modules) is correct, I will update the schematic to make it more clear.

Originally I had a large number of 0.5V cells in series, and I had bypassed those with diodes. I though that non-illuminated PV cells act like open circuits.

The design was modified to parallel connections, since each module (consisting of several cells) supplies enough voltage to drive the charging circuit and regulators. Given the initial assumption that dark cells equal open circuits, it was an easy to connect every cell in parallel.

If in fact that diode exists in the EC, each module would need a series diode in its branch on the parallel circuit, otherwise the dark cells would short out the whole circuit.

I will study the EC of the PV cells further, and if this is all true, I will simply add a diode to each module, allowing current only to flow out. Shottkey might be wise since reverse voltage won't be too high, and forward voltage drop is minimal.

This being a cube shape, it is obviously impossible that all cells are illuminated at the same time. The design does need to handle some cells generating no power. The power budget (i.e. required average and instantaneous current) takes this into account.

 

[gnurf]

It does have that nice feature, but it's disabled (the TS (temp-sensing) pin has been set to Vcc/2).

The reason is this: Ok, the battery is hot. Now what? Switch it off, and the whole satellite dies. Leave it on, and the battery burns out, and the whole satellite dies.

I don't get this. If the battery charger senses an over-temperature condition, won't it simply stop charging? What do you mean when you way the whole satellite dies? Temp out of bounds --> stop charging --> run satellite on solar cells and/or discharging batteries. No?

There is also a possibility that things get really hot because of the sun, not the battery. I wouldn't want a false-positive to switch off the battery charger.

Some liion cells have internal temp sensors which you can access via a third (or forth) terminal on the battery. Regardless, you must of course make sure that your temperature sensor returns the information you intended (the bat temp), so it's up to you to place the sensor in such a way that it does in fact measure the battery temp as accurately as possible (sorry if that was obvious).

I actually think the biggest problem for your battery might not be heat, but the very cold temperatures it will be exposed to during the typical half-hour eclipse in a low-earth orbit. Throw your liion cells in the freezer and see how they react (i.e. compare the discharge curves in different temperatures). I see now that both your regulators have pretty good head-room relative to a potential detrimental temperature effect on the battery voltage, so this might not be an issue.

 

[p1ayaone1]

https://www.physicsforums.com/showthread.php?t=345524&highlight=lithium-ion

A quick read over this thread seems to indicate that charging LiIon in extreme cold might be dangerous, but I suppose that's not a problem.

I could of course throw the circuit in the freezer, but maybe I'll just wait till February and leave it outside overnight. Not quite space-cold, but nearly. I'm sure I'll get a few nights of -20C soon enough.

 

[skeptic2]

Can your experienced space engineer give you an idea of how cold the satellite is likely to get. Nearly half of the satellite's sky will be filled by the Earth at an average temperature of around 12 C, plus the fact that the satellite will be generating some heat, so it seems to me it will take a long time, much longer than 1/2 hour, for the satellite to cool off.

 

[p1ayaone1]

Nearly half of the satellite's sky will be filled by the Earth at an average temperature of around 12 C

Sorry, I don't understand. What is 12 C?

The actual temp of the satellite will depend on how much solar radiation is absorbed/reflected. There is no ambient temperature in space afaik.

 

[p1ayaone1]

Thank's not to say the temp is 0K, but temperature is a property of matter, of which there is very little floating around at ~800km.

 

[skeptic2]

Sorry, I don't understand. What is 12 C?

The actual temp of the satellite will depend on how much solar radiation is absorbed/reflected. There is no ambient temperature in space afaik.

I should have said 12 degrees Celsius. Agreed, without matter it's difficult to get rid of heat. It can be said that the temperature of space is average temperature of the radiation from all sides. Since the Earth is about 55 deg Fahrenheit, and if the satellite were not rotating, the side facing the Earth would not cool down much below that. The other side of the satellite, though receiving very little thermal radiation, also would not be able to radiate its own heat very rapidly.

 

[p1ayaone1]

I got that 12 C meant 12 deg celsius, I meant what body has that temp? The Earth? So what you're saying is that the side of the satellite facing the Earth will have the same temperature as it?

If I've understood, then the side of the satellite facing the Sun will have the same temperature as the Sun's surface? Seems a bit hot?

I've been told that while there will be no convective heat dissipation, there will still be conductive and radiative. I.e. the aluminum chassis should sink some heat, as well as IR emitted by the hot satellite. Any need to put a CPU-style passive heatsink, or do those work purely on convective principle?

And then the side(s) of the satellite facing empty space will simply sink heat from the other sides?

Why oh why did I never take a course in thermodynamics?

 

[skeptic2]

I got that 12 C meant 12 deg celsius, I meant what body has that temp? The Earth? So what you're saying is that the side of the satellite facing the Earth will have the same temperature as it?

Since the Earth covers nearly 180 deg of the sky, if the satellite were to get cooler than the earth, would it not receive more heat from the Earth than it is able to radiate and so warm back up. It may still conduct some of its heat to the other side of the satellite away from the earth.

If I've understood, then the side of the satellite facing the Sun will have the same temperature as the Sun's surface? Seems a bit hot?

Yes, except that instead of covering half the sky, the sun is only 0.5 degrees in diameter. In all directions other than that of the sun, the satellite would lose heat (assuming that while on the sunny side of the earth, the satellite is hotter than the earth).

I've been told that while there will be no convective heat dissipation, there will still be conductive and radiative. I.e. the aluminum chassis should sink some heat, as well as IR emitted by the hot satellite. Any need to put a CPU-style passive heatsink, or do those work purely on convective principle?

I was working at Motorola when they developed the first flip phone. Though I didn't personally work on that phone, the story I heard was that they had a serious heat dissipation problem and used a somewhat unconventional solution. Instead of a fiberglass circuit board, they used solid copper. Onto the copper they silkscreened a number of layers. The first layer was an insulating layer with holes for ground connections. The next layer was for circuit traces and wherever a ground was needed, that trace was routed over a hole in the insulating layer. That layer was covered by another insulating layer again with holes left for interlayer connections. The surface mount components were added last and soldered. The copper was able to conduct the heat away much better than fiberglass making the phone possible.

And then the side(s) of the satellite facing empty space will simply sink heat from the other sides?

...and radiate it away, but not very efficiently.