Satellite MPPT solar circuit design

I'm in the process of developing a pocketqube and I am currently working on the power system.
I am hoping to use a switching regulator such like the following (https://www.digikey.es/product-detail/es/texas-instruments/TPS62120DCNR/296-27815-1-ND/2407328) and an SPV1040 MPPT IC to simplify the design and maximize efficiency.
Each side of the craft will have 8 Trisolx GaAs solar cells with an open voltage of 2.6 and Voltage (max power) of 2.3, they have a max current of 14.6ma. Obviously all these characteristics are in a direct sunlight ideal scenario at ambient temperature. The battery charged will be a 700 to 900 mah 3.7v Li-Ion and the output voltage needed is 3.3v.

On another note I would also like to regulate 3.3v from 2 or 3 of those solar cells, the output power need though will be minimal at around 80mw, just a direct regulation without a battery.

I have some basic circuit design experience but would just like to make sure I'm not making any errors, any sort of existing schematics would really help.

Thanks for your help and support,
Julian

Julianfer5:
Each side of the craft will have 8 Trisolx GaAs solar cells with an open voltage of 2.6 and Voltage (max power) of 2.3, they have a max current of 14.6ma. Obviously all these characteristics are in a direct sunlight ideal scenario at ambient temperature. The battery charged will be a 700 to 900 mah 3.7v Li-Ion and the output voltage needed is 3.3v.

Working with such ridiculously low voltages is going to be tricky; I cannot readily comment on it.

One interesting aspect of a spacecraft is that sunlight is essentially binary - either full "direct sunlight" or none. Orientation is really the only variable. So you will have six panels?

Paul__B:
Working with such ridiculously low voltages is going to be tricky; I cannot readily comment on it.

One interesting aspect of a spacecraft is that sunlight is essentially binary - either full "direct sunlight" or none. Orientation is really the only variable. So you will have six panels?

Ah ok, I have managed to run the satellite using linear voltage regulator but I am trying to improve efficiency, Its definitely possible though. In reality I will only have 5 panels of 8 cells since 1 panel will be taken up by deployable doors that will launch other satellite.

Thanks!

The satellite will be passively stabilized so we can easily predict its orientation on each stage of the orbit regarding earth's magnetic field.

Deployable cells are an option but I really want to keep it simple and avoid possible failure points.

The IC I quoted has an input voltage of 0.3 - 5.5v and the satellite automatically adjusts for the amount of power it has available.

From what I can see everything falls under spec and should work, just need to figure a schematic since I cant figure where to put caps and resistors etc.

Switching regulators like the TPS62120 are very sensitive to PCB layout. If you are having difficulty with the schematic then you have no hope to get the PCB right on your own.

If you are certain that you need a switching regulator then try working down this decision tree:

  1. Can you buy the regulator as a module on a PCB with all the required components already soldered? Pololu make some good ones.

  2. Can you buy the PCB design as a module to be pasted onto your PCB design? (This is probably going to blow your budget.)

  3. Can you find a free design for the PCB layout with all the components fully specified with part numbers? Some datasheets will show this level of detail or you may find an application note from the chip manufacturer.

  4. Can you afford the time to learn PCB layout and make and test 5 different versions? You are also going to have to learn to use circuit simulation software such as LTSpice.

  5. Do you have a friend who is an expert in designing switching power supplies? The first thing he will tell you is that you chose the wrong chip to begin with and then you find that his recommended chip has a good module that you can buy on ebay for a few bucks. Go back to step 1.

Are you following any of the schematics or the layout diagram in the datasheet?
http://www.ti.com/lit/ds/symlink/tps62120.pdf

And there is always a flight proven example of a MMPT and Lithium Ion charger that was used for the first of the PocketQubes here;

However you really need to start with a good understanding of how you charge Lithium batteries in this type of application. Then you can design your charger.

srnet:
And there is always a flight proven example of a MMPT and Lithium Ion charger that was used for the first of the PocketQubes here;

Dropbox - Maximum Power Point Tracker - Simplify your life

However you really need to start with a good understanding of how you charge Lithium batteries in this type of application. Then you can design your charger.

Hey Stuart thanks for your reply,
I was looking at your design yesterday going through the 50$sat plans, as you said I really need to get a grasp of the subject working on such a project, or find someone to help me out.

I'm going to base myself off something already made as pointed out by other members but your design is really helpful, seems like a little overkill for my needs though.

CrossRoads:
Are you following any of the schematics or the layout diagram in the datasheet?
http://www.ti.com/lit/ds/symlink/tps62120.pdf

Just read the the design procedure, had missed it last time, going to design a schematic on easyeda now and see if it works.

MorganS:
Switching regulators like the TPS62120 are very sensitive to PCB layout. If you are having difficulty with the schematic then you have no hope to get the PCB right on your own.

If you are certain that you need a switching regulator then try working down this decision tree:

  1. Can you buy the regulator as a module on a PCB with all the required components already soldered? Pololu make some good ones.

  2. Can you buy the PCB design as a module to be pasted onto your PCB design? (This is probably going to blow your budget.)

  3. Can you find a free design for the PCB layout with all the components fully specified with part numbers? Some datasheets will show this level of detail or you may find an application note from the chip manufacturer.

  4. Can you afford the time to learn PCB layout and make and test 5 different versions? You are also going to have to learn to use circuit simulation software such as LTSpice.

  5. Do you have a friend who is an expert in designing switching power supplies? The first thing he will tell you is that you chose the wrong chip to begin with and then you find that his recommended chip has a good module that you can buy on ebay for a few bucks. Go back to step 1.

First of all thanks for your reply, really helps

  1. Wow thanks for the link, they are nearly exactly what I need.
  2. Pasted onto my pcb? are you talking about a PCB stencil or having the Module connect to the main PCB with removable pins?
    3 & 4. I have used EasyEda until now and have designed and printed out schematics for a couple projects, they also offer circuit simulation in the editor which I have tested out before.
  3. Unfortunately I do not have a friend that can help me on this though I appreciate any collaborations with the project and all help is really appreciated. The chip I quoted was actually recommended by another forum user.

Regardless, the regulator is something I am not that worried that much about since worse case scenario I can use the linear voltage regulator for now. The MPPT circuit has much less documentation online and I have only managed to find 1 or 2 schematics.

Thanks a lot for your help!

So using a linear regulator is the 'worse case scenario' ?

I would tend to disagree. I realise that the perceived wisdom these days is that switch mode supplies are always more efficient, but this is not the case.

If the MPPT is doing the switching for you, as part of the MPPT processs, then you can arrange to have the Lithium IoN voltage fairly low say 3.8v.

In those circumstance you will probably find that using a switcher to go from 3.8V to 3.3v is less efficient than a linear. And the linear may well have a very low standby current as an added advantage.

srnet:
So using a linear regulator is the 'worse case scenario' ?

I would tend to disagree. I realise that the perceived wisdom these days is that switch mode supplies are always more efficient, but this is not the case.

If the MPPT is doing the switching for you, as part of the MPPT processs, then you can arrange to have the Lithium IoN voltage fairly low say 3.8v.

In those circumstance you will probably find that using a switcher to go from 3.8V to 3.3v is less efficient than a linear. And the linear may well have a very low standby current as an added advantage.

Looking at it now 3.8v from a battery to 3.3v seems like it could work efficiently with a linear regulator, the sprites however need a step up in voltage so a the TPV1040 can do that with the integrated MPPT algorithm. Seems like I can use the TPV1040 in both situations.

In your circuit is the battery constantly charging to 3.9v or does it stop at a certain voltage? I believe your satellite stopped transmitting after a certain voltage, did it have over-discharge protection in case for some reason it did not charge properly?

Thanks

The battery was charged to circa 3.8V and transmit stopped when the battery fell below 3.3V.

Why would you need over discharge protection, your not going to be able to replace the battery if it fails ?

srnet:
Why would you need over discharge protection, your not going to be able to replace the battery if it fails ?

You mean to say "You are not going to be able to replace the battery if it fails, so you seriously need over discharge protection to make sure it never over-discharges and fails as a result (because it is a lithium battery)."

srnet:
The battery was charged to circa 3.8V and transmit stopped when the battery fell below 3.3V.

Why would you need over discharge protection, your not going to be able to replace the battery if it fails ?

Paul__B:
You mean to say "You are not going to be able to replace the battery if it fails, so you seriously need over discharge protection to make sure it never over-discharges and fails as a result (because it is a lithium battery)."

I understand the thought process behind thinking it does not need protection since it will never be replaced but I was reading the alba orbital launch requirements and I think its pretty clear the pocketqube needs to have protection from battery mishaps, people will be dealing with it, it will be enclosed with 6 other pocketqube during launch and you never know what can occur.

Also, where did you request permission for the launch of the satellite? Specifically tracking since I believe we might have some issues with that.

Thanks!

Did you consider a LiFePO₄ battery.
Then you wouldn't need a 3.3volt regulator, because that battery type has an almost flat discharge voltage of 3.2volt, and is fully charged at about 3.5volt. It also seems to have a longer cycle life.
Leo..

Wawa:
Did you consider a LiFePO₄ battery.
Then you wouldn't need a 3.3volt regulator, because that battery type has an almost flat discharge voltage of 3.2volt, and is fully charged at about 3.5volt. It also seems to have a longer cycle life.
Leo..

The electronics I am using require regulated 3.3v anyways, I did consider using a LiFePo4 battery but I cant seem to find a suitable form factor for what I need, they do indeed have longer cycle life and battery temperature handling but unfortunately they don't exist in such small sizes.

I could only find them in round cells.
15mm diameter x 27mm long (600mAh).
Leo..

Paul__B:
You mean to say "You are not going to be able to replace the battery if it fails, so you seriously need over discharge protection to make sure it never over-discharges and fails as a result (because it is a lithium battery)."

You manage the battery with software, keeping well away from the failure point.

If the battery still gets over discharged then something has failed, like the solar charging, so additional over discharge protection does not help.

I understand the thought process behind thinking it does not need protection since it will never be replaced but I was reading the alba orbital launch requirements and I think its pretty clear the pocketqube needs to have protection from battery mishaps, people will be dealing with it, it will be enclosed with 6 other pocketqube during launch and you never know what can occur.

The small Lithium Ion batteries will have there own internal protection, which should keep the Alba orbital people happy. However in reality you cannot (and must not) rely on this circuit to manage the battery, the cut off will be at 2.4V, your satellite has failed if the battery gets this low.

Also, where did you request permission for the launch of the satellite? Specifically tracking since I believe we might have some issues with that.

I dont believe you need permission to launch, but you do need a frequency allocation for the transmitter. For amateur satellites contact the IARU.

Wawa:
I could only find them in round cells.
15mm diameter x 27mm long (600mAh).
Leo..

Seems like its the smallest they make them, gonna look around a bit more but looks like the flat li-ion cells are going to be the easiest to work with.

on another note I found out the provider of the IC offers a circuit simulation tool to calculate resistor and cap values, seems like part of the question is resolved.
Here is the picture of the simulation.

srnet:
The small Lithium Ion batteries will have there own internal protection, which should keep the Alba orbital people happy. However in reality you cannot (and must not) rely on this circuit to manage the battery, the cut off will be at 2.4V, your satellite has failed if the battery gets this low.

I dont believe you need permission to launch, but you do need a frequency allocation for the transmitter. For amateur satellites contact the IARU.

Obviously the battery should never get this low and if it does there is a major design flaw but you never know what can happen such as a cable shorting for some reason. Yeah the integrated battery protection is what I was talking about, some dont have them though.

Talked with the chaps at Alba orbital and they told me I might have issues with tracking since the pocketqube will be launching sprites, still don't have a rely from him on who to contact to receive approval for the launch. They did point me to IARU and ITU.

Hi,
See you are at SIS, is that a University?
Have you tried to use the facilties and the brains of tutors and lecturers to help you?
Is this a academic project?
Make sure your batteries are happy going into a vacuum and operating at extreme temperatures, you are not going to be in the sun all the time.

Tom... :slight_smile: