Topic: Supercap voltage - when does it die?

[azchrisf]

Does anyone know what the voltage of the Supercap will be when it is considered "dead" or fully discharged to the ISS?
I'm sure it's not 0V. And I'm not talking about the backup battery, I'm talking about the Supercap itself.

[clweb]

Fully discharged is not the same as "dead". Fully discharged some millivolts may still be here.
But there are a few "dead" causes. The supercap may be a short circuit, so the votage is zero volt. When it does no more charge the it
may look as it is discharged.
Most of the time a dead supercap is physically damaged with chemical leaking.

[miraculon]

I was interpreting this question a bit differently, as in "when is the super-cap voltage too low to sustain the SIM electronics?".

The TI RF chipset works down to 2V. I am not sure about this, but I have the impression that there is diode-steering between the solar panel, super-cap and battery. If we assume that they are Schottky diodes, they typically have a 0.3V forward voltage drop. This would imply that at least 2.3V or so would be required to sustain the SIM circuitry.

My experience with the batteries is that once it discharges to about 2.7V there will no longer be any data coming from the SIM. The diagnostic LED will not flash below 2.48V according to an experiment I ran years ago. You could infer a similar voltage for the super-cap, I suppose.

The way to be sure would be to remove the battery and plastic cover over the super-cap and monitor the cap voltage until the SIM stops transmitting.

This is not an "authoritative" answer, so if anyone knows for sure it would be good to know.

Greg H.

[azchrisf]

Greg thank you that's exactly what I was asking.
The cap is 2.7v fully charged so I'd guess 2.0v is when it would be considered dead to the ISS. But your right it would require an experiment to know for sure.

[miraculon]

Greg thank you that's exactly what I was asking.
The cap is 2.7v fully charged so I'd guess 2.0v is when it would be considered dead to the ISS. But your right it would require an experiment to know for sure.

Actually a little higher than 2V because of the diode forward voltage drop. If my guess about the diode type is correct, that would be 0.3V, which would make the dropout  voltage about 2.3V.

Greg H.

[azchrisf]

That's a very narrow range of power for the cap...2.7 to 2.3V and it's dead?

[miraculon]

That's a very narrow range of power for the cap...2.7 to 2.3V and it's dead?

It should transfer to the battery, assuming that it is good. I could be mistaken about the diode, they could be using a MOSFET to switch power around, then it would be minimal voltage drop in the millivolt range and make it closer to 2V. The innards of the SIM seems to be a deep dark secret, at least I haven't been able to find a schematic for it.

Greg H.

[azchrisf]

Ah yes I dismissed the battery keeping the cap charged.
It's one of those things where someone who has deep electrical engineering knowledge would need to sit down and follow the circuit from start to finish and draw out a schematic. It's all SMD components so they would need a magnifier too

If someone could do that it would help the community a GREAT deal.

[azchrisf]

So I got an extra board to experiment with, and I found with zero voltage in the Supercap, and I add in the battery, the voltage stays steady at 0.90 volts, and the ISS is transmitting. Remove the battery and of course it drops over time, I did not watch where it stops transmitting. Now with the TI chipset being a 2.0V chipset, that would have to say there is a boost converter in the circuitry somewhere. I see a couple coils in the circuitry, whether that's there for buck/boost I dunno.

But still interesting. I'll look at it further.

[doubleohwhatever]

So I got an extra board to experiment with

Any idea what pins 1 and 6 are for on the ISS RJ12 connector? I've been meaning to investigate but don't have a spare board on my desk at the moment.

[azchrisf]

Pinouts I have no idea. Wish I could be of more help

[johnd]

So I got an extra board to experiment with

Any idea what pins 1 and 6 are for on the ISS RJ12 connector? I've been meaning to investigate but don't have a spare board on my desk at the moment.

Which connector are you asking about? Presumably you're using a wireless VP2? So you must mean one of the sensor connectors?

[doubleohwhatever]

I'm referring to the connector in the upper right corner of the ISS board. Pins 3 and 4 are the cables ISS RS422 connection. Pins 2 and 5 are power. Pins 1 and 6 are the pins I'm wondering about.

[johnd]

I'm referring to the connector in the upper right corner of the ISS board.

Ah OK. When used on cabled SIM boards for normal data transfer, the cable is obviously just 4-conductor and so 1/6 would not then be needed of course. But maybe this connector is also used eg for flashing the ISS firmware (just a speculation, but there clearly are other options for flashing), so perhaps other connections might be needed during manufacture. Unfortunately there's zero technical documentation available from Davis on aspects like this, even on a confidential basis AFAIK.

[azchrisf]

I'm sure there are no firmware upgrade facilities for the SIM boards, the chips are preburnt at the fab with the code Davis provides for economy of scale. If they must make a major code revision, new chips are burnt. What they would do with the existing stock I don't know.

[Mattk]

I'm sure there are no firmware upgrade facilities for the SIM boards, the chips are preburnt at the fab with the code Davis provides for economy of scale. If they must make a major code revision, new chips are burnt. What they would do with the existing stock I don't know.

For that exact reason of economy of scale nobody would make a board that could not be modified. I find it rather strange that some believe the ISS can not be modified and programmed?
   

[johnd]

I'm sure there are no firmware upgrade facilities for the SIM boards, the chips are preburnt at the fab with the code Davis provides for economy of scale. If they must make a major code revision, new chips are burnt. What they would do with the existing stock I don't know.

Having one flash method for mass manufacture doesn't rule out having another for post-manufacture updates. By that logic we shouldn't be able to flash the console with firmware updates. But firmware flashing was just one thought. Production testing might be another. Or setting the frequency range post-manufacture.

Overall there are several possible uses for that connector, but I'm not sure anyone outside of Davis actually knows.  (It's possible that a Davis distributor might have done post-supply reprogramming - it's never been necessary for major markets like the US or Europe, but some territories like Australia have their own frequency ranges and I've got a vague recollection (BICBW) that eg AU used to do some final programming locally at one time in the past. So maybe someone at the distributor there might know more?)

[azchrisf]

I'm sure there are no firmware upgrade facilities for the SIM boards, the chips are preburnt at the fab with the code Davis provides for economy of scale. If they must make a major code revision, new chips are burnt. What they would do with the existing stock I don't know.

Having one flash method for mass manufacture doesn't rule out having another for post-manufacture updates. By that logic we shouldn't be able to flash the console with firmware updates. But firmware flashing was just one thought. Production testing might be another. Or setting the frequency range post-manufacture.

Overall there are several possible uses for that connector, but I'm not sure anyone outside of Davis actually knows.  (It's possible that a Davis distributor might have done post-supply reprogramming - it's never been necessary for major markets like the US or Europe, but some territories like Australia have their own frequency ranges and I've got a vague recollection (BICBW) that eg AU used to do some final programming locally at one time in the past. So maybe someone at the distributor there might know more?)

Um, no. My logic does not say that. We know different already so...
There are no apparent user-facing methods available for firmware updates, not to mention no updates I've seen or read about for the ISS itself, but that does not mean there is not a factory method such as JTAG or direct pin to pin jig writing...but considering the board is encapsulated that itself would be near impossible without damaging the pins removing the encapsulation. Second, as far as we know, the processing is handled on the console. That is where a majority of your bug fixes would happen. You can manipulate the data any way you want to fix problems, etc. after it is at ultimate destination. And second to that, even doing firmware updates on the consoles is not easy. But we know it's possible... Davis designed all this stuff before user upgrades became a norm in the later 2000's.

In the end, who cares. Noone is going to know except Davis either which way.

[EA1EF]

Sorry if offtopic, but LIFEPO4 seem good fit with the voltage range of 3.3v electronics.

May be necessary limit peak voltage at charging, because lifepo4 when full 100% charged  up easy from 3.6v to 4.2 volts, probably must limit to 3.4 or 3.45v for reach 85 or 90% charged.

Eventually lifepo4 and solar charging with 7000 mA 32700 would be autonomous

[azchrisf]

I'm not sure it would work with a 3.3v Battery. It might though... Be an interesting experiment.
Anyways, incase anyone is interested the cap is rated for 2.7V, but it stops charging/full voltage on the cap is 2.56V (~5% safety margin), and it stops transmitting at approx. 0.65V. So about 2V is available until it's fully depleted.

Of course, the battery helps keep it charged.

[azchrisf]

Now, I thought I'd add in for anyone interested:

You can actually calculate the time the Supercap will last (the built-in one is 10F), if you know the voltage draw. That I haven't figured out.
But I believe Davis specs the 10F cap for short of overnight, so about 10-12 hours. So a 50F cap should last 2.5 days, and a 100F cap about 5 days and a 200F cap over a week.

When you go larger than 100F you will probably have to leave the cover off the cap.