Need help with a project involving solar cells and SLAs

[poobah]

Yeah... that trick never really occured to me until I blew a Powerex IGBT module to smitherines... 1200 Volt / 600 Amp jobbie. Once we cleaned away the ektoplasm you could see the Zee wiring of the individual chips...

[Saturnus]

Originally posted by poobah
Depending on the size of his batteries, sag should not be a problem... SLA's pretty damn stiff as far as that goes.

OH... and BTW you don't need to disconnect the solar panel when using the charger. You WILL want a diode in series with the charger to prevent discharging the batteries through the charger's internal reisitors though.

My SLAs are two standard Panasonic 12V 7Ah types mostly because they're dirt cheap here. Around half the price of either a 4Ah or a 9Ah.

Strangely enough I have intuitively wired my SLAs in the Zee connetion you suggest. It just seemed logical to me.

Could I just wire it as shown below? It's schottkys all the diode shown except the regulating zener. And can I even use a zener here? It's my understanding that it dissipates only surplus voltage above the zener point but I'm not sure that totally correct.

Sorry for the lousy drawing. It's free hand *shrug*

Thanks for the advice.

Johnny

[classd4sure]

If it's only dissipating surplus, it's not regulating anything. It's really not a good idea. Use enough capacitance like you normally would anyway, and it will absorbe recirculating current nicely.

What of the possibility of removing that third schottky so that any pumping can help charge the batteries?

[Saturnus]

Originally posted by classd4sure
If it's only dissipating surplus, it's not regulating anything. It's really not a good idea. Use enough capacitance like you normally would anyway, and it will absorbe recirculating current nicely.

Yes, but the output needs to be 13.2V maximum with no limit to minimum voltage. The solar cells/SLAs/charger is driving a T-amp which only takes up to 13.2V supply voltage. The idea was to ensure that only 13.2V was outputted by using 4 3.3V 5W zener in series. You're saying that this won't work?

I want minimum voltage loss so probably a shunt regulator then if the above doesn't work?

With regards to you edited suggestion, I suppose I could move the cap to before the schottky and have the same effect?

[poobah]

Aha...

Your Solar Current is 1 amp and your batteries are 7 Amp-Hour...

You will need some charge limiting after all...

Hang on... I'll give you a circuit...

[classd4sure]

Originally posted by Saturnus


Yes, but the output needs to be 13.2V maximum with no limit to minimum voltage. The solar cells/SLAs/charger is driving a T-amp which only takes up to 13.2V supply voltage. The idea was to ensure that only 13.2V was outputted by using 4 3.3V 5W zener in series. You're saying that this won't work?

I want minimum voltage loss so probably a shunt regulator then if the above doesn't work?

With regards to you edited suggestion, I suppose I could move the cap to before the schottky and have the same effect?

I like the shunt regulation alot better of course, but yah that could work I'm guessing. (Haven't done the math tho).

How about a high power TVS as an option to series'd zeners?

I don't think you'd want to move the caps to before the diodes, then the zener would have to drop the pumping current, and you want the caps right at the rails of the amps anyway.

Don't mean to complicate things BTW, but it's always fun to explore options, poobah will keep me straight

[poobah]

This is a linear shunt regulator that I have built thousands of...

The TO220 device needs a good heat sink.

Calculate everything to yield 13.8 to 14.1 at the battery and you are good to go...

[Saturnus]

When I said a total electronics design n00b I weren't kidding. So when you off-handedly wrote "calculate everything", I had no idea where to even start looking for information that would enable me to do that.

A few hours on the internet later, I think I'll take a stab at it though.

Firstly, let me see if I get this circuit correctly in it base function. It's basically a transistor assisted shunt regulator, right?

If that's the case then I think the R(collector) should be a 10ohm 10watt type.

The R(ref1) and R(ref2) sets the V(ref) for the shunt, so with an V(input) of 16V and a wanted V(ref) of 14V then R(ref2) should 7 times larger than R(ref1). For convinience let's set it those at 1Kohm and 6.8Kohm respectively. I(ref) for the shunt regulator would then be 8mA which is within the acceptable range.

I'm a little at a loss as for what R(emiter) should be as it would also depend on the used BJT but a stab in the dark could be any value from 1Kohm to 10Kohm, based on assuming it's supposed to be R(collector)*DC gain.

Hope I'm at least partly right. *fingers crossed*

EDIT: I should note that I'm a woodworker by profession and didn't go further than high school, so this tech stuff is really new to me. I have though been programming CNC machinery for a number of years, so the math is somewhat familiar at least.

[poobah]

You're doing pretty damn good!

The collector resistor should ideally be 7 ohm (use 7.5), this unloads as much heat as possible from the transistor (1/2 the total voltage).

Rbase can be in the range of 1 to 10 kOhm. Best to calculate that based on minimum Beta of the transistor; realizing that you only have about 10 Volts or so across the base resistor to achieve that. This resistor "degenerates the gain in the TL somewhat... no harm in that as it imporves stabilty.

Iref can easily be cut to 1 mA. So that leaves 14.4 kOhm for the total divider impedance.

The ref voltage should be 2.495 Volts at 14.4 Volts (adding in an extra 0.4 for the diode drop) The lower resistor should be 2.49 kOhm; the upper should be 11.8 kOhm. You should use 1%'s... 0.1% better still (or a pot).

Carry on!

[Saturnus]

This crash course in electronic really has me on the edge on my seat, and I'm trying really hard to figure things out first before asking.

Let's go back to the design, and work backwards. The T-amps has a maximum rated input voltage of 13.2 (12V nominal) which I guess is Tripath just setting a 10% tolerance from the nominal voltage. If they are feed from the SLAs through a diode with a voltage drop 0.6V that is exactly the same 13.8V that is specified on my SLAs as maximum constant voltage. It's my opinion (I could by wrong) that it's best to leave as much voltage regulation to the shunt as possible, so I'll use a schottky which has a voltage drop of 0.2 as isolation. That should leave a regulated voltage of 14V. And 14.4V as you suggest, if I use a regular diode. That is a correct assumption, right? Or maybe you mean that the shunt drops 0.4V?

With V(out)=14V and we accept a slightly higher I(ref) than you suggest (~3mA) then V(ref)=V(out)/(1+R(upper)/R(lower)=14/(1+4.7K/1K)=2.456V, which would use standard values for the voltage divider.

So R(colletor) should be V(out)/2? Interesting! So I could use 1 7.5ohm 10watt type, or 3 22ohm 3watt types.

I think it's probably getting too late for going into choosing a suitable BJT and a R(base) tonight but I ponder over it tomorrow at work, whih basically means another day of not doing as much as I should.

Cheers

Johnny