More battery charger, voltage shunt reference.

Uh-uh, nothing like adding a $10 chip to the mix. Like I said, just a bunch of hand waving.

No biggie, just swap the destinations of the collector leads. The old time schematics used to stylistically crisscross them so maybe that was missed in the transcription.

It isn't cheap, but it is good and low-power. Lets see your design ...

That ought not to be a problem in real life. A battery is a perfectly enormous "capacitance", and should swamp any other lag in the circuit.

You might use a smaller PNP in the long-tailed pair. The 2N4208(National Process 65 where the 2N306 is process 66) is smaller and has slightly lower capacitances. It's gold-doped, and mainly intended for switching, but might be worth trying.

I've tinkered with your .asc file it to put in the HP HLMP-D150 LED as the voltage reference. The Spice model is just the LTSpice QTLP690C Fairchild red LED model with Is tweaked up a bit to give an exactly 1.5V reference voltage.

I've fiddled the resistors R1, R4.R5 and R6 to match the lower reference voltage.

I wouldn't trust it as far as I could kick it, but it does show what I had in mind. Version 4 SHEET 1 1612 680 WIRE 560 -352 512 -352 WIRE 672 -352 560 -352 WIRE 800 -352 672 -352 WIRE 896 -352 800 -352 WIRE 1120 -352 896 -352 WIRE 1296 -352 1120 -352 WIRE 1312 -352 1296 -352 WIRE 672 -320 672 -352 WIRE 800 -304 800 -352 WIRE 896 -304 896 -352 WIRE 1120 -304 1120 -352 WIRE 896 -192 896 -224 WIRE 896 -192 800 -192 WIRE 992 -192 896 -192 WIRE 1296 -160 1296 -352 WIRE 672 -144 672 -256 WIRE 736 -144 672 -144 WIRE 1120 -144 1120 -224 WIRE 1120 -144 1056 -144 WIRE 800 -80 800 -96 WIRE 1040 -80 800 -80 WIRE 1120 -80 1120 -144 WIRE 512 -64 512 -352 WIRE 992 -48 992 -96 WIRE 992 -48 896 -48 WIRE 800 -16 800 -80 WIRE 992 -16 992 -48 WIRE 896 32 896 -48 WIRE 896 32 864 32 WIRE 928 32 896 32 WIRE 1120 32 1120 0 WIRE 1200 32 1120 32 WIRE 1296 32 1296 -80 WIRE 1296 32 1200 32 WIRE 1312 32 1296 32 WIRE 1120 64 1120 32 WIRE 672 112 672 -144 WIRE 800 112 800 80 WIRE 992 112 992 80 WIRE 1040 144 1040 -80 WIRE 1056 144 1040 144 WIRE 1072 144 1056 144 WIRE 512 224 512 16 WIRE 560 224 512 224 WIRE 672 224 672 192 WIRE 672 224 560 224 WIRE 800 224 800 192 WIRE 800 224 672 224 WIRE 992 224 992 192 WIRE 992 224 800 224 WIRE 1120 224 1120 160 WIRE 1120 224 992 224 FLAG 800 -304 0 FLAG 1200 32 Vout FLAG 560 -352 VIN+ FLAG 560 224 VIN- FLAG 1056 144 VG SYMBOL res 1136 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R1 SYMATTR Value 33k SYMBOL res 912 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R6 SYMATTR Value 68k SYMBOL pnp 1056 -96 R180 WINDOW 3 55 66 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q1 SYMBOL pnp 736 -96 M180 WINDOW 3 51 67 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q3 SYMBOL npn 928 -16 R0 WINDOW 3 -39 109 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q2 SYMBOL res 816 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL res 1008 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R3 SYMATTR Value 100k SYMBOL res 1136 16 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R4 SYMATTR Value 180k SYMBOL npn 864 -16 M0 SYMATTR InstName Q5 SYMATTR Value 2N3904 SYMBOL res 688 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R5 SYMATTR Value 680k SYMBOL nmos 1072 64 R0 SYMATTR InstName M1 SYMATTR Value IRF530 SYMBOL voltage 512 -80 R0 SYMATTR InstName V1 SYMATTR Value 14 SYMBOL current 1296 -160 R0 WINDOW 3 57 154 VLeft 2 SYMATTR Value PULSE(0 -1 10u 0 0 100u) SYMATTR InstName I1 SYMBOL LED 656 -320 R0 SYMATTR InstName D1 SYMATTR Value HLMP-D150 SYMATTR Description Diode SYMATTR Type diode TEXT 264 -248 Left 2 ;.ac oct 1000 10k 1e6 TEXT 768 -400 Left 2 ;Solar fed lead-acid charger TEXT 328 -208 Left 2 ;.tran 250u TEXT 320 -168 Left 2 ;.dc V1 10 20 0.1 TEXT 1160 -128 Left 2 ;Load for\nsimulation TEXT 320 -128 Left 2 !.dc I1 0 2 0.1 TEXT 1336 -352 Left 2 ;VOUT+ TEXT 1328 32 Left 2 ;VOUT- TEXT 288 256 Left 2 !.model HLMP-D150 D(Is=2.5e-22 Rs=6 N=1.5 Cjo=50p Xti=100 Iave=160m Vpk=5 mfg=HP type=LED)

I don't know how you managed to do that, but the input differential pair transistors are missing, the diode and biasing resistors for the load mirror are missing, parts of the resistor divider for the battery voltage sample are missing, It's impossible to determine what your intentions are with this incomplete circuit, especially the part about replacing the battery with a current source.. Maybe try updating your version of LTSpice, looks like you're years out of date.

Fred - I think the problem is on your end.

I imported Bill's LTSpice schematic without any problem and it was complete.

I use Google Groups and I had to be careful how I imported it as it clipped some of the lines in the middle of the circuit - maybe you had a similar problem.

kevin

That fixed it. Sloman screwed up the circuit. He returned R5 (ref voltage divider) to chassis gnd instead of BATT (-) terminal. And none of his simulations do a temperature sweep, which was kind of the point of the exercise.

Kevin did that. I just changed his circuit to use a different (and lower) voltage reference - the red LED at D1. I changed resistors R1, R4.R5 and R6 to match the lower reference voltage.

They are Kevin's simulations, not mine, and who would bother to get LTSpice to do a temperature sweep?

The red LED sits at about 1.5V and decreases by 2mV per degree. If you scaled that up to match the 13.5V you get out of a fully charged 6 cell lead-acid battery (which Kevin hasn't done) you'd get the -18mV/C which you seem to think is required. That would mean increasing R4 from 10k to 264k (R1 at 30k, R5 at 240k would work with E24 values) - but throwing in a pot would make more sense

Precisely what the circuit is actually supposed to do isn't clear. I'd have to dig a solar panel and it's data sheet to work out what a sensible circuit would look like, and who wants to re-invent the wheel?

It looks as if he has decided that his solar cell (or whatever V1 is supposed to be ) will always deliver 14V and he wants to use M1 to turn off the charging current when the battery - his "load for simulation" - gets charged up far enough.

He did mention that he'd left out the diode that stopped the battery forcing current through the solar cell when the sun wasn't shining.

But it was you who posted the spice netlist.

The temperature dependence of charger threshold is to compensate for lead acid battery cell voltage corresponding to 100 % state-of-charge, SOC for short. The 6-cell needs -18mV/oC. Some of the semis, like Motorola and Unitrode had this built-in to their charger ICs.

The PV isn't going to overcharge the battery without reasonably strong insolation, and since the panels designed for 12 Volt charging have open circuit voltages of 22 Volt, it's safe to assume the PV is operating past its "knee" while doing any significant charging. This means a simple voltage source with open circuit voltage the PV knee voltage in series with a Voc/Isc resister is an adequate model for the PV in the circuit sim. No point in more precision than that because the insolation puts the PV I-V all over the map anyway. You need to sweep the circuit temperature from 0-70oC and plot the terminal charge voltage across the battery. Charge times would involve integrating the charging current and scaling by an efficiency factor. Although that can be done in principle, good luck working with LTSpice getting it done. I can't get it to add 2 + 2 without it fussing about some bullsh_t.

On Saturday, 20 March 2021 at 07:15:11 UTC-7, Fred Bloggs wrote: ...

I (kevin) did that, and commented in my post that it was done to enable simulation into a resistive load or the circuit wouldn't start-up - it would work into a battery as the battery would provide the initial drive and also provide a more stable voltage to improve the operation of the reference.

...

There wasn't much point as the Ltspice models didn't model reverse Veb breakdown so I replaced it with a proper zener.

kevin

On Saturday, 20 March 2021 at 08:03:49 UTC-7, Bill Sloman wrote: ...

I wanted to float charge a gel-cell from a small photovoltaic array to power an experiment I was playing with and didn't want the quiescent current from the usual op-amps (eg LM324, LM358) to consume much of my available power.

That 14V was for simulation purposes - a photovoltaic array acts as a constant current generator in parallel with a string of diodes. As such when loaded its voltage drops significantly.

In this application, the FET is turned on completely when the battery is below the target voltage so the solar array collapses to the battery voltage plus the few millivolts drop across the FET. ...

kevin

On Saturday, 20 March 2021 at 08:36:42 UTC-7, Fred Bloggs wrote: ...

Here is Bills' version with an LED reference with a temperature sweep added.

The temperature coefficient seems a bit high - I'm getting about 61mV/deg at the battery (~7mV/deg at the diode).

I haven't looked into the diode model at all.

Agreed, At 14.5V output, a 36-cell array is a bit lower voltage than its maximum power point.

Here is the LTspice netlist that does that (well 0-60C anyway). ...

Version 4 SHEET 1 1612 680 WIRE 560 -352 512 -352 WIRE 672 -352 560 -352 WIRE 800 -352 672 -352 WIRE 896 -352 800 -352 WIRE 1120 -352 896 -352 WIRE 1200 -352 1120 -352 WIRE 1296 -352 1200 -352 WIRE 1312 -352 1296 -352 WIRE 672 -320 672 -352 WIRE 800 -304 800 -352 WIRE 896 -304 896 -352 WIRE 1120 -304 1120 -352 WIRE 1200 -304 1200 -352 WIRE 896 -192 896 -224 WIRE 896 -192 800 -192 WIRE 992 -192 896 -192 WIRE 1296 -160 1296 -352 WIRE 672 -144 672 -256 WIRE 736 -144 672 -144 WIRE 1120 -144 1120 -224 WIRE 1120 -144 1056 -144 WIRE 800 -80 800 -96 WIRE 1040 -80 800 -80 WIRE 1120 -80 1120 -144 WIRE 512 -64 512 -352 WIRE 992 -48 992 -96 WIRE 992 -48 896 -48 WIRE 800 -16 800 -80 WIRE 992 -16 992 -48 WIRE 896 32 896 -48 WIRE 896 32 864 32 WIRE 928 32 896 32 WIRE 1120 32 1120 0 WIRE 1200 32 1200 -224 WIRE 1200 32 1120 32 WIRE 1296 32 1296 -80 WIRE 1296 32 1200 32 WIRE 1312 32 1296 32 WIRE 1120 64 1120 32 WIRE 672 112 672 -144 WIRE 800 112 800 80 WIRE 992 112 992 80 WIRE 1040 144 1040 -80 WIRE 1056 144 1040 144 WIRE 1072 144 1056 144 WIRE 512 224 512 16 WIRE 560 224 512 224 WIRE 576 224 560 224 WIRE 672 224 672 192 WIRE 672 224 640 224 WIRE 800 224 800 192 WIRE 800 224 672 224 WIRE 992 224 992 192 WIRE 992 224 800 224 WIRE 1120 224 1120 160 WIRE 1120 224 992 224 FLAG 800 -304 0 FLAG 1200 32 Vout FLAG 560 -352 VIN+ FLAG 560 224 VIN- FLAG 1056 144 VG SYMBOL res 1136 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R1 SYMATTR Value 24k SYMBOL res 912 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R6 SYMATTR Value 68k SYMBOL pnp 1056 -96 R180 WINDOW 3 55 66 Left 2 WINDOW 0 84 32 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q1 SYMBOL pnp 736 -96 M180 WINDOW 3 51 67 Left 2 WINDOW 0 84 32 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q3 SYMBOL npn 928 -16 R0 WINDOW 3 -39 109 Left 2 WINDOW 0 56 32 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q2 SYMBOL res 816 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL res 1008 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R3 SYMATTR Value 100k SYMBOL res 1136 16 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R4 SYMATTR Value 180k SYMBOL npn 864 -16 M0 WINDOW 0 56 32 Left 2 WINDOW 3 56 68 Left 2 SYMATTR InstName Q5 SYMATTR Value 2N3904 SYMBOL res 688 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R5 SYMATTR Value 680k SYMBOL nmos 1072 64 R0 WINDOW 0 56 32 Left 2 WINDOW 3 56 72 Left 2 SYMATTR InstName M1 SYMATTR Value IRF530 SYMBOL voltage 512 -80 R0 SYMATTR InstName V1 SYMATTR Value 14 SYMBOL current 1296 -160 R0 WINDOW 3 11 88 Left 2 SYMATTR Value 0 SYMATTR InstName I1 SYMBOL LED 656 -320 R0 SYMATTR InstName D1 SYMATTR Value HLMP-D150 SYMATTR Description Diode SYMATTR Type diode SYMBOL res 1216 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R7 SYMATTR Value 10 SYMBOL schottky 640 208 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D2 SYMATTR Value RB060M-40 SYMATTR Description Diode SYMATTR Type diode TEXT 184 -288 Left 2 ;.ac oct 1000 10k 1e6 TEXT 768 -400 Left 2 ;Solar fed lead-acid charger TEXT 184 -256 Left 2 ;.tran 250u TEXT 184 -224 Left 2 !.dc V1 0 20 0.1 TEXT 1264 -48 Left 2 ;Load for\nsimulation TEXT 184 -192 Left 2 ;.dc I1 0 2 0.1 TEXT 1336 -352 Left 2 ;VOUT+ TEXT 1328 32 Left 2 ;VOUT- TEXT 248 288 Left 2 !.model HLMP-D150 D(Is=2.5e-22 Rs=6 N=1.5 Cjo=50p Xti=100 Iave=160m Vpk=5 mfg=HP type=LED) TEXT 184 -320 Left 2 !.temp 0 10 20 30 40 50 60

And I made it perfectly clear that it changed Kevin's netlist only to the extent of replacing his voltage reference with my red LED.

You could, but why bother.

That's computer simulation for you. It's worth doing when there is something messy and complicated going on, but this doesn't seem to be such as case.

The 680k resistor in series with my red LED would have to cope with quite a lot of charging voltage variation.

I've edited Kevin's most recent circuit to upgrade the reference generator to a ring-of-two which should fix the reference current to 5uA over a wide range of charging voltages.

I suppose you could exploit the stable voltage across D3 to bias a constant current load, which would let you get rid of Q2 and R3, at the price of reducing the voltage gain a little (but that's mostly going to be determined by the Early effect in Q5). R2 would get pushed up to 140k.

Version 4 SHEET 1 1612 680 WIRE 320 -352 272 -352 WIRE 432 -352 320 -352 WIRE 624 -352 432 -352 WIRE 800 -352 624 -352 WIRE 896 -352 800 -352 WIRE 1120 -352 896 -352 WIRE 1200 -352 1120 -352 WIRE 1296 -352 1200 -352 WIRE 1312 -352 1296 -352 WIRE 432 -320 432 -352 WIRE 624 -304 624 -352 WIRE 800 -304 800 -352 WIRE 896 -304 896 -352 WIRE 1120 -304 1120 -352 WIRE 1200 -304 1200 -352 WIRE 896 -192 896 -224 WIRE 896 -192 800 -192 WIRE 992 -192 896 -192 WIRE 1296 -160 1296 -352 WIRE 432 -144 432 -256 WIRE 736 -144 432 -144 WIRE 1120 -144 1120 -224 WIRE 1120 -144 1056 -144 WIRE 624 -112 624 -224 WIRE 800 -80 800 -96 WIRE 1040 -80 800 -80 WIRE 1120 -80 1120 -144 WIRE 272 -64 272 -352 WIRE 432 -64 432 -144 WIRE 560 -64 432 -64 WIRE 992 -48 992 -96 WIRE 992 -48 896 -48 WIRE 800 -16 800 -80 WIRE 992 -16 992 -48 WIRE 432 16 432 -64 WIRE 896 32 896 -48 WIRE 896 32 864 32 WIRE 928 32 896 32 WIRE 624 64 624 -16 WIRE 624 64 496 64 WIRE 624 112 624 64 WIRE 1120 128 1120 0 WIRE 1200 128 1200 -224 WIRE 1200 128 1120 128 WIRE 1296 128 1296 -80 WIRE 1296 128 1200 128 WIRE 1312 128 1296 128 WIRE 432 160 432 112 WIRE 1120 160 1120 128 WIRE 800 208 800 80 WIRE 992 208 992 80 WIRE 1040 240 1040 -80 WIRE 1056 240 1040 240 WIRE 1072 240 1056 240 WIRE 272 320 272 16 WIRE 320 320 272 320 WIRE 336 320 320 320 WIRE 432 320 432 240 WIRE 432 320 400 320 WIRE 624 320 624 176 WIRE 624 320 432 320 WIRE 800 320 800 288 WIRE 800 320 624 320 WIRE 992 320 992 288 WIRE 992 320 800 320 WIRE 1120 320 1120 256 WIRE 1120 320 992 320 FLAG 800 -304 0 FLAG 1200 128 Vout FLAG 320 -352 VIN+ FLAG 320 320 VIN- FLAG 1056 240 VG SYMBOL res 1136 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R1 SYMATTR Value 24k SYMBOL res 912 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R6 SYMATTR Value 68k SYMBOL pnp 1056 -96 R180 WINDOW 3 55 66 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q1 SYMBOL pnp 736 -96 M180 WINDOW 3 51 67 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q3 SYMBOL npn 928 -16 R0 WINDOW 3 27 59 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q2 SYMBOL res 816 304 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL res 1008 304 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R3 SYMATTR Value 100k SYMBOL res 1136 16 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R4 SYMATTR Value 180k SYMBOL npn 864 -16 M0 SYMATTR InstName Q5 SYMATTR Value 2N3904 SYMBOL res 448 256 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R5 SYMATTR Value 180k SYMBOL nmos 1072 160 R0 SYMATTR InstName M1 SYMATTR Value IRF530 SYMBOL voltage 272 -80 R0 SYMATTR InstName V1 SYMATTR Value 14 SYMBOL current 1296 -160 R0 WINDOW 3 11 88 Left 2 SYMATTR Value 0 SYMATTR InstName I1 SYMBOL LED 416 -320 R0 SYMATTR InstName D1 SYMATTR Value HLMP-D150 SYMATTR Description Diode SYMATTR Type diode SYMBOL res 1216 -208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R7 SYMATTR Value 10 SYMBOL schottky 400 304 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D2 SYMATTR Value RB060M-40 SYMATTR Description Diode SYMATTR Type diode SYMBOL pnp 560 -16 M180 WINDOW 3 51 67 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q4 SYMBOL res 608 -320 R0 SYMATTR InstName R8 SYMATTR Value 180k SYMBOL LED 608 112 R0 SYMATTR InstName D3 SYMATTR Value HLMP-D150 SYMATTR Description Diode SYMATTR Type diode SYMBOL npn 496 16 M0 SYMATTR InstName Q6 SYMATTR Value 2N3904 TEXT -56 -288 Left 2 ;.ac oct 1000 10k 1e6 TEXT 768 -400 Left 2 ;Solar fed lead-acid charger TEXT -56 -256 Left 2 ;.tran 250u TEXT -56 -224 Left 2 !.dc V1 0 20 0.1 TEXT 1264 -48 Left 2 ;Load for\nsimulation TEXT -56 -192 Left 2 ;.dc I1 0 2 0.1 TEXT 1336 -352 Left 2 ;VOUT+ TEXT 1328 128 Left 2 ;VOUT- TEXT 248 384 Left 2 !.model HLMP-D150 D(Is=2.5e-22 Rs=6 N=1.5 Cjo=50p Xti=100 Iave=160m Vpk=5 mfg=HP type=LED) TEXT -56 -320 Left 2 !.temp 0 10 20 30 40 50 60