Varivolt PV (4 selectable voltages, no potentiometer)

[o4_srt]

Just a simple, variable voltage (I'm setting it to 3.7, 5, 5.5 volts), with 6.1 volts available if I close all the DIP switches.

Using an LM317T (rated at 2.2 amps if Vin-Vout <= 20 volts) because they are cheap, and readily available to me. (I know the schematic says K, ignore)

I decided to go against using a potentiometer simply because I don't want to have to play around with it. Just flick a switch and it's where I want it.

Everything SHOULD fit into a 3xAA battery box, using 2x14500's. Might be a little tight, but i SHOULD be able to make it fit.


God I love multisim



Should cost under $10, not including shipping.

 

[Rocketman]

Pretty interesting idea. The proper selection of resistors would give you several voltages. Closing all three switches would give you a voltage too low though.

 

[o4_srt]

Pretty interesting idea. The proper selection of resistors would give you several voltages. Closing all three switches would give you a voltage too low though.

you are correct, closing all 3 gives me 2.3 volts, i MEANT to say OPENING all the switches gives 6.1 volts



simply changing the values of R2, R3, and R4 can give you many, many different options. Still have not decided if I want to use a DIP switch, or SP3T slide switch to select voltage.

 

[WillyB]

you are correct, closing all 3 gives me 2.3 volts, i MEANT to say OPENING all the switches gives 6.1 volts

I'm confused as to how you figured this. With R1 @ 120, R2 would have to be about 470Ω for 6.1V.

Also everything I've read pegs the LM317T as a 1.5A reg.

The LM317T can handle average currents of up to 1.5 Amps

LM117/LM317A/LM317 Package Options

Output
Current
LM117, LM317 K TO-3 1.5A
LM317A, LM317 T TO-220 1.5A
LM317 S TO-263 1.5A
LM317A, LM317 EMP SOT-223 1.0A
LM117, LM317A, LM317 H TO-39 0.5A
LM117 E LCC 0.5A
LM317A, LM317 MDT TO-252 0.5A

 

[o4_srt]

I'm confused as to how you figured this. With R1 @ 120, R2 would have to be about 470Ω for 6.1V.

Also everything I've read pegs the LM317T as a 1.5A reg.

multisim (SPICE simulator/design software) predicts 6.1 volts with all 3 DIP switches open.

http://www.datasheetcatalog.org/datasheet2/a/0a5g8w9ft4l5px3509xfla6uksyy.pdf


Take a look at maximum output current. It is greater if the voltage differential is less than 15v (some manufacturers are 20v). 1.5 amps is guaranteed, greater depends on application.

"This monolithic integrated circuit is an adjustable 3-terminal
positive voltage regulator designed to supply more than 1.5A
of load current with an output voltage adjustable over a 1.2
to 37V."

 

[Rocketman]

Average current, total dissipation within the regulator will be small compared to it's capabilities. Minimal heatsinking would be required if the model used had some thermal mass to it, a TO-220, TO-3 might fit in a box mod.
Without a sense return path (r2/r1 some finite value) some uses may not be stable. A high static unswitched resistance value like 1200 ohms (r2/r1 =10) might be an advantage as it would also group the switched voltages (8 of them with 3 switches) a little closer.

The 1200 ohm static r2 would give max output (battery - drop across regulator).
r2 values like maybe 610, 670, and 730 would give several output voltages close to:
6.0, 5.75, 5.5, 4.3, 4.0, 3.9, 3.25. Maybe a little tweeking on the r values would give better values. Someone try 630 ohms, 670 ohms, and 710 ohms for the 3 r2 values (don't forget the parallel 1200 ohm static r2 resistor).
My brain is tired Hope someone checks my math here

The LM317K and other adjustable regulators, some with even lower voltage drops could be used with a dip switch programming resistor network. I have some 10 switch dip switches, wow

 

[Rocketman]

Pretty interesting idea.
This could really get complicated though.

What if you set up a spreadsheet for r1=120
r2 (static) and 3 r2 (switched) and ran them through your simulator?

Max out with only r2static=470 ohms,minimum voltage with all 4 resistors.
r2 switched values of 970 ohms, 1370 ohms, 1940 ohms, or whatever is the closest standard value to each?
8 voltages from about 3.7 to 6.1, I think
Rocket

 

[o4_srt]

Pretty interesting idea.
This could really get complicated though.

What if you set up a spreadsheet for r1=120
r2 (static) and 3 r2 (switched) and ran them through your simulator?

Max out with only r2static=470 ohms,minimum voltage with all 4 resistors.
r2 switched values of 970 ohms, 1370 ohms, 1940 ohms, or whatever is the closest standard value to each?
8 voltages from about 3.7 to 6.1, I think
Rocket

thought about that as well, but decided I only need 3 voltages. I am, indeed, planning on using a TO-220 package, securing it with DP-190 to the case. TO-220 fits in a battery well inside a 3xAA battery case, tested it at work the other day. The DP-190 should give it more than enough thermal mass to adequately heat sink what little heat it will make. It's not like we run the things constantly.

there are plenty of LM317 calculators online that give you what voltage is supplied with which resistors.

Also thought of using a tri-color LED tied to the three set lines so I know what I'm vaping at, but haven't found one that will work without a little more thought/circuitry.

 

[Rocketman]

Just 3 voltages? when the others are almost free?
How about adding one of those 10 led blocks and an LM3914.
First led for on, 8 for voltage, one for vape, and an MP3 player (just kidding)

But for real, right selection of resistors in your circuit would give you 8 voltages.

Have fun

 

[o4_srt]

Just 3 voltages? when the others are almost free?
How about adding one of those 10 led blocks and an LM3914.
First led for on, 8 for voltage, one for vape, and an MP3 player (just kidding)

But for real, right selection of resistors in your circuit would give you 8 voltages.

Have fun

certainly would, BUT, can't see myself using that many. It is VERY easily adaptable to do so, for anyone else who might attempt this, just a matter of a bigger block of DIP switches, and a few more resistors.

 

[WillyB]

Average current, total dissipation within the regulator will be small compared to it's capabilities. Minimal heatsinking would be required if the model used had some thermal mass to it, a TO-220, TO-3 might fit in a box mod.
Without a sense return path (r2/r1 some finite value) some uses may not be stable. A high static unswitched resistance value like 1200 ohms (r2/r1 =10) might be an advantage as it would also group the switched voltages (8 of them with 3 switches) a little closer.

The 1200 ohm static r2 would give max output (battery - drop across regulator).
r2 values like maybe 610, 670, and 730 would give several output voltages close to:
6.0, 5.75, 5.5, 4.3, 4.0, 3.9, 3.25. Maybe a little tweeking on the r values would give better values. Someone try 630 ohms, 670 ohms, and 710 ohms for the 3 r2 values (don't forget the parallel 1200 ohm static r2 resistor).
My brain is tired Hope someone checks my math here

The LM317K and other adjustable regulators, some with even lower voltage drops could be used with a dip switch programming resistor network. I have some 10 switch dip switches, wow

Trying to find some real specs on dropout voltage is tough. Seems a R2 value (R1 @120) of 800Ω will tap these out. Various folks list it in excess of 1.5V, as high as 2.5V (the spec sheet, figure 3, seems to confirm this). To get 6V out of these your batteries would have to at about 8+V, 6V output seems impossible, even 5V seems like a struggle (a very small window). These regs seem better suited for 12Vin power supplies like in computers, then you have the headroom to power 6V devices.

The TI283/383 regs at the current we need have a drop out voltage of only .5V (.6v @ 3A)

UCC283T-ADJ Texas Instruments Low Dropout (LDO) Regulators

These regs look interesting with only a 0.34V drop @ 3A

Digi-Key - LT1764EQ#PBF-ND (Manufacturer - LT1764EQ#PBF)

Both should handle our 2.4A needs (Joye 510) without a heat sink as long as they aren't buried under wires and epoxy. The LM317T will be running over spec and will definitely need a HS.

The Fairchild LM317T is priced right @ 57¢, but doesn't seem up to the task.

 

[o4_srt]

Trying to find some real specs on dropout voltage is tough. Seems a R2 value (R1 @120) of 800Ω will tap these out. Various folks list it in excess of 1.5V, as high as 2.5V (the spec sheet, figure 3, seems to confirm this). To get 6V out of these your batteries would have to at about 8+V, 6V output seems impossible, even 5V seems like a struggle (a very small window). These regs seem better suited for 12Vin power supplies like in computers, then you have the headroom to power 6V devices.

The TI283/383 regs at the current we need have a drop out voltage of only .5V (.6v @ 3A)

UCC283T-ADJ Texas Instruments Low Dropout (LDO) Regulators

These regs look interesting with only a 0.34V drop @ 3A

Digi-Key - LT1764EQ#PBF-ND (Manufacturer - LT1764EQ#PBF)

Both should handle our 2.4A needs (Joye 510) without a heat sink as long as they aren't buried under wires and epoxy. The LM317T will be running over spec and will definitely need a HS.

The Fairchild LM317T is priced right @ 57¢, but doesn't seem up to the task.

as stated before, straight from the data sheet, the LM317T will provide 2.2 amps if Vin-Vout is less than 20 volts, which it is in our application. Also, specs given are for constant operation, not 3 seconds here, 5 seconds there, like our application. The LM317T will be more than adequate. I've confirmed with several engineers at work.


I doubt that fairchild would put out a faulty SPICE model for their parts, and it's DC, not a whole lot of fluctuation in usage. If it would be RF, it would be a different story, but DC is pretty cut and dry.


in addition, had I not been able to get the LM317T for FREE, I probably would go with an LM350, rated at 3 amps. But since I only use cartos with 3.2(ish) ohms of resistance, I am sitting at 300 mA away from Imax at 6.1 volts.

 

[WillyB]

I've confirmed with several engineers at work.

Have you confirmed with these engineers the voltage drop? I may be reading it wrong, but at high amp loads this reg seems very inefficient.

The chart doesn't go high enough for your application (the drop seems to increase with amp draw). But looking at 1.5A we see the differential of in to out to be at least 2.5V. Assuming your batts are supplying 7.2V (that sounds about right for loaded volts) subtracting the 2.5V drop from the chart leaves only 4.7V, tops/max, coming out of the reg, but that's with freshly charged batts, within a few minutes that voltage will be dropping in the same manner that an 18650 or 14500 (mAh related) unregulated, naked mod would.

Contrast that to a standard 5V TI reg. It will provide a constant 5V till the batts drop to about 5.5V.

I maybe wrong.

Good luck with your project.

 

[Rocketman]

Yeah, but I still like the dip switch idea for 8 voltages.
A low dropout regulator would allow you to set 8 voltages with only 5 total resistors.
With a stack of 3.7 volt batteries, you could waste anywhere from 0.5 volts to about 3.7 volts and get outputs from 6.9 all the way down to 3.7.
At the lowest setting you could use almost half of your watt hours to vape with
Still like the dip switch idea though.