OT: Jfet constant current source.

[Shafto]

I'm using a picaxe to control the sequencing of LED candles from the dollar store, but they aren't quite bright enough. I'd like to build some LED candles that run an LED at about 100mA as compared to the usual 20mA.

I have some LEDs that can take 100mA and have a Vf of around 3V. My source voltage is from a rechargeable Li-Ion battery with a nominal voltage of 3.7V. If I just use a resistor to drive the LED I'll get a large current change as the battery depletes and the voltage changes.

I've been reading about using a Jfet as a current source, as I'd like to keep things cheap and simple, as I'll be building around 20 of these. I've been reading about the use of a Jfet as a current source, and I think it may be difficult with the low battery voltage I'll be using?

Here are a couple of the articles I have been reading:

http://socrates.berkeley.edu/~phylabs/bsc/PDFFiles/bsc4.pdf

http://www.vishay.com/docs/70596/70596.pdf

So I think I have the basic understanding down. The negative voltage developed over the source resistor is fed back to the gate to "close" the Jfet more as current increases, as current decreases the voltage will rise closer to zero, which will open the Jfet to more current flow.

So, I have found some Jfets with an Idd of 200mA:
http://www.fairchildsemi.com/ds/J1/J105.pdf

Using the formula provided for basic source biasing on the first page of the vishay document:

-4V Gs(off)
0.1A Id
0.2A Idd
2.0K

I end up with 11.7 ohms.

This however, doesn't make sense to me, since no current flows through the gate of the Jfet, it all must go through the 11.7 ohm resistor, but if I calculate what resistor I'd need to run the LED with no Jfet I end up with:

3.7V supply, 3Vf drop, and 100mA = 7ohms.

So I can't be using more resistance on the Jfet circuit, doesn't make sense to me, where am I misunderstanding? Is my battery voltage just not high enough to make this work properly? If that's the case should I look into cascading Jfets or running two with a lower Gs(off) in parallel?

Or maybe there's a much more simple way to achieve a 100mA current source from a 3.7V Li-Ion cell?

Thanks a bunch for any help offered.

 

[JimPerry]

You probably won't get 100mA for very long - the calculated 7R resistor is to limit current - try connecting a disposible LED over a disposible battery with a 100mA diode in series (1n4001 or similar) and see what happens

 

[Shafto]

You misunderstand my intentions Jim. The 100mA diode is the LED.

Do you have any information on how to use a Jfet as a 100mA current source with a battery voltage of 3.7V and an LED voltage of 3V? That's what I'm trying to do.

 

[JimPerry]

Thge 1N4001 will drop 600mV giving you 3V1 over the LED - which it may just take - no JFET required. Similar to the way Stickie LED toys are made

 

[Shafto]

using a diode in series with the LED to drop the voltage would essentially be the same thing as using a resistor only, which I would prefer not to do because of how the current will change as the battery depletes and voltage drops. That's why I'm after a constant current source with a Jfet. I'm ordering a couple to play around with but I was hoping someone might have some insight to help me along.

 

[inglewoodpete]

I have many hours of work in recent months working with MOSFET current controllers for LED lighting.

A good, simple current regulator is described here. I have made both N-channel and P-channel versions of it and used a PICAXE to provide dimming using PWM. It uses MOSFETs rather than JFETs.

 

[Jeremy Harris]

The really big problem here is that you only have about 0.7V to play with across whatever current limiting circuit you choose to use and you only have a maximum supply voltage of 3.7V.

Using MOSFETs is out, as you don't have anything like enough voltage available to provide enough gate drive. The circuits that are linked to above probably won't work with a total supply of only 3.7V because of this (MOSFETs generally need a Vg of more than 3 or 4V to turn on).

Similarly I can't see the classic JFET circuit working well either, for the same reason. Most JFETs need more than 3V to turn off (the example you quoted needs 4V).

I've used an LM334Z for something similar in the past, but even it only works down to a voltage differential of about 0.8V, so not low enough for your application. It would also need an external pass transistor to increase the current handling capability in this application.

I think you're really going to have to look at increasing the supply voltage, as with it as low as it is now I can't see a way you can do this current limiting very simply. There are more complex solutions, for example you could use an 08M2, an external transistor switch and a low value sense resistor and simple filter to build a closed loop PWM current regulator, but that's probably more complex than you are looking for. You could also look at using a simple switcher as a boost current regulator, but again it would add complexity.

 

[BeanieBots]

I fully agree with the comments made by Jeremy Harris but I think there might be some hope for the design posted by inglewoodpete. Certainly worth a try but as the voltage headroom is very small, expect to see significant variation with tempertature.

 

[AllyCat]

Hi,

What about good old bipolars (e.g. a BC337)? You could replace the FET in the circuit above with a bipolar, but I'd try a single NPN with about 3 ohms in its emitter (i.e. 300 mV drop). Then the base needs to be biassed at about 1 volt, for example using a normal silicon and a Schottky diode in series. Or you might find a red LED which drops about a volt at a few milliamps (which has a possible advantage of compensating the negative tempco of the "reference" against the VBE), or use another NPN in a "VBE multipler" configuration.

The transistor will probably require about 1mA of base current so you may need to "tweak" the value of the "reference" pull-up resistor a little to give the desired 100 mA collector current.

Or for greater precision (and lower voltage drop), a 1 ohm resistor monitored by a single-rail Op-Amp or comparator (perhaps even one in the PICaxe) driving an NPN series control element.

Cheers, Alan.

 

[fernando_g]

For a JFET to work in CC mode, one must operate it above the linear region into the saturated region. There must be sufficient Vds and Vgs differential to reach that region.

 

[AllyCat]

I'm using a picaxe to control the sequencing of LED candles from the dollar store, but they aren't quite bright enough. I'd like to build some LED candles that run an LED at about 100mA as compared to the usual 20mA.

Hi,

How many LEDs controlled by each PICaxe?

If not too many, then perhaps just use the method I devised for my "battery tester" (which measures Voltage/Time at a 100 mA constant drain from any single-cell battery). Sorry, it's not formally documented yet as it needs some more elegant logging routines.

Basically just an NPN (BC337) with a few ohms (as available) in its emitter, and the base driven from a PWM pin via a LP filter of say 1 kohms, 22 uF. Then a PICaxe A/D pin connected to the emitter monitors the current with a software loop adjusting the PWM for the correct current. 10 bit A/D with a 2 volt (FVR) reference can resolve to within about 1 mA. The saturation voltage of the NPN should be only a few hundred mV at 100 mA, so adding the resistor drop, the constant current into the collector (i.e. through your LED) could be maintained down to half a volt, or less, above ground.

Cheers, Alan.

 

[Shafto]

I've ordered some JFETs to play with, I might have to weed out the ones with a higher Gs(off) and stick to the lower ones to work properly though, I'm not sure, I suppose I'll see.

Something like the AMC7135 would be perfect; a low loss simple and cheap constant current regulator that works at low voltage. Only problem is that it's set to 350mA and I need less than that per unit. I'm even considering a 555 PWM circuit to drop the average current of an AMC7135 chip to 100mA, but this is more complex than I'd prefer. It seems like there should be an easier way to achieve what I'm after, but I'm not sure what route to take.

I've also looked into low loss linear voltage regulators that can be used as current regulators, like the LM317, but much lower voltage drop. I haven't been able to find anything suitable yet.

I'm hoping to find a simple solution to a low voltage linear constant current source of 100mA, if I have to go more complex I will, but I'd really prefer to keep it as simple as possible.

Here's the AMC7135:

http://pdf1.alldatasheet.com/datasheet-pdf/view/202788/ADDTEK/AMC7135.html