Using a P-FET as an "Ideal Diode" - Will this work?

[timb]

Hey guys, so I'm working on a project that's powered by three small, low voltage indoor solar cells. I'm using one of LT's lovely little energy harvesting converters plus a couple of rechargeable coin cells.

Now, I've got that part of the design done, however I want to be able to multiplex the solar cells with USB power. The easy way to do this would be to simply use two diodes, one from the USB port and one from the solar cells (and indeed this is given as an example in the datasheet). However, the particular solar cells I'm using are only 800mV open circuit voltage (and I'm only using three of them) so every mV matters and I'd really like to avoid the voltage drop of even a Schottky diode.

So, I was thinking of using the following arrangement:





I've found some P-channel MOSFETs with a VgsTh of 1V. Basically, when 5V is coming in from the USB port, it would pull the gate up to at least 2.2V, turning it off. When the USB power is disconnected a 1M resistor connected to the gate will pull it down, turning it on and allowing current to flow.

I've simulated this arrangement in LTspice (with this particular MOSFET) and it seems to work just fine. Am I missing anything or should this work just as I intend it to?

Thanks!

[wraper]

Don't forget about parasitic diode in the MOSFET.

[The Soulman]

Why not use a depletion mode fet?

[timb]

Don't forget about parasitic diode in the MOSFET.

Right... I guess I could swap it so the drain is connected to the solar cells? It seems to work and at these current levels I can't imagine it causing any damage...

[timb]

Why not use a depletion mode fet?

I can't seem to find one with a really low Vgs(Th) that's not huge. This will only be pushing uA up to 10mA or so. Something in a SOT-23 or SC70 package (or smaller) would be ideal due to size constraints. (I can find quite a few enhancement mode fets that meet this criteria, which is what I used in my LTspice simulations.)

[Benta]

Depletion-mode FETs are rare, and mostly available for RF applications.

May I suggest a super-primitive idea?

A reed relay with changeover contacts (DT, double throw). Power the coil from the USB port and the NO contact will let USB power flow. If USB power disappears, it'll fall back to NC and transfer power from the solar cells.
This would be the lowest loss solution, I think.

[langwadt]

a really silly idea; a cmos inverter, use Vcc as output, the gate output connected to the solarcells, gate input input connected to Vusb

[mmagin]

Why not use a depletion mode fet?

I can't seem to find one with a really low Vgs(Th) that's not huge. This will only be pushing uA up to 10mA or so. Something in a SOT-23 or SC70 package (or smaller) would be ideal due to size constraints. (I can find quite a few enhancement mode fets that meet this criteria, which is what I used in my LTspice simulations.)

Maybe look at some of the parts Advanced Linear Devices makes?

[timb]

a really silly idea; a cmos inverter, use Vcc as output, the gate output connected to the solarcells, gate input input connected to Vusb

That's actually pretty creative! Quiescent current or leakage might be an issue, but I'll breadboard it tonight and see!

Depletion-mode FETs are rare, and mostly available for RF applications.

May I suggest a super-primitive idea?

A reed relay with changeover contacts (DT, double throw). Power the coil from the USB port and the NO contact will let USB power flow. If USB power disappears, it'll fall back to NC and transfer power from the solar cells.
This would be the lowest loss solution, I think.

Hmm, not a bad idea, actually. Something like this:






However, there are a couple of issues I can see:

I don't have a lot of space available... I think I could find one small enough in the L and W dimensions,  but I might have a hard time finding one that's under 2mm in height! (Though I could get away with 3mm on the bottom of the board).

The solar side of the relay would only be putting out tens to hundreds of microamps (possibly going to ~5mA if someone takes it into the sun). A lot of relays really need quite a bit more current than that to keep their contacts clean.

Though I suppose that really only applies to power relays; a signal relay might not be as vulnerable because a lot of small, low current signal relays have contact resistances in the tens of ohms, which wouldn't really drop the solar side that much (I don't care about the USB side as it's going down to 3.3V anyway).

Still, not a bad idea and worth checking what's available on Digi-Key!

---

So, I did a test last night with a random PFET I had sitting around (a TO-220 IRF part). I hooked it up backwards (with the source towards the higher voltage) and it seemed to work exactly as it did in my simulations.

Barring the availability of a tiny relay or the use of a CMOS inverter, I think I might try this method out on the actual board and see how it works.

Here's my LTspice data using the actual FET I planned on using:

[NiHaoMike]

What about a small linear regulator to keep the supercap charged?

[David Hess]

A low Vgs p-channel MOSFET will work fine.

Using an old CMOS inverter in reverse as described is a 20+ year old trick for doing the same thing.

[SeanB]

How bad would feeding the solar cells with 5V through a resistor and a blocking diode to the 5V USB supply be? If the converter will work off the 5V rail, and the solar cells will simply act as a large area forward biased diode in any case, and as yours are likely totalling around 4v forward drop in any case just use them as a regulator.

[Kleinstein]

the solar cells will not like a much higher voltage from extern - they are just diodes after all. Also the super cap might not like more than 2.5 V.  Depending in the load, the leakage current through the solar cells when dark could be an issue - though with a boost converter, the load current could be much higher anyway.

For low current one could consider a JFET, if you need a depletion mode FET.
Normally the p-mosfet should work OK if put in the right way. One might be able to have the FET at the negative side and than use a N-channel - more choice of low threshold types there.

[vealmike]

http://www.ti.com/product/LM74610-Q1/datasheet

[timb]

http://www.ti.com/product/LM74610-Q1/datasheet

That's a pretty cool part, however there's some potential issues in my application:

Minimum drain current is 1ma (to maintain enough of a voltage drop across the MOSFET's body diode); whereas in my case I might only be pushing a few hundred uA.

It also requires the voltage source to be able to sink current, albeit briefly, like a battery can.

Still, that's a damn clever design!

[exmadscientist]

If you can tolerate the losses (11uA quiescent current and 20mV drop across the external FET), the LTC4412 ideal diode controller is a robust and simple solution. It's marginal at 2.4V though, so perhaps another member of the family such as the LTC4419 would be a better choice. Have a look, there's a few ideal diode choices available from LTC and one or two more from TI. (I don't recall other vendors offering any last time I went looking.)

[Cerebus]

Two thoughts, both a bit 'off-the-cuff' as I have to shoot out and don't have time to think things through fully.

Firstly, what about JFETs? You're only switching a few mA, well within the capabilities of small signal devices (and by implication SOT-23 packages or similar), depletion mode is the norm and they tend to have lower V_GS(OFF)MAX values.

Secondly, have you thought about turning your thinking 'upside down' and switching the supply negative rail? That puts you looking for N-channel devices and gives you a much, much bigger selection to choose from.

[timb]

Two thoughts, both a bit 'off-the-cuff' as I have to shoot out and don't have time to think things through fully.

Firstly, what about JFETs? You're only switching a few mA, well within the capabilities of small signal devices (and by implication SOT-23 packages or similar), depletion mode is the norm and they tend to have lower V_GS(OFF)MAX values.

Secondly, have you thought about turning your thinking 'upside down' and switching the supply negative rail? That puts you looking for N-channel devices and gives you a much, much bigger selection to choose from.

Most of the JFETs I found with low threshold voltages were rated for 2mA or less (and they had insanely high RdsOn). The ones that could do 10 or 20mA had threshold voltages of 5V or higher. This is just from a cursory search on DigiKey, I'll do some more in depth research later tonight.

Switching the ground rail of the solar cells isn't a bad idea, actually! Though, I would need to use a NOT gate or another FET to switch it based on the status of the USB rail... (Though that wouldn't be a problem, TI has single inverter gates that have Iq's of 500nA.

The only downside to doing it that way is I'd need to run a trace from the USB input all the way across the board, which is doable but is going to be a major pain (this is packed pretty tightly).

Still, if this PFET thing doesn't work out it's a very good alternative. Nice outside the box thinking!

If you can tolerate the losses (11uA quiescent current and 20mV drop across the external FET), the LTC4412 ideal diode controller is a robust and simple solution. It's marginal at 2.4V though, so perhaps another member of the family such as the LTC4419 would be a better choice. Have a look, there's a few ideal diode choices available from LTC and one or two more from TI. (I don't recall other vendors offering any last time I went looking.)

Yeah, those were the first things I checked. I'd rather not have the 10uA IQ if I can help it. There was another reason I ruled them out, can't remember what it was though...