Hello all. I am looking for an N channel mosfet or NPN bjt for a capacitance multiplier with a very low voltage drop. I am working with some Lithium iron batteries and I would like to clean up their noise a bit to make them perform a bit more like Nicad or Alkaline batteries without sacrificing power density. The circuit is drawing 650ma, and I am having a tough time finding decent parts.
So far I have found the Si5442DU which drops about 0.8 volts due to the limits of Vgs, but I would ideally like to be closer to 0.5v or lower. I'm not even sure that is possible at this point though.
Any thoughts would be appreciated.
So far I have found the Si5442DU which drops about 0.8 volts due to the limits of Vgs, but I would ideally like to be closer to 0.5v or lower. I'm not even sure that is possible at this point though.
Any thoughts would be appreciated.
The lowest Vgs threshold I've come across in a MOSFET is from Rohm's RYC002N05 but its such a small package (SOT-23) it would never handle the current you're looking for. I rather think 0.5V is too tall an order for a MOSFET. Try a low saturation voltage bipolar from Diodes or Nexperia.
ixys makes depletion mode fets with 64mohms, 100V 16A, to247 , -4V turn off, IXTH16N10D2
Yeah, I have looked at a bunch of BJTs. The 0.7v vbe is a problem. I also don't have a problem paralleling a bunch of parts together so long as they are cheap enough.
Yes, I would love to be able to use a depletion mode mosfet or better yet a Jfet.
Unfortunately those parts are either very costly, or are no longer being made. IXYS, panasonic, and a few others use to have a decent selection of reasonable parts. Now they only have very expensive high power parts or very low power parts.
The one depletion mode mosfet that would work is CPC3701CTR (I could parallel a bunch of them together). Sadly it is no longer being produced. If it is my only option I may buy a few hundred, but I'm hoping there is a better -more sustainable- solution.
I doubt it, these are typically very high current. But since lithium ion batteries have intercalation in the electrodes, perhaps that is a noise generating process, or at least a source of excess noise (resistance noise).
Just buy a cheap, low current, isolated DC-to-DC converter and connect it up to create (Vin + 12V). Use that to drive the gate of a conventional enhancement mode MOSFET. Here are a bunch of them at (mouser.com).
The schematic below shows a high current, low drop switch; you can convert it to a capacitance multiplier with a little cleverness. The big idea is to have an auxilliary supervoltage supply that exceeds the enhancement mode MOSFET's required gate-to-source voltage.
_
The schematic below shows a high current, low drop switch; you can convert it to a capacitance multiplier with a little cleverness. The big idea is to have an auxilliary supervoltage supply that exceeds the enhancement mode MOSFET's required gate-to-source voltage.
_
Attachments
Last edited:
LDO linear regulators use a pass transistor operating in common emitter mode, not common collector. So, for a positive regulator, you want a PNP or P-channel MOSFET. Of course, the circuit is a bit more complicated, ie CFP with gain. If you have a second ~voltage-doubler rectifier then you have a source of bias above the primary raw DC and then any N-channel common collector/drain will do. As others have said, if you only want 650mA then a LDO chip would be a good choice, unless you want more than 40V, but then you probably would not care about the voltage drop. You may also want to look at buck-boost ~battery management chips.
You can use a Vishay VO1263AA optocoupler to bias the gate more positive than the source, I saw this idea in EDN a few years ago. (You can even make a current controlled high voltage power supply this way.)
If you allow the use of a transformer, you can build an electronic filter having close to zero dropout, independent of the devices used:
https://www.edn.com/smart-ripple-canceller-offers-near-zero-dropout/
If you use opposite polarity transistors and reverse diodes and Ecaps, it can also work for the negative side
https://www.edn.com/smart-ripple-canceller-offers-near-zero-dropout/
If you use opposite polarity transistors and reverse diodes and Ecaps, it can also work for the negative side
These are both super neat ideas. Both give me the ability to filter or otherwise condition the voltage going to the gate/base of the pass device.
On second thought, that's not going to work out too well, at least not in its simplest terms.
Germanium transistors have approx. .32 Volt loss. As seen in the 'link', there are powerful ones.
https://www.nteinc.com/semiconductors/Germanium.php
https://www.nteinc.com/semiconductors/Germanium.php
Based on what Mark Johnson recommended, I wanted to get your guy's thoughts about a charge pump converter?
I found this one on JLPCB for ~10 cents. https://jlcpcb.com/partdetail/Shouding-SD6210A/C250809
No doubt the noise on the output will be pretty high so it will have to be filtered somewhat. But In theory I should be able to use the output voltage as rail voltage for a low noise - high psrr opamp to feed the mosfet.
I have never worked with a charge pump before, so I am curious if there are any gotchas.
I found this one on JLPCB for ~10 cents. https://jlcpcb.com/partdetail/Shouding-SD6210A/C250809
No doubt the noise on the output will be pretty high so it will have to be filtered somewhat. But In theory I should be able to use the output voltage as rail voltage for a low noise - high psrr opamp to feed the mosfet.
I have never worked with a charge pump before, so I am curious if there are any gotchas.
Charge pump converters in integrated form (ICL7660 started the genre I believe) tend to only exist in low supply voltages (<= 10V). For higher rails you'll likely need to roll your own. Boost converters are an option to consider, in contrast to charge pumps they have inductors which need to be screened or segregated from sensitive circuits.