EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Peabody on August 14, 2024, 01:56:10 pm
-
I'm trying to understand how a constant current source can be designed using a depletion-mode N-channel JFET (not a MOSFET). I don't have any JFETs on hand at the moment, and I ran across a video suggesting I should add a resistor directly between gate and source to make the current source stable. I can't find this idea anywhere else. Can someone explain what he's doing, and what this is all about?
https://www.youtube.com/watch?v=mG2Gc02030c (https://www.youtube.com/watch?v=mG2Gc02030c)
-
If you connect the gate to the source, either directly or through a resistor, the drain current will equal Idss so long as the drain-source voltage is sufficient.
All JFETs are depletion-mode, most MOSFETs are enhancement mode, but there are a few depletion-mode MOSFETs available from IXYS (now owned by Littelfuse).
-
He gets the odd behaviour with the gate floating. it will than pick up charge from leakage and rectified hum.
With the floating gate one can end up with a positive gate voltage and this way get more than Idss - possibly too much for the LED or the FET.
For practical purpose it would be a good idea to have always some 10 K in series with the gate for such tests to avoid too much gate current (e.g. with the power supply reversed). For switching purposes there is no low resistance needed from gate to source. If there is little hum, some 1 M are OK and 10 K would already allow relatively fast switching with a capacitance of the 10 pF.
There are quite a few depl. mode FETs available, also for higher voltages and more current than the usual JFETs, not just from IXYS, but also from Infineon (e.g. BSS129). However the MOSFETs have the internal diode and are thus not symmetric.
-
..I can't find this idea anywhere else. Can someone explain what he's doing, and what this is all about?
He is collecting likes and subscribers..
-
Thanks for the responses. So I'm concluding that I don't have to add that resistor, and the current draw will not bounce around like he shows. In other words, he's wrong.
-
One needs a defined gate voltage to get a stable current. If the gate is free to float and pick up noise or hum, the current will follow that. So he is not really wrong, but not making clear that it is the floating gate that causes his unstable current. So he kind of misses the point that different from the base of a BJT a not connected gate is not a good idea - it can store the previous state for quite some time to give kind of random / noisy / drifting results.
How a defiend gate voltage is realized does not matter, it does not have to be a gate to source resistor.
-
Or get the two-pin version, called CLD: https://en.wikipedia.org/wiki/Constant-current_diode
-
Or get the two-pin version, called CLD: https://en.wikipedia.org/wiki/Constant-current_diode
If you can find them. The only manufacturer I know of these days is Linear Systems in California. And they're not cheap.
-
From https://www.linkedin.com/pulse/mapping-current-limiting-diodes-market-trends-challenges-bngze: (https://www.linkedin.com/pulse/mapping-current-limiting-diodes-market-trends-challenges-bngze:)
- Microsemi
- Caloric
- Central Semiconductor
- Infineon
- Maxim Integrated
- Micro Commercial
- Nexperia
- ON Semiconductor
- Semtech
- Vishay
-
Anyone else hate idiots posting Youtube videos of themselves randomly plugging stuff into a breadboard without a schematic in sight? ::)
As far as I can see, it's the old magazine floating gate 'electrostatic field detector'.
-
Anyone else hate idiots posting Youtube videos of themselves randomly plugging stuff into a breadboard without a schematic in sight? ::)
As far as I can see, it's the old magazine floating gate 'electrostatic field detector'.
Oh, yes indeed.
I thought of posting something along the lines of 99.99% of yoootoob talking head vids are a waste of time, and, in order to prove that for any one video, you have to watch all of it. Much better to short-circuit that and just ignore them all.
A key skill nowadays is learning how to realise quickly that something should be ignored. The other key skill is unchanged: knowing how to search for worthwhile sources.
-
LM334. done 3 pin current source. set current using a resistor.
-
and this one : https://www.ti.com/lit/an/snoaa46/snoaa46.pdf (https://www.ti.com/lit/an/snoaa46/snoaa46.pdf)
lots of current sources.
-
From https://www.linkedin.com/pulse/mapping-current-limiting-diodes-market-trends-challenges-bngze: (https://www.linkedin.com/pulse/mapping-current-limiting-diodes-market-trends-challenges-bngze:)
- Microsemi
- Caloric
- Central Semiconductor
- Infineon
- Maxim Integrated
- Micro Commercial
- Nexperia
- ON Semiconductor
- Semtech
- Vishay
Did you check? I know most of those manufacturers, and NO, they do not make constant current diodes. What they do make are constant current ICs, which is a different story. Check it yourself, before posting some LinkedIn commercial.
-
Here are the equations for the simple JFET current source from "Current Sources and Voltage References - A Design Reference for Electronics Engineers - Linden T. Harrison".
-
One argument for having the gate resistor, I have blown up a few JFET's in CCS configuration, that was at high voltages 350V. edit: IXCP 10M45S (https://www.mouser.com/datasheet/2/205/98704-13273.pdf) 450V 2-100mA but I think it's a N-ch depletion mosfet, not a JFET. Fix was adding a reverse-diode across it.
Their (N-ch) weakness is reverse-polarity transients flow back into the gate and can damage it, so the resistor current-limits and protects the diode junction. It's not a mosfet with glass insulator to the gate. Or include a reverse-diode across the JFET.
In 1975 as a kid I used 2N3819 and had a blast with it, small wire on the gate and LED+Speaker on the output, it could pick up a person about 6' away.
-
TAoE x-chapters shows how a cascode increases the output impedance, how to stack depletion MOSFETs/JFETs to increase voltage up to several kV, and how to control power dissipation.
-
Did you check? I know most of those manufacturers, and NO, they do not make constant current diodes. What they do make are constant current ICs, which is a different story. Check it yourself, before posting some LinkedIn commercial.
these exist and are available :
Semitec US E-5xx series
Microchip !n5406
Diotec CL20 series
Linear systems J50x
Central Semi CMJ / CMJH series
Interfet
https://www.mouser.com/c/semiconductors/discrete-semiconductors/diodes-rectifiers/current-regulator-diodes/?srsltid=AfmBOopuDl-LNBAAOu-Ht4d0Qou5kTUN0A4-rjF_6GTteoc1eJM_wgHo (https://www.mouser.com/c/semiconductors/discrete-semiconductors/diodes-rectifiers/current-regulator-diodes/?srsltid=AfmBOopuDl-LNBAAOu-Ht4d0Qou5kTUN0A4-rjF_6GTteoc1eJM_wgHo)
https://www.digikey.com/en/products/filter/current-regulation-diodes-transistors/1164?s=N4IgjCBcoLQExVAYygFwE4FcCmAaEA9lANogCsIAugL74wCciIKkGO%2BRkpEN11QA (https://www.digikey.com/en/products/filter/current-regulation-diodes-transistors/1164?s=N4IgjCBcoLQExVAYygFwE4FcCmAaEA9lANogCsIAugL74wCciIKkGO%2BRkpEN11QA)
-
Did you check? I know most of those manufacturers, and NO, they do not make constant current diodes. What they do make are constant current ICs, which is a different story. Check it yourself, before posting some LinkedIn commercial.
these exist and are available :
Semitec US E-5xx series
Microchip !n5406
Diotec CL20 series
Linear systems J50x
Central Semi CMJ / CMJH series
Interfet
Nice list, thanks.
1N5406 must be a mistake, but the rest are kosher.
Especially Interfet looks interesting for someone working with JFETs,
-
I feel like reading the first few google results for "jfet current source" and then coming back for clarification would be useful:
https://www.vishay.com/docs/70596/70596.pdf (https://www.vishay.com/docs/70596/70596.pdf)
https://www.electronics-tutorials.ws/transistor/fet-current-source.html (https://www.electronics-tutorials.ws/transistor/fet-current-source.html)
https://www.electronicdesign.com/technologies/power/article/21808012/whats-all-this-jfet-constant-current-stuff-anyhow (https://www.electronicdesign.com/technologies/power/article/21808012/whats-all-this-jfet-constant-current-stuff-anyhow)
-
Yes, I had read AN103 and found several good videos, and thought I understood it. But this particular video made me question whether I really understood it. It turns out I did understand it after all.
-
Here are the equations for the simple JFET current source from "Current Sources and Voltage References - A Design Reference for Electronics Engineers - Linden T. Harrison".
GOOD BOOK ! Added to the collection
-
I remember reading that book and disliking quite several things about it. Off the top of my head:
VBE(sat) voltage directly affects gain, because it determines the base current. Too low a base voltage will not permit enough base current to flow, which will in turn restrict the amount of collector current. Too high a base-emitter voltage can destroy the transistor, although most have an absolute maximum rating of several volts. This specification ranges from about 3 volts for some devices up to about 6 volts for others; the 2N3904 is rated for up to 6 volts, while the 2N3906 is rated for up to 5 volts. (Darlingtons have a higher absolute maximum rating often reaching 10 volts, but they are not really suitable as potential current sources.)
Leaving aside the fact that anyone who talks "Vbe controls base current which controls collector current" appears to have a fairly sloppy understanding of how things work, the author clearly confuses Vbe(sat) with junction reverse breakdown rating. Horowitz or Hill he is not :P
(Not sure what's the deal with Darlingtons not being suitable for current sources either.)
-
I also have the Linden T Harrison book and have mixed feelings about it. The relentless flagwaving irritates and he's clearly not seriously technical. But he clearly has technical people around him. Perhaps he's a mananger in a technical environment. Despite my reservations, the book makes some good points and I have acted on tips in there. For instance, rather than temperature compensating an LM334 with the data sheet's recommended diode, I now use an MPSA42's base-emitter junction. Harrison claims this is a better match to the LM334's behaviour. Maybe it is, maybe it isn't. But what's definite is that you can epoxy the flat face of an MPSA42 to the flat face of an LM334, achieving thermal bonding that is way better than can be achieved using a glass-bodied diode.
-
It's like Wikipedia - a collection of useful information mostly written by others, mixed with dangerous traps for the unwary. Not to be trusted immediately. I mean, try switching 2N3904 with a 5V gate driver and see what happens >:D
-
I also have the Linden T Harrison book and have mixed feelings about it. The relentless flagwaving irritates and he's clearly not seriously technical. But he clearly has technical people around him. Perhaps he's a mananger in a technical environment. Despite my reservations, the book makes some good points and I have acted on tips in there.
I think it is one of the better books with a good compromise between math and practical examples.
For instance, rather than temperature compensating an LM334 with the data sheet's recommended diode, I now use an MPSA42's base-emitter junction. Harrison claims this is a better match to the LM334's behaviour. Maybe it is, maybe it isn't. But what's definite is that you can epoxy the flat face of an MPSA42 to the flat face of an LM334, achieving thermal bonding that is way better than can be achieved using a glass-bodied diode.
Any small signal high conductance low leakage non-gold doped silicon diode should work well. When the LM334 was first released, this diode was the 1N457 and it is still readily available. The 1N458 or FDH300 should also work well. A base-emitter junction of a non-gold doped small signal transistor makes an excellent high conductance low leakage diode, albeit at low voltage, and like you mention, it is easier to attach for better thermal tracking. Little s-clips or aluminum blocks used to be available for thermally bonding TO-92 and similar small packages.
-
i use the 1n457 with lm334 to stabilize reference zeners.