current regulator

Started by John Larkin May 30, 2018
https://www.diodes.com/assets/Datasheets/AL5890.pdf

The block diagram shows this as a closed-loop regulator, not just some
jfet sort of thing.

The high minimum voltage and unspecified capacitance/speed limits its
use, but it is cute.



-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

On 05/30/2018 11:23 AM, John Larkin wrote:
> https://www.diodes.com/assets/Datasheets/AL5890.pdf > > The block diagram shows this as a closed-loop regulator, not just some > jfet sort of thing. > > The high minimum voltage and unspecified capacitance/speed limits its > use, but it is cute.
Fun. Might be a good safety feature for George's APD supply if it's quick enough. It's hard to get into much trouble with 10 mA, and that's way more than an APD needs. Mine uses LND150 depletion MOSFETs because I want a lower current limit. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On Wed, 30 May 2018 12:09:32 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>On 05/30/2018 11:23 AM, John Larkin wrote: >> https://www.diodes.com/assets/Datasheets/AL5890.pdf >> >> The block diagram shows this as a closed-loop regulator, not just some >> jfet sort of thing. >> >> The high minimum voltage and unspecified capacitance/speed limits its >> use, but it is cute. > >Fun. Might be a good safety feature for George's APD supply if it's >quick enough. It's hard to get into much trouble with 10 mA, and that's >way more than an APD needs. Mine uses LND150 depletion MOSFETs because >I want a lower current limit. > >Cheers > >Phil Hobbs
DN2530 with a source resistor is pretty good too. I have values. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Wednesday, May 30, 2018 at 12:09:37 PM UTC-4, Phil Hobbs wrote:
> On 05/30/2018 11:23 AM, John Larkin wrote: > > https://www.diodes.com/assets/Datasheets/AL5890.pdf > > > > The block diagram shows this as a closed-loop regulator, not just some > > jfet sort of thing. > > > > The high minimum voltage and unspecified capacitance/speed limits its > > use, but it is cute. > > Fun. Might be a good safety feature for George's APD supply if it's > quick enough. It's hard to get into much trouble with 10 mA, and that's > way more than an APD needs. Mine uses LND150 depletion MOSFETs because > I want a lower current limit.
Way too much current for the Spad. The spec sheet says ~200uA CW I've got an lnd150 and 10k ohm (G-S) for 100uA. (It's a few hundred volts, 100V @ 100 uA ~10 mW... it might take more than that but who wants to blow up a $250 part, I've been afraid to test the current limit until I make more measurements.) George H.
> > Cheers > > Phil Hobbs > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC > Optics, Electro-optics, Photonics, Analog Electronics > > 160 North State Road #203 > Briarcliff Manor NY 10510 > > hobbs at electrooptical dot net > http://electrooptical.net
On 05/31/2018 08:30 AM, George Herold wrote:
> On Wednesday, May 30, 2018 at 12:09:37 PM UTC-4, Phil Hobbs wrote: >> On 05/30/2018 11:23 AM, John Larkin wrote: >>> https://www.diodes.com/assets/Datasheets/AL5890.pdf >>> >>> The block diagram shows this as a closed-loop regulator, not just some >>> jfet sort of thing. >>> >>> The high minimum voltage and unspecified capacitance/speed limits its >>> use, but it is cute. >> >> Fun. Might be a good safety feature for George's APD supply if it's >> quick enough. It's hard to get into much trouble with 10 mA, and that's >> way more than an APD needs. Mine uses LND150 depletion MOSFETs because >> I want a lower current limit. > Way too much current for the Spad. The spec sheet says ~200uA CW > I've got an lnd150 and 10k ohm (G-S) for 100uA. (It's a few hundred > volts, 100V @ 100 uA ~10 mW... it might take more than that but who wants > to blow up a $250 part, I've been afraid to test the current limit until > I make more measurements.) >
I was talking about the safety current limit, not the device protection one. Unless I'm mistaken, UL wants the gizmo to be safe even when it's broken. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On 05/30/2018 07:59 PM, John Larkin wrote:
> On Wed, 30 May 2018 12:09:32 -0400, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 05/30/2018 11:23 AM, John Larkin wrote: >>> https://www.diodes.com/assets/Datasheets/AL5890.pdf >>> >>> The block diagram shows this as a closed-loop regulator, not just some >>> jfet sort of thing. >>> >>> The high minimum voltage and unspecified capacitance/speed limits its >>> use, but it is cute. >> >> Fun. Might be a good safety feature for George's APD supply if it's >> quick enough. It's hard to get into much trouble with 10 mA, and that's >> way more than an APD needs. Mine uses LND150 depletion MOSFETs because >> I want a lower current limit. >> >> Cheers >> >> Phil Hobbs > > DN2530 with a source resistor is pretty good too. I have values. > >
I bought a bunch of depletion MOSFETs recently, including some of those. Useful gizmos. The DN2530's specs are pretty loosey-goosey--Vdg_off is -1 to -3.5V, and there's only a minimum for transconductance and a maximum I_DSS. Do you find that it's consistent enough in real life to make a good current limiter? The On Semi 120V, 20 mA model has +-15% tolerance on I_DSS, and looks like about 35 ohms at room temperature, which wouldn't be bad for input protection of lowish-noise amps. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On Thursday, May 31, 2018 at 9:19:29 AM UTC-4, Phil Hobbs wrote:
> On 05/31/2018 08:30 AM, George Herold wrote: > > On Wednesday, May 30, 2018 at 12:09:37 PM UTC-4, Phil Hobbs wrote: > >> On 05/30/2018 11:23 AM, John Larkin wrote: > >>> https://www.diodes.com/assets/Datasheets/AL5890.pdf > >>> > >>> The block diagram shows this as a closed-loop regulator, not just some > >>> jfet sort of thing. > >>> > >>> The high minimum voltage and unspecified capacitance/speed limits its > >>> use, but it is cute. > >> > >> Fun. Might be a good safety feature for George's APD supply if it's > >> quick enough. It's hard to get into much trouble with 10 mA, and that's > >> way more than an APD needs. Mine uses LND150 depletion MOSFETs because > >> I want a lower current limit. > > Way too much current for the Spad. The spec sheet says ~200uA CW > > I've got an lnd150 and 10k ohm (G-S) for 100uA. (It's a few hundred > > volts, 100V @ 100 uA ~10 mW... it might take more than that but who wants > > to blow up a $250 part, I've been afraid to test the current limit until > > I make more measurements.) > > > > I was talking about the safety current limit, not the device protection > one. Unless I'm mistaken, UL wants the gizmo to be safe even when it's > broken.
Oh right, something like 10 mA might be perfect for that. I haven't thought that far our yet. Heck at the moment I'm using a hunking linear HV supply, and haven't made the HV power supply.... Lot's to do. George h.
> > Cheers > > Phil Hobbs > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC > Optics, Electro-optics, Photonics, Analog Electronics > > 160 North State Road #203 > Briarcliff Manor NY 10510 > > hobbs at electrooptical dot net > http://electrooptical.net
On Thursday, May 31, 2018 at 10:45:04 AM UTC-4, Phil Hobbs wrote:
> On 05/30/2018 07:59 PM, John Larkin wrote: > > On Wed, 30 May 2018 12:09:32 -0400, Phil Hobbs > > <pcdhSpamMeSenseless@electrooptical.net> wrote: > > > >> On 05/30/2018 11:23 AM, John Larkin wrote: > >>> https://www.diodes.com/assets/Datasheets/AL5890.pdf > >>> > >>> The block diagram shows this as a closed-loop regulator, not just some > >>> jfet sort of thing. > >>> > >>> The high minimum voltage and unspecified capacitance/speed limits its > >>> use, but it is cute. > >> > >> Fun. Might be a good safety feature for George's APD supply if it's > >> quick enough. It's hard to get into much trouble with 10 mA, and that's > >> way more than an APD needs. Mine uses LND150 depletion MOSFETs because > >> I want a lower current limit. > >> > >> Cheers > >> > >> Phil Hobbs > > > > DN2530 with a source resistor is pretty good too. I have values. > > > > > > I bought a bunch of depletion MOSFETs recently, including some of those. > Useful gizmos. > > The DN2530's specs are pretty loosey-goosey--Vdg_off is -1 to -3.5V, and > there's only a minimum for transconductance and a maximum I_DSS.
I bought DN2540's -1.5 to -3.5V.. but that's no different than the lnd150's spread. Have you measured any? Hmmm if I get a batch of lnd150's with closer to -3V than whatever I'm seeing now that could be a problem. I'll have to measure some... I'm mostly machining today. I found this 2-56 gun tap in our drawer. It's really slick, shoots the chip out the far end of the hole, rather than having to work the tap in and out. (Wouldn't be so good for a blind tapped hole though.) George H.
> > Do you find that it's consistent enough in real life to make a good > current limiter? > > The On Semi 120V, 20 mA model has +-15% tolerance on I_DSS, and looks > like about 35 ohms at room temperature, which wouldn't be bad for input > protection of lowish-noise amps. > > Cheers > > Phil Hobbs > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC > Optics, Electro-optics, Photonics, Analog Electronics > > 160 North State Road #203 > Briarcliff Manor NY 10510 > > hobbs at electrooptical dot net > http://electrooptical.net
On 05/31/2018 01:12 PM, George Herold wrote:
> On Thursday, May 31, 2018 at 10:45:04 AM UTC-4, Phil Hobbs wrote: >> On 05/30/2018 07:59 PM, John Larkin wrote: >>> On Wed, 30 May 2018 12:09:32 -0400, Phil Hobbs >>> <pcdhSpamMeSenseless@electrooptical.net> wrote: >>> >>>> On 05/30/2018 11:23 AM, John Larkin wrote: >>>>> https://www.diodes.com/assets/Datasheets/AL5890.pdf >>>>> >>>>> The block diagram shows this as a closed-loop regulator, not just some >>>>> jfet sort of thing. >>>>> >>>>> The high minimum voltage and unspecified capacitance/speed limits its >>>>> use, but it is cute. >>>> >>>> Fun. Might be a good safety feature for George's APD supply if it's >>>> quick enough. It's hard to get into much trouble with 10 mA, and that's >>>> way more than an APD needs. Mine uses LND150 depletion MOSFETs because >>>> I want a lower current limit. >>>> >>>> Cheers >>>> >>>> Phil Hobbs >>> >>> DN2530 with a source resistor is pretty good too. I have values. >>> >>> >> >> I bought a bunch of depletion MOSFETs recently, including some of those. >> Useful gizmos. >> >> The DN2530's specs are pretty loosey-goosey--Vdg_off is -1 to -3.5V, and >> there's only a minimum for transconductance and a maximum I_DSS. > > I bought DN2540's -1.5 to -3.5V.. but that's no different than the > lnd150's spread. Have you measured any?
All the LND150s I've seen were right around 1.6 mA I_DSS. If the DN2530s are the same way, we can probably use them for our own products, but not in designs for licensing or work-for-hire. The On Semi NSIC2020JB is 20 mA +- 15%, and has a small signal resistance of around 35 ohms at room temperature. For input protection on a high-Z input, it would contribute only about 0.76 nV 1-Hz noise at 300 K, and I could use low-capacitance Schottkys to the supply rails with no worries.
> > Hmmm if I get a batch of lnd150's with closer to -3V than whatever > I'm seeing now that could be a problem. > > I'll have to measure some... I'm mostly machining today. I found this > 2-56 gun tap in our drawer. It's really slick, shoots the chip out the > far end of the hole, rather than having to work the tap in and out. > (Wouldn't be so good for a blind tapped hole though.) >
Fun. Cheers Phil Hobbs (doing Gerbers and proposals today--less fun) -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On Thursday, May 31, 2018 at 4:34:40 PM UTC-4, Phil Hobbs wrote:
> On 05/31/2018 01:12 PM, George Herold wrote: > > On Thursday, May 31, 2018 at 10:45:04 AM UTC-4, Phil Hobbs wrote: > >> On 05/30/2018 07:59 PM, John Larkin wrote: > >>> On Wed, 30 May 2018 12:09:32 -0400, Phil Hobbs > >>> <pcdhSpamMeSenseless@electrooptical.net> wrote: > >>> > >>>> On 05/30/2018 11:23 AM, John Larkin wrote: > >>>>> https://www.diodes.com/assets/Datasheets/AL5890.pdf > >>>>> > >>>>> The block diagram shows this as a closed-loop regulator, not just some > >>>>> jfet sort of thing. > >>>>> > >>>>> The high minimum voltage and unspecified capacitance/speed limits its > >>>>> use, but it is cute. > >>>> > >>>> Fun. Might be a good safety feature for George's APD supply if it's > >>>> quick enough. It's hard to get into much trouble with 10 mA, and that's > >>>> way more than an APD needs. Mine uses LND150 depletion MOSFETs because > >>>> I want a lower current limit. > >>>> > >>>> Cheers > >>>> > >>>> Phil Hobbs > >>> > >>> DN2530 with a source resistor is pretty good too. I have values. > >>> > >>> > >> > >> I bought a bunch of depletion MOSFETs recently, including some of those. > >> Useful gizmos. > >> > >> The DN2530's specs are pretty loosey-goosey--Vdg_off is -1 to -3.5V, and > >> there's only a minimum for transconductance and a maximum I_DSS. > > > > I bought DN2540's -1.5 to -3.5V.. but that's no different than the > > lnd150's spread. Have you measured any? > > All the LND150s I've seen were right around 1.6 mA I_DSS. If the > DN2530s are the same way, we can probably use them for our own products, > but not in designs for licensing or work-for-hire. > > The On Semi NSIC2020JB is 20 mA +- 15%, and has a small signal > resistance of around 35 ohms at room temperature. For input protection > on a high-Z input, it would contribute only about 0.76 nV 1-Hz noise at > 300 K, and I could use low-capacitance Schottkys to the supply rails > with no worries. > > > > > Hmmm if I get a batch of lnd150's with closer to -3V than whatever > > I'm seeing now that could be a problem. > > > > I'll have to measure some... I'm mostly machining today. I found this > > 2-56 gun tap in our drawer. It's really slick, shoots the chip out the > > far end of the hole, rather than having to work the tap in and out. > > (Wouldn't be so good for a blind tapped hole though.) > > > > Fun.
Yeah, I don't get to do it much these days, but I enjoy machining. (turning a hunk of metal into something useful.)
> > Cheers > > Phil Hobbs > (doing Gerbers and proposals today--less fun)
Sorry, there's always scut work, not sure how to classify proposals. Someone's gotta pay the bills. George H.
> > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC > Optics, Electro-optics, Photonics, Analog Electronics > > 160 North State Road #203 > Briarcliff Manor NY 10510 > > hobbs at electrooptical dot net > http://electrooptical.net