Author Topic: High voltage, high side current sense amplifier.  (Read 815 times)

0 Members and 1 Guest are viewing this topic.

Offline Greg Robinson

  • Regular Contributor
  • *
  • Posts: 82
  • Country: au
High voltage, high side current sense amplifier.
« on: December 09, 2017, 07:32:45 pm »
Hi everyone, just after a quick sanity check on this high side current monitor I'm working up for a high voltage circuit. As Pmos devices aren't commonly available in high enough voltage ratings for my application, I've cascoded a couple. Does this look like a workable design?
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 1268
  • Country: gb
Re: High voltage, high side current sense amplifier.
« Reply #1 on: December 09, 2017, 08:26:26 pm »
It looks OK, although you are pulling more current than necessary through R1-R2.

What MOSFET's are you using?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 9823
  • Country: us
  • DavidH
Re: High voltage, high side current sense amplifier.
« Reply #2 on: December 09, 2017, 08:27:33 pm »
That will work but there has to be something better than a string of p-channel MOSFET cascodes.

I see that you are using a resistive bias source.  If the high side voltage is that stable, then why not replace all but one high voltage p-channel MOSFET cascode with stacked zener diodes?

We were discussing this circuit in a different forum thread and I was thinking about how to use PNP transistors while compensating for their bias current but that does not really gain much if anything in this situation since PNP transistors are limited in voltage rating also.
 

Offline Greg Robinson

  • Regular Contributor
  • *
  • Posts: 82
  • Country: au
Re: High voltage, high side current sense amplifier.
« Reply #3 on: December 09, 2017, 09:04:19 pm »
It looks OK, although you are pulling more current than necessary through R1-R2.

What MOSFET's are you using?
Thanks fcb, yes, I am pulling more current than needed through those resistors, I had actually calculated for a cascode of three pmos, with three 330k resistors for biasing, which would have the zener current 480uA at the 500V idle voltage. I just got rid of one of the pmos to make the schematic more clear, I should have changed the resistors to 470k.

I haven't picked the fets yet, the voltage rail will be able to swing to about 950V, so I was planning 500-600V rated parts with three in cascode. Mouser/Digikey/RS/element14 all have a few examples, I haven't studied datasheets closely yet.

That will work but there has to be something better than a string of p-channel MOSFET cascodes.
Any suggestions on that front? I want to avoid extra isolated/floating supplies if at all possible, isolation amplifiers were something else I had considered, but most of the ones I looked at couldn't accept inputs all the way to the positive rail, so I couldn't use the floating zener trick.

I see that you are using a resistive bias source.  If the high side voltage is that stable, then why not replace all but one high voltage p-channel MOSFET cascode with stacked zener diodes?
That was not something I had considered. The rail will actually swing between about 50V and 950V (I was actually planning on 3 cascoded pmos rated 500-600V each, I just omitted the third for clarity), but the measurement will only ever be done while the rail is at a relatively stable 500V (not particularly low impedance source, so it will shift a bit with different current draw) during startup, so that could work.

We were discussing this circuit in a different forum thread and I was thinking about how to use PNP transistors while compensating for their bias current but that does not really gain much if anything in this situation since PNP transistors are limited in voltage rating also.
Yes, I noticed that thread just after I posted my own, I was certainly not expecting to see high voltage high side current sensing in a topic titled "high voltage op amp". I've got notifications turned on for the topic and will follow along of course.
« Last Edit: December 09, 2017, 09:05:50 pm by Greg Robinson »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 13318
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: High voltage, high side current sense amplifier.
« Reply #4 on: December 09, 2017, 09:21:39 pm »
That will work but there has to be something better than a string of p-channel MOSFET cascodes.

I see that you are using a resistive bias source.  If the high side voltage is that stable, then why not replace all but one high voltage p-channel MOSFET cascode with stacked zener diodes?

We were discussing this circuit in a different forum thread and I was thinking about how to use PNP transistors while compensating for their bias current but that does not really gain much if anything in this situation since PNP transistors are limited in voltage rating also.

Yeah, not so helpful, as HV BJTs have kind of low hFE.  'Least the current is low enough not to worry about 2nd breakdown.

Besides zeners, resistors can be used too, with the added restriction that output voltage and current need to be relatively stable to obtain that benefit.

Would be nice if there were a way to use NCH here, but it'd have to be, like, a folded cascode or something.  Which just makes the problem worse... :P

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 1268
  • Country: gb
Re: High voltage, high side current sense amplifier.
« Reply #5 on: December 10, 2017, 12:14:55 pm »
At this stage, i'd probably just select a MOSFET (IRFR9310?) and prototype.
 

Offline Greg Robinson

  • Regular Contributor
  • *
  • Posts: 82
  • Country: au
Re: High voltage, high side current sense amplifier.
« Reply #6 on: December 10, 2017, 12:55:21 pm »
I'll need to track down something with a DC SOA plot, as even with three devices in cascode, there may still be up to around 1mA through the fets with ~320V across them. I suspect most high voltage PMOS should manage this, but I don't want to risk it and don't have the time or money to characterise a device that doesn't specify linear operation.
 

Offline Greg Robinson

  • Regular Contributor
  • *
  • Posts: 82
  • Country: au
Re: High voltage, high side current sense amplifier.
« Reply #7 on: December 10, 2017, 01:21:15 pm »
And a quick look on Digikey, cheapest pmos in low quantities rated >400V, ZVP0545GTA, 450V, 2mA at 450VDC, three of those in cascode should be generous overkill.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf