Current Limiting for High side mosfet switch.

I have a high side P Channel Mosfet switch.

How do I limit switched current to 4 Amps?

The problems are:

- I need small shunt resistor due to power dissipation limit under

1/2W, but that reduces the drop across the shunt resistor. Hence Vbe technique fails.

- I can use an opamp differential comparator across the shunt, but the problem is that the high side of the shunt is close enough to the supply voltage and unfortunately there are no truly rail-to-rail input comparators I could find (Even those CMOS ones have 0.2V offset from rail).

Is there a way to do this simply?

Mike

Reply to
siliconmike
Loading thread data ...

--
Hall effect sensor?
Reply to
John Fields

Yeah, kinda. instead of wiring the comparator directly across the shunt you use a divider string to bring both sides down just enough to keep it within the rail limitations of the comparator.

Shunt is in the high side line somewhere. One (relatively low value) resistor from each side of the shunt forms a voltage divider with one each relatively high valuer resistor to ground. The junction forms the input to the comparator - all you've done is removed the comparator from the need to sense close to the supply rail by droping the inputs down a little.

You do sacrifice a little gain that way but the open loop gain of the comparator is on the order of 200K so that shouldn't be a problem. I don't see why a pair of diodes or diode strings to just offset the voltage by .6 volts below the high supply rail wouldn't work also if you are just limiting current and don't care about measuring it. Same as the above example but diodes in place of the low value resistors connected to the shunt.

I did something like this awhile back to add an electronic ammeter to a motorcycle. It lowered the signal into my op amps and I had to recalculate the gain but it worked like a champ, Didn't try using diodes though.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Reply to
default

The simple way is to hang an isolated +/- power supply across the + rail and run all the electronics from there. Those little 1W DC-DC converters are cheap enough. john

Reply to
John Jardine.

The LM101 op amp has an input voltage range that includes V+. I've used it in power supplies I designed a long time ago to form part of the current-limiting circuitry. Can you still get them>

Graham H

Reply to
Graham Holloway

Can be as simple as two PNP transistors and a few resistors...

What is the value and how stable is the power supply voltage?

Value of shunt resistor?

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Last time I did this, I used a small shunt resistor and an LM7301 r-r opamp as the "comparator". It directly drove the gate of the current-limit fet, although you may need some logic if the current-limiter fet is also a controlled switch. You can power the amp from the + rail and ground, up to 30 volts or so, I think.

John

Reply to
John Larkin

Yep, but have a look at linear's LTC6101, it's a high side current monitor, that may be useful to you. You will need a comparator as well, to wiggle the fet

martin

Reply to
martin griffith

Hello Jim,

Absolutamente, as long as the limit doesn't have to be very precise. I'd do it with only one PNP up there and pre-bias it a bit so that the shunt only drops a couple hundred mV or whatever is acceptable. Then a resistor from its collector to a FET, NPN or whatever down at logic level, with a bleeder resistor to GND.

If really hard pressed on the BOM budget there is the option of, gasp, running the resistor from that PNP's collector right into a CMOS Schmitt input and let it ease into the VCC substrate diode upon conduction. Needs a bleeder as well.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

If you use two PNP's you lose the temperature problems and can easily detect accurately a 100mV drop.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Hello Jim,

Yes, if you need more than just a crude limit against a catastrophic event two transistors would be better.

A "reverse" TLV431 would be a really sweet thing to have here.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

Don't know about the LM101 - which was specified for the military -55C to +125C temeprature range and always came in the TO-5 can - but Farnell still stock the LM301AN for a euro (in small quantities) - at the moment that's $1.28 in God's own currency.

There is also the Burr-Brown INA159 which is a difference amplifier aimed at this sort of job.

formatting link

--
Bill Sloman, Nijmegen
Reply to
bill.sloman

Thank you all for your valuable suggestions. First, hall effect and 'high side op-amp' techniques are out of budget.

Next, the max shunt drop could be around 100 mV at 4 Amps.

Power supply is the 12V Auto battery.

Jim, are you suggesting a differential amplifier? If yes are there any cheap dual transistor packages out there?

Could you post a schematic?

Thanks, Mike

Reply to
siliconmike

Application note AN39.

See the section on their ZXCT100xx range.

--
Tony Williams.
Reply to
Tony Williams

--
How much can you spend?
Reply to
John Fields

That is so so so so boring!!!

Reply to
Fred Bloggs

What a waste you are, Fred.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Well, yes.

John

Reply to
John Larkin

I'll draw you a picture, conceptually, adapt to your current...

formatting link

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

But, does it get the job done?

Thanks, Rich

Reply to
Rich Grise

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.