He's been told this before, but it's too difficult!!!
He needs his 'webmaster' to create a folder on his business website for personal use, where he can stick things.
-- It's easy to think outside the box, when you have a cutting torch.
You don't necessary have to get that low for Rtop: why do you want 3mA through the transistor? 300uA FS would be plenty enough, and less power dissipated in it, depending on how high is your voltage source.
No worries about the DS leakage since it'll be pretty small and included in the source current anyway, so won't be an error source :-)
Why not setting Rset at 10K?
Plenty of 5V low cost CMOS opamps. The AD8541 has 100pA max Ibias @ 85°C and typical 100fA/°C from room temp to 85°C. For your kind of needed accuracy, just drop Rfb. (if I've understood well enough)
No, you don't know the opamp offset polarity. It being 6mV max, 7mV over temp, that's half an amp on your 15mR shunt.
The circuit will work if the offset has the right sign, but would have a huge up to 0.5A sticking dead zone at origin if not. You definitely need the added offset. Say circa 10mV (which BTW makes interesting the lower drain current).
If your voltage source isn't stable enough (how much relative variation, BTW? It's just you battery voltage, so...), you have a regulated PSU for your uC. Just use it:
Vsource V | .-. | | Ioffset | | V '-' | | | +----. | | | | \| | |/ |--+---| | | VCC ------' .-. | | | | '-' | === GND
Or maybe even delete the diode connected BJT and tie the other's base to Vcc, if it's high enough so the 2mV/°C over temp range gives you a good enough reading stability. Hmmm, wait, if you do the zeroing say once a second, you don't need the second transistor at all. And if your source voltage is also stable enough over the offset sampling period, you don't even need the first transistor :-)
Now, to power the opamp, just use a 5V6/5V1 zener diode. And you have a nice stable offset ref... So:
___ Load BSS84 | | |/| ||-+ | | ___ | | | ___ '-|___|--+-------------|---------+-|___|->GND 470K | o-----> current ___ | Vref o-------|___|----o Rshift | .-. | | | | 10K '-' | ===
And in case you worry about the bias current switch to a 100R/47K bias network. Or maybe just do it anyway...
-- Thanks, Fred.
Have a look at my last circuit :-)
OK, no LM358, but...
Well, maybe even the 358 will do, but you'll have to work that out :-)
-- Thanks, Fred.
this then?
.97$ @500
-Lasse
That needs one more resistor, to define the zener current.
Or use a 79L05 upside-down.
Doesn't need Vref/Rshift, either.
The opamp/BSS84 loop might oscillate, one of a couple of ways.
Hey, this might work.
This is not an attempt to be a diffamp; the resistor to ground just adds a small offset to keep the zero live. The output is referenced to V+ and drops as load current increases. Presumably the uP is already digitizing V+, so it can do the "common-mode rejection" in math.
John
But higher cost than a zener+R
Depends on the ADC (assuming it's going into one) input CM range, and we don't know about that.
Not so: no C load on source and the CMOS RR opamps have pretty high and lossy output Z. I have exactly that running into prod right now and have tested them thoroughly. At 10K the source R is a bit higher though. If you're paranoid enough it's better to zobel the gate to the + source rail. Hey, I just invented a new english verb... Now that's really fun :-)
-- Thanks, Fred.
It's kinda comparing apples to oranges (it's nearly $2.00 in single-piece quantities, as opposed to $0.60 for a 'cheap' op-amp). But it sure scrapes some components off the board!
-- www.wescottdesign.com
I've seen both mosfets and bipolars do high-frequency oscillation when driven from opamp outputs, particularly with resistance in the source/emitter. I figured that the opamp Zout would replace a gate/base resistor, and was wrong.
The overall loop should be checked, too.
Hey, you're not allowed to do that!
John
Big snip somewhere...
Maybe not, but that circuit is even more fun :-)
And the AD8541 with it's 45typ/70uA max Iq will fit nicely. Still, the input offset offseting network is required.
___ Load
[snip[
Save me going thru all the thread... what's the voltage at the input? ...Jim Thompson
[On the Road, in New York]
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Varies between 5V and 12V.
-- www.wescottdesign.com
me
s=A0
e___|---o----------------------o-------> to load >
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 =A0.-.
=A0 =A0 =A0 =A0 =A0 =A0 =A0| |
=A0 =A0 =A0 =A0| | Rtop
=A0 =A0 =A0 =A0 =A0 =A0 =A0'-'
=A0 =A0 =A0 =A0Rfb =A0 =A0 |
=A0 =A0 =A0 =A0___ =A0 =A0 |
___|----o
=A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 |
-. =A0 =A0 =A0|
=A0 =A0 =A0| =A0 =A0 =A0|
=A0C1 =A0 | =A0 =A0 =A0|
=A0 =A0 =A0 | =A0 =A0 =A0|
=A0 =A0|<
| =A0>-----o----| =A0 Q1
=A0 =A0 =A0 =A0 =A0 |\ =A0MMBT3906
=A0|/ =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 U1 =A0 =A0 =A0 =A0 =A0 =A0 |
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0o-----> current V >
=A0 =A0 =A0 =A0 =A0___ =A0 =A0 |
---|___|----o
=A0 =A0 =A0 =A0 Rshift =A0 |
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 .-.
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | |
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | | Rset
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 '-'
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 GND
ISTR Zetex has a complementary current mirror ZBT1009 or something like that with the capacity to do hi side sensing. regards, al
[snip]
Do it like this, then measure P-P using the micro....
Subject: Wescott Current Measurement (S.E.D) - WescottCurrentMeasurement.pdf (1/1) Date: Fri, 29 Jul 2011 20:17:52 -0400 Message-ID:
(Chopper could be 4066) ...Jim Thompson
[On the Road, in New York]
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
As long as there's nothing you care about on that machine, or on any machine it's connected to inside your firewall, you have no security problems. You can put an old sacrificial machine in the DMZ of your firewall, for instance. I pay GoDaddy about $5 a month for hosting electrooptical.net, so it's their worry.
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 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
Cool. But dump the fet, dump Rshift, and maybe put a cap across the
10K. Like this:
But the zener just exploded and we're all dead.
John
The machine belongs to our ISP, which is LMI in Berkeley. They gave me some ftp space just because they like me, I guess.
John
Sure, but we were talking about a web site--which you can attach to your own network, in which case you have to worry about the vulnerability of the host machine, or have hosted someplace as you do with your FTP site.
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 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
They host our web site and do our email, too. If I do a personal web site some day (grrr, snarl) I'll just do one of those google things or something like that. Mo has her little business on a Yahoo-hosted web site, all drag-and-drop, with her own URL, and that seems to work fine.
John
Late at night, by candle light, Tim Wescott penned this immortal opus:
Except that at power-up you put voltage on the input pins through a low impedance a bit before the converter's charged up and powering the opamp. Better watch out for latch-up and other unexpected effects, like phut-bang!
- YD.
-- Remove HAT if replying by mail.
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.