Current divider mirror

Do you mean things like this?

ftp://jjlarkin.lmi.net/Mirror1.JPG

ftp://jjlarkin.lmi.net/Mirror2.JPG

John

Version #1 is NOT accurately made variable/adjustable.

Version #2 is. ...Jim Thompson

--
| 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.

Are you looking for a totem pole current mirror?

No, offset is an issue. Example: a bipolar current mirror will work over as many decades as the transistors do (assuming equal temperatures, which works for monolithic, not as well for discrete, but is capable of working roughly), but an op-amp mirror gets lost in Vos. I'd rather not tune microvolts of offset, or amplify the noise. A bipolar mirror would work if I basically had a 1:100 variable emitter width.

Specifically I want something that works well from 1uA to 5mA within

+/-5 or 10V rails.

Tim

Specs? You have specs? How about noise, bamdwidth, cost? Do you intend it to be an adjustable ratio? How would it be adjusted?

Use a chopper amp and biggish resistors to keep the offset low.

John

Noise: low enough. It's charging a timing capacitor, so HF noise doesn't matter as much as 1/f noise. Let's say current noise of 2nA or less. Likewise, the current itself is a DC signal (< 100Hz), bandwidth doesn't really matter. The output needs ~MHz compliance, which won't matter around a drain / collector output (if nothing else, it can be cascoded). Cost: low, and should be simple (i.e., that rules out a logarithmic ADC/DSP/DAC).

Potentiometer, or something else panel adjustable. Doesn't have to be electronically controlled.

I suppose a chopper amp could be filtered to remove its noise. Something less than 1k would generate somewhat less than 5V at full current, and less than 100uV at the lowest range (assuming something like the opamp-FET-feedback resistor circuit).

On principle, I don't like using something additive (a linear amp) when the nature of the signal is logarithmic. So the obvious answer is a logarithmic converter, which is basically a glorified diode. Maybe a simplified log-exp amp would work anyway with reasonable stability, while keeping noise and offset sensible? Gain would also be electronically controllable by adding to the log signal. Of course, the simplest log-exp converter is the current mirror, but unequal current densities aren't temp compensated. Maybe they can be. Temp compensation isn't something I've explored much and may be worth a romp to figure out...

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

That is what a variable resistor in the emitter is for..

And introduces a TC if emitter:emitter and resistor:resistor aren't

1:1.

OpAmp driven transistor is the only way a variable resistor scheme is accurate and temperature stable. ...Jim Thompson

--
| 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.

out

The schemes John Larkin gave are unipolar, and 'current from +V or -V' sounds like bipolar to me.

If it's for bipolar current, use one op amp as current/voltage converter (this means the pseudoground input is your input load, which is ideal), then a second op amp with current mirrors on its power pins to convert back to a current; the ratio of resistors sets the current gain

(warning bad ASCII art follows)

+-----Rf----+ | |\ | I input +--------+---|- \ | | >--+--+ GND--|+ / | |/ | | | | +---------------------+ | (+V) | | | {PNP mirror} | | | | |\| | +------|+ \ +------- I_out=3D I_input*Rf/Rg | >-+ | GND --Rg---+----|- / | | | |/| | | | | | | +-----------+ | | | {NPN mirror} | (-V)

That's the topology of my MRI gradient amps. We sell units that go from 3 amps to 120 amps peak output. The big one peaks at about 20 kilowatts out.

That wouldn't be a bad audio amp topology, either. You can make the crossover essentially perfect, and it uses 100% of the available supplies, all biasing is tightly controlled, fet load matching is excellent, and it has some nice supervision/protection hooks. No, on second thought, this is way too precise for audio.

John

No, it's unipolar, I meant the currents are sourced from the +V rail, or sunk into the -V rail.

Your circuit reminds me of,

which isn't to be left alone on the breadboard unless the transistors are on a fairly large heatsink. ;-)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

Do you need continuously adjustable or can you live with discrete steps=20 on the current ratio?

It does very much look odd, the first time, looking at that big resistor to ground, and nothing else on the 'output' pin.

There's a bit of difficulty if the intended output is a low current; the quiescent op amp drain and the deviation from unity of the mirror gain interact. That makes an output offset current of Iq* (Gpnp - Gnpn). To trim that, you put a pot from the op amp V+ supply to the V- supply and ground the wiper (through a big resistor).

The original inquiry mentioned a range in the low microamps, will probably require such trim. It looks as odd as the 'output' resistor.

I use active mirrors, very precise, with an opamp per fet, so all the fet currents are very well controlled and matched. Various units use one to 16 fets per mirror.

ftp://jjlarkin.lmi.net/Amp.jpg

My quiescent current problem is the opposite: most opamps have too much Iq compared to Imax, so my mirror idle currents get too high. That's fixed by finding the right opamp and adding some off-direction offset to the mirrors. I do a pot from opamp V+ to V- to set fet idle currents.

The crossover graphs are esentially perfect.

John

Using the opamp rail currents gives essentially perfect crossover behavior. At zero signal, both mirrors run at their Iq. As you apply signal in one direction, one mirror current goes up and the other stays at Iq. That's different from most class AB stages, where as one side conducts, the other is turned off... which can lead to weird transient distortions.

John

Continuous, or sufficiently continuous (>= 8 bits?).

I suppose an alternate formulation might be, if it can't be varied reliably over a wide range, then it could be varied over a small dither range while the switching handles the wide range, but that would quickly get ugly..

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

.

-V'

or=20

ifer.gif

are on=20

And to think i cribbed that trick from datasheets and application=20 notes in 1965 to build my first self designed stereo.

This config is ideal for driving current into inductive loads, like MRI coils. The inherently high output impedance makes closed-loop dynamics mostly independent of load.

For audio, where you want a low output impedance, it's not ideal. The fix is to apply massive negative feedback, which requires tons of GBW to waste at higher frequencies. That was probably hard to do in 1965.

Got a schematic?

John

MC1524, actually was an I/C plus a 2N2222 and a 2N2906. ...Jim Thompson

--
| 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.