el cheapo micro-ammeter

[dannyf]

In a separate discussion, someone asked how to measure small current. So I thought I would experiment with a cheap ammeter that can go down to ua range -> it actually has the potential to go further down but we will not discuss that now.

The design goal is:

1) cheap -> utilize easy to get jfet or cmos opamps. I will start with a TL082 now but will migrate to LMC6042 later;
2) easy to build -> dual opamp, minimal parts count, perfboard construction;
3) work as an adapter for other meters, like a dmm.

[dannyf]

The basic topology is simple: an opamp plus a big resistor -> virtual ground feedback ammeter.

we will use another opamp as rail spliter. Battery powered -> TL082 consumes 5 ma so 60 hours of operation. LMC6042 consumes 30ua and no need for a battery switch.

[george graves]

subscribed

[dannyf]

I already built a prototype on a breadboard, with a 12v source.

When measuring current in a 9v battery + 1.5Mohm resistor, the output is 0.666v -> fairly close to what you would expect.

I will share a picture later.

[dannyf]

Range selection will be done by wiring different value R3.

10k  -> 1v/100ua
100k -> 1v/10ua
1Meg -> 1v/1ua
10Meg-> 1v/0.1ua

Anything below that gets a little bit tricky as the Ibias for TL082 is 30pa+, depending on the source of the datasheet. Now, Ibias doesn't change very fast so if you just zeroed the meter, the measurement is still good.

But that becomes tedious. A low Ibias opamp would be needed if you intend to go below 1v/1ua. That's where something like LMC6042 would come into play, both for its super low Ibias (2fa) and low current consumption (20ua or so if I recall right).

[dannyf]

The construction will be a little bit tricky, if you wish to go below 1v/1ua: guard ring would be a must as you cannot trust most pcb substrates.

For me, I would just use 3d / point-to-point construction on the input pins.

[dannyf]

Some pictures.

First, the micro-ammeter on a breadboard. Red arrow shows the range setting resistor R3 (=100k here). Green box shows Ibias generator.

[dannyf]

Running the meter at ~10v.

Measuring current in a loop with 9v battery + 1.5Mohm resistor -> 0.6v output expected. We measured twice, reversing the battery polarity.

If the meter is perfectly zeroed, we would expect identical readings and with opposite signs.

[dannyf]

Same meter, unzeroed, running under 5v. Swinging shouldn't be a problem but we will have offset. So the difference between the two readings should be about 1140 (571mv - (569mv) = 1140mv).

We got two readings, 539mv and (599mv), for a difference of 1138mv, vs. 1140mv expected. Not bad.

[dannyf]

The Ibias generator has an adjustment range of about +/- 17mv, at 10v power supply, vs Vos of 10mv.

That range is greatly reduced at 5v so I could never zero the readings with this particular Ibias generator. However, if I were to reduce the value of R6 to 1Meg, zeroing becomes possible.

So when you design for your particular implementation, keep that in mind.

[dannyf]

You may also notice that I used 100K resistors rather than 1Meg resistor for my build, purely for convenience as I don't have 1Meg resistors handy. It also tells you how unimportant those resistors are, value-wise.

I will play with other dual jfet opamps and see how that they perform.

[dannyf]

Here is the meter utilizing a KF353 (LF353 by Fairchild). LF353 has higher Vos so I was not able to quite zero the meter (1mv off) but good enough.

So the meter is actually quite tolerant of the opamp choices. I am sufficiently comfident that if you build it around other jfet opamps (including the quads) it would work just as well.

[dannyf]

Voltage sensitivity:

Ran the meter at 12v and then 10v on the same current.

Reading: 5.71ua vs. 5.72ua.

[dannyf]

Identical implementation by someone else: http://www.vk2zay.net/article/251

His unit was built around lmc662(the same family with LMC6042.

Notice how the constructed the input? Absolutely necessary for a picometer.

Well done.

[Vgkid]

Is the bias generator used to null out the opamp input bias currents?

[dannyf]

Yes.

The difficult with nulling that current is that its direction changes from one opamp to another. The schematic showed a Vcc connection and if you notice my breadboard construction it was actually connected to Vee.

Just something to keep in mind.

[dannyf]

Measuring the photodiode current from a led, using a 10Meg ohm feedback resistor -> 1v/0.1ua. The 9v battery is not in the circuit.

6.5mv -> 0.65na.

The readings would fluctuate under a led lightbulb, and as you wave your hands. So in real life, shielding would be a must.

[dannyf]

The reading when the led is reverse biased by the 9v battery is 11mv -> 1.1na.

[babysitter]

If you are on a budget and dont wan't to solder, there is something else you have to measure low currents.El-even-cheaper.
Basically a voltmeter in parallel with 10 Megaohm...a voltmeter in parallel with 10 Megaohm...a voltmeter in parallel with 10 Megaohm...

Yes, your garden variety dmm with 10 Megaohm impedance :-)

(exact value may vary but no one prohibits you from finding out )

[Kalvin]

Identical implementation by someone else: http://www.vk2zay.net/article/251

His unit was built around lmc662(the same family with LMC6042.

Notice how the constructed the input? Absolutely necessary for a picometer.

Well done.

The 1 gigaohm resistor maybe something which is not easily found. You can split that huge resistor into three smaller ones (See R2, R3 and R4 in the schematics attached). The schematics shows a measurement circuit for 1nA using more common component values. For the simulation purposes the voltage source provides a 1V step voltage which is fed through a large (1 gigaohm) resistor into the measurement circuitry giving a 1nA to be measured.

[DmitryL]

In a separate discussion, someone asked how to measure small current. So I thought I would experiment with a cheap ammeter that can go down to ua range -> it actually has the potential to go further down but we will not discuss that now.
1) cheap -> utilize easy to get jfet or cmos opamps. I will start with a TL082 now but will migrate to LMC6042 later;
2) easy to build -> dual opamp, minimal parts count, perfboard construction;
3) work as an adapter for other meters, like a dmm.

What's wrong with existing uCurrent design ? Works pretty well.

[dannyf]

You can split that huge resistor into three smaller ones

In an I-V converter, you try your best to avoid the use of small value resistors to minimize johnson noise. If you look at commercial picometers, they all use high value resistors, some as high as 1Tohm.

Yes, they are expensive, though glass resistors are a lot more available now, especially from China. The polymer types however aren't that great.

[dannyf]

Put an electrolytic capacitor under test, for its leakage current.

Feedback capacitor is 100Kohm. So gain is 1v/10ua.

When the capacitor is charged under the right polarity, the leakage current is 0.34ua - picture on the right;
When the capacitor is charged under the wrong polarity, the leakage current is 11.7ua - picture on the left, ~40x bigger.

Really leaky capacitors, if I may say so.

[dannyf]

What's wrong with existing uCurrent design ?

I haven't seen it.

[DmitryL]

What's wrong with existing uCurrent design ?

I haven't seen it.

LOL! What a fun try to search by this keyword on this site.