# Measure low current with variable voltage

I want to make a system that always have the same current (about 1ma) in a water with silver electrode. At start it will be purified water so conductivity will be really low.

But with time silver particle will be released in water, raising his conductivity so if I don't lower the voltage the current will raise too (meaning that bigger particle will be released which I don't want too).

So, is there any way I can measure low current and adapt voltage with the result (since deducting resistance wouldn't be hard).

I know about ASC712 but It seem that he can't be precise enough (steps of 13ma max) to my use.

I precise it's not that long I'm trying to get back in electronics with arduino so many school things have been forgotten.

What you want is called a constant current source. You can actually build a simple one using one transistor ; but a better one uses two transistors - as in the left example below - see this page for explanations.

Here the leds would be your load, R=680 Ohm would give you about 1mA - and you would need to increase the value of the 2k2 resistor to somewhere around 22k.

The problem is that, as John Williams noted, forcing 1mA through a high resistance requires high voltage : 1MOhm would need 1kV, 10MOhm needs 10kV etc. That's very dangerous voltages! and quite some power as well : 10kV at 1mA means 10W.

I think you need to define your needs first :

• What is a tolerable error on the absolute value of current? (Would constant 1.1mA be OK?)
• What is a tolerable error on the current variations? (Would current varying between 1.19mA and 1.21mA be OK?)
• Do you absolutely need 1mA right from start? Or can you accept a lower current at start, growing with increased conductivity until 1mA is reached and then being held nicely constant?

If the above examples are fine, a current source as described above would do the job nicely for less than 1USD.

If you really need a constant current source right from the start, then be prepared to work with expensive and dangerous pieces of equipment.

• I maybe not need 1mA from the start I think that's mostly for the end of the process but that's not very well documented anywhere :/. And I don't need to be precisely 1ma either, I even think that I can go to 10 or 50ma without issue but again there is no really exact value to refer out there Commented Aug 2, 2014 at 13:21

Alternatively, you can buy ready made unit. Random search from web give example like this has 3 models for, 1.5, 3 and 6kV 1mA constant current. http://www.matsusada.com/high-voltage/sc/

Safety first if self make (do you need a unit for use or you want to do exercise trying to make it). May not be a easy project.

Your post stated 'purified water', resistance will be really high, in many mega ohm range, if it is really pure.

First measure resistance with multitester to get realistic ohm reading in pure and later stage when "with time silver particle will be released in water".

The hi and then lower ohm will get you voltage needed. to drive 1mA through multiple mega ohm need kilo volt and need special circuit and suitable experience for hi volt safety.

• if you may describe more about your intended application, people may be more help in alternate idea/method. Like, it that hobby or 'real' use? Why you need to drive that much current into purified water? At 10 watts, 10KV, it will be quite a bit of hydrogen/ oxygen gas for REAL explosion risk. Today news, gas explosion kill 100, hurt 300, damaged kilo meters of road and car flied 30 foot and landed on rooftop.
– EEd
Commented Aug 2, 2014 at 20:32
• Correction. 10 watts make a lot of gas at more typical low volt case. At high voltage (if purified in original post refers to distilled or RO water, at commercial/industrial/factory/lab quality level), 1 to 50mA, gas quantity is much less.
– EEd
Commented Aug 3, 2014 at 3:13