AC constant current source

Discussion in 'The Projects Forum' started by mcike, May 27, 2011.

1. mcike Thread Starter New Member

Apr 17, 2011
12
0
Hello,
unfortunately i'm quite a newbie concerning electronics.
I want to design a constant AC current source of current variable up to 100mA rms.
The frequencyrange of the current source should be from 1Hz to 100KHz. The load is mounted in a nitrogen-flow-cryostat an therfore will change it's resistance with temperature between 50 and 100Ohms.

The AC input signal originates from a lockin amplifiers Oscillator output, and can be chosen between 0 and 4Vrms. one important feature of this current source should be a very low excitation of higher harmonics.

i already found a few papers which describe how in principle such a current source can be built using 1 or 2 operational amplifiers.

but unfortunately all the papers were outdated or written in a very general way. so my question would be, how to build such a current source in practice, and which operational amplifiers should i use??

i would be grateful for any help.
thanks, matthias.

2. #12 Expert

Nov 30, 2010
16,657
7,302
I'm having difficulty imagining an AC constant current source. Do you mean constant current at all phase angles of the sine wave, but reversing polarity, which would be a square wave into a fixed load? Or are you going to use a load that varies resistance as fast as the sine wave?

3. mcike Thread Starter New Member

Apr 17, 2011
12
0
i'm sorry for my unclear declaration of my problem. with constant i wanted to say, independant of load resistance.

4. ErnieM AAC Fanatic!

Apr 24, 2011
7,434
1,625
I get what you mean by a constant current source, so this 0-4Vrms AC signal is going to set the current supplied.

Here's a general scheme: U1 is an op amp capable of the high current output, or a more general purpose op amp followed by a current buffer.

U2 is a differential amplifier that senses the voltage across R1. U1 will then drive R1 such that the input voltage is put on R1, giving an output current of Vin/R1.

With Vin max at 4V and R1 at 40 ohms, I max = 4/40 = 100 mA.

mcike likes this.
5. #12 Expert

Nov 30, 2010
16,657
7,302
Something like this?

File size:
30.5 KB
Views:
287
6. mcike Thread Starter New Member

Apr 17, 2011
12
0
Thanks for your quick help!

still i would be thankful, if somebody had recomendations about specific OPamps, which i could use. beside the already mentioned requirements, offsetvolatges on the load should be prevented as good as possible.

thanks, matthias.

7. Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
You can do it, in principle, with a Howland current pump. However, you will need an op amp that can put out ±150mA and ±15V at up to 100kHz. This might require a combination of an op amp and a high-current buffer, and a heat sink.

mcike likes this.
8. #12 Expert

Nov 30, 2010
16,657
7,302
You are welcome.

9. Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
You need to change the feedback. As drawn, it is positive.
You might also be hard pressed to get 100kHz bandwidth with two op amps in a feedback loop.

10. ErnieM AAC Fanatic!

Apr 24, 2011
7,434
1,625
You are quite correct sir. My bad.

However, the Improved Howland Current Pump looks like a much better configuration.

Last edited: May 28, 2011
11. Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
I've been working on (designing and simulating) a circuit, which is working well, but it's not probably something a novice should attempt. The circuit is a straightforward improved Howland ckt, but the op amp is a SMD (DDPAK) with a tab that needs to be soldered to the PC board, and it would have to be reflow soldered, because the tab is not accessible. I think it would run without heat sinking, but the temp would be about 65-70°C above ambient - a finger-blistering hazard, at the very least.
Maybe I can find a more robust op amp with lower junction-ambient thermal resistance.

Last edited: May 28, 2011

Aug 23, 2012
4,924
777