Positive and Negative Adjustable Current Supply (3.3mA)

[ericdmurphy]

Hello, I was looking for some sort of chip/IC that would accept a digital input of say 4-5 bits and that would adjust an output current. I am looking to have a +/-3.3mA current applied to a resistive load. It would also be helpfull (for testing purposes) if I could adjust the inputs so that I could get a resolution of .1mA. The final project will be set somewhere between 3-4mA. I have tried using LM317 regulators as well as just a constant voltage through a 1% resistor. These don't provide quite steady of a current out for me. Also, because the positive and negative sides aren't directly linked, they tend to drift a little. Any ideas?

 

[berkeman]

Hello, I was looking for some sort of chip/IC that would accept a digital input of say 4-5 bits and that would adjust an output current. I am looking to have a +/-3.3mA current applied to a resistive load. It would also be helpfull (for testing purposes) if I could adjust the inputs so that I could get a resolution of .1mA. The final project will be set somewhere between 3-4mA. I have tried using LM317 regulators as well as just a constant voltage through a 1% resistor. These don't provide quite steady of a current out for me. Also, because the positive and negative sides aren't directly linked, they tend to drift a little. Any ideas?

Welcome to the PF.

Some DACs have current-mode outputs. Have you looked into using a DAC?

 

[ericdmurphy]

Thanks for the quick reply.

In theory, I understand how they work, but I have never used them in practice. Any examples of one that I could look into, maybe give you a better idea of what aspects that I like/dislike about it and why it may or may not work?

I looked over a couple on digikey, and it seems that the output is an adjustable voltage. Would that then have to put through a resister? That may work. But without having the exact same resistor (can get pretty close, but never exact) the positive and negative sides wouldn't match. Is there any that provide a current output?

 

[berkeman]

I've worked with them in the past, but it's been a while. I just now did a Google Images search on dac with bipolar current output, and got some good hits. Maybe try that search to see if you see a circuit that will work for you.

 

[ericdmurphy]

Bipolar current output, deferentially haven't tried that search phrase yet. I'll check it out.

 

[donpacino]

another note, you can use the voltage level dac you mentioned earlier and combine it with a voltage to current amplifier

 

[ericdmurphy]

I'm currently looking up digital pot's with dual outputs that are non-serial interface controlled. The DAC's seem like they may be a little more than I need.

donpacino - any idea of a circuit to give me a better idea of what to look for? Any specific amp?

Actually, while we are on the topic of amps: This whole project is basically a high speed fully differential instrumentation amplifier with adjustable gain and VERY low noise. Sometimes the input I have to deal with is on the same level as the noise. Makes the output very difficult to ascertain. The current device that is in play (that I am trying to re-create and eventually make better and faster, stronger, newer, all that.) is the INA110. Any ideas for a circuit or better replacement amplifier?

 

[meBigGuy]

You are trying to build a better amp than the INA110 out of discrete components? Good luck with that!

What specific specs do you need to improve (what new specs do you require)?

http://cds.linear.com/docs/en/datasheet/1102fb.pdf
http://www.ti.com/lit/ds/symlink/ina111.pdf

 

[ericdmurphy]

Actually I said quite the opposite. I am looking for a better instrumentation amplifier. Price isn't really relevant. I don't care if the chip cost $50 each. As long as its quick, and extremely low noise. With the INA110 right now, the engineers from 88' developed a circuit board that on the output, at G=100, the noise is about 2mv pk-pk. Right now my design, on a breadboard, in a metallic grounded enclosure is getting around 4-5mv pk-pk. I realize that once I make this on a printed circuit board it will improve some of the noise just because of the components being soldered and I can place grounding strips running parallel to the input's and outputs. However, if I can get something that is better, faster, less noise, and just an all around better chip and therefore a better amplifier, that would be ideal. I should also point out that the schematic from 1988 has been lost for some time, hence me re-engineering the wheel.

Also, any ideas about cleaning up the noise in general would be of great assistance.

Thanks again for the reply's and advice.

Eric.

 

[meBigGuy]

I am looking for a better instrumentation amplifier.

Better in what way? WHat specific specs needs to be better, and by how much?
Sending an email to TI or ADI support might clue you into their newest products.

Or, maybe you do need to roll your own out of high performance amps.
http://www.analog.com/media/en/trai...dbooks/5812756674312778737Complete_In_Amp.pdf

I listed two different in-amps in a previous post. Were those not "better"?
Is this one "better"?
http://www.digikey.com/en/product-highlight/a/analog-devices/ad8421-instrumentation-amplifier

Have you tried the parametric searches at different vendors?
http://www.analog.com/en/parametricsearch/10078#/p4133=In-Amp

No one seems to have that simple answer you are looking for "OH, just change to the XXX or do YYY". Sorry.

 

[ericdmurphy]

The INA110 works well for what we need it to do. I am just looking for something that has less noise, or faster, or both. I looked at the INA111, that definitely seems to be a good possibility. The fixed gains of the other amp amp won't cut it though. The sets gains of the INA110 is a good starting point, I do like the idea of being able to throw any resister in there and get any gain that I would like though. I will probably order up a few of the INA111's and give them a shot.

I have another question for any of you EE types that may have made your own PCB in the past. I've been downloading almost every software I can find today, and I haven't found any that are very intuitive, and that have the footprints and the parts that I've been using. Is there a program out there that allows you to slap down some generic footprints, choose some static locations for some specific ones, connect some traces, and have it auto-route the traces in the most intelligent manor?

I'm not looking for some professional quality board with all the bells and whistles, just something I can have made up that I can solder all the parts that I already have to and see if the board works.

 

[berkeman]

Here are a couple previous PCB layout threads that may give you some ideas...

https://www.physicsforums.com/threads/pcb-design-software.89198/#post-750947
https://www.physicsforums.com/threads/new-to-pcb-help.418407/#post-2815114

 

[meBigGuy]

Did you even look at the AD parametric link I posted? You can filter on how the gain is determined, etc

Did you look at the AD8421 spec I linked to? What is it you don't like about it?

I asked about what specs you want better for a reason.

another possibility is:
http://www.analog.com/media/en/technical-documentation/data-sheets/AD8429.pdf

Am I wasting my time?