On LED slowly dies with AD5206 +, -, and GRND connected.

Hello,

I've been working through the AD5206 SPI communications tutorial. I had it working at one point. I started adapting it to the project I had in mind and ever since, the "On" LED on my Arduinos (tried three separate ones!) won't stay on. The second I connect the positive (5v) and negative (ground) to my circuit that LED either goes out, or dies a slow death. Things I've tried:

  • A Due with the SPI coms running off of the SPI header, and two Unos and SPI running on 10, 11, and 13.
  • Three different breadboards.
  • Uploaded the Blink example thinking it was my code. Still had the problem.
  • Minimized all of the connections to just the ones for the chip (+, -, and GRND). I've attached a photo of it with SPI hooked up as well. Blink was running at the time so notice: (1) 2V on the rails from the 5V source. If I used rails with nothing on them, I get 5V, so I know the chip is consuming power; (2) The lack of lights for LED 13 and On.
  • Yes, GRND (pin 4) is dodgy in that last one as it is connected to + in the tutorial. Connecting it to - didn't make a difference though. I did check the documentation on the chip to confirm it should be ground and it should.
  • Flipped the chips around.
  • Regular power supply rather than powering off of my computer.
  • About 11 AD5206 chips... my project was going to use a lot.
  • Used winAVR and avrdude to confirm the fuses are ok.
  • Searched the forums and read every AD5206 post I could find. Also tried finding info on the LEDs going off, but the blink example fills all of the results.

I'm a programmer who can solder but not much of an electronics person, beyond building my own PCs. I get that I've backed myself into a corner with all of this... What's left to replace when I've done everything but the kitchen sink? I'd be grateful if any of you can think of something I haven't tried, a common factor I haven't seen, or could just walk me through a way to get some diags on what is going on. I'm new to avrdude but happy to dig in more if it'll be worth it.

Thank you!

Jason

A power LED that dies on connection usually means a short-circuit. Could you post a photo of your set-up?

Photo's up there now. The initial post rejected it due to the size limit. Thanks.

The breadboard on the Arduino page has a different + and - supply rail than YOUR breadboard.

You have just blindly coppied that, so your IC's supply is reversed.
I hope the IC will survive that.
Leo..

edit.
I also have other doubts about that Arduino page.
IMHO, this IC should not be used to drive LEDs.
The IC's absolute max pot current according to the datasheet is 11mA.
I calculate ~14.5mA with a standard red LEDs and a 220ohm resistor.

Also, do yourself a favour and put a DMM on your Christmas list.
That 2Kohm/volt meter could easily introduce reading errors.
At that 10volt range, it always puts a 10k resistor across everything you're measuring.
Good for battery/supply testing, but not ok for anything else.

I knew it was going to be something stupid... thanks Wawa. That was it! The one time I had it working I must have reversed the wires by mistake! Thank you! I'll put a DMM on my list as well. I didn't know that about them.

Not to hijack my own thread, but your point about the 11mA is worrying me a bit and is probably saving me some trouble later on. I need to cover 0-200mA. Is there a way to step it up, or should I look for a different chip?

What are you trying to do with this chip.
Drive LEDs?

If so, there are much better solutions for that.
Leo..

It's for an experiment I'm working on, though I'm just a hobbyist so it's been slow going. It'll be powering an electric mesh, but I want the Arduino to control the amount of power based on some sensors I have. 200mA is about the tops it'll need to put into the mesh though.

I have no idea what that "electric mesh" means.
Had to googled it, and saw "electric fences".
Tell us what you want to do. Maybe we can throw in some ideas.
Leo..

That's actually close. Think of mosquito netting, but made from a conductive wire such as copper. I want to pass a very small amount of current through it to generate a small magnetic field. However, I need to be able to change the amount of current from 200mA down to 0, if possible. Second best would be having a base amount of 150mA so I could cover a range above and below it. Either way, it's something I want the Arduino to manage based on the input of two sets of sensors which will set the upper and lower bounds for the needed current range.

First you need to know the resistance of your "mesh", so you can pick a suitable voltage source.
Then you think of measuring/controling it.
Only one current sensor is needed. Boundaries can be set in software.
Leo..

Have you calculated how small that magnetic field will be? The Earth's field will totally swamp it and any experement will need to be shielded from it in order to see anything.