DAC161S997: HART Connection with AD5700-1

Christina,

The connections look correct. Just to be sure, what is the loop supply voltage? Also do have the correct current going through the loop after you set up the DAC161S997?

Just to make sure there's a good connection from the HART modem, I'd use an oscilloscope to check the signal coming out of the AD5700 to make sure you see the 500mVpp signal. After that, check the signal at the C2 pin on the DAC161S997. This would be on the opposite side of the C68 capacitor. Going into the device, the signal is attenuated and should be about 15mVpp. 

What's the value of the sense resistor you're using on the loop? 

Joseph Wu

Hi Joseph,

thank you for the reply! The loop ist powered by 24V and I get the desired current, so it is working fine.

I checked the signal at C68. The input signal should be fine and around 500mVpp (blue). Right after C68, I cannot measure any signal beside some noise which ist approx 100mVpp (pink).

I have a Gateway in my loop with a built in 250 Ohm resistor and another 250 Ohm resistor to measure the signal inside the loop.

Is the value of the capacitance wrong?

Christina,

I looks to me that the capacitance values are correct. The schematic you put up matches the HART signal injection diagram from Figure 20 in the datasheet (C1=390nF, C2=220nF, C3=1nF, and the pass through cap is 6.8nF). Here's the diagram:

I'm not sure what prevents you from seeing the signal. In the past, I've run a board with this device, and I can see the 15mV signal at the C2 pin (and see the HART sinusoid across the load resistor).

To be sure, I would check a couple of things. First, I'd try to clean up the scope shot. I wouldn't have expected it to be so noisy. You might try using a different ground note. I think I probably used the DAC local ground instead of the supply ground. I would also use the scope's bandwidth limiting to get a cleaner signal.

If you still don't see a signal, you might try increasing the 6.8nF capacitance by an order of magnitude. You really shouldn't have to do this, but it should pass the signal better and see if there's a problem with one of the capacitances on your board. Then try measuring C2 again.

One last thing that I noticed is that you have a really large capacitance on the loop input node. C16 is 4.7uF, and that's a bit high. For most two-wire transmitter devices, the input capacitance/impedance is stays pretty high over a bandwidth, so the 4.7uF would really reduce that.

Joseph Wu

Hi Joseph,

I tried some different values for C68 an it turns out that my original smd capacitor was'nt working. Don't ask me why but the value itself was no problem. I still changed it to 10 nF since I was recieving some false values and that helped.

Can you explain the impact of C16? What value would you recommend?

Thanks so much! This really helped.