DC Power Line Sinusoidal ripple/noise Injection

Hi Thomas, 

This for the detailed explanation, that's really helpful. BTW, May I know what's the simulation software you using in attached pic? Thanks.

Regards,

Wang Xiao

Hi Xiao,

The OPA548 summing circuit was simulated using TI's free, TINA-TI Spice simulation software. It is very easy to learn and will allow you to create and simulate your own circuits. Here is the link where you can download the TINA-TI:

http://www.ti.com/tool/tina-ti&DCMP=hpa_amp_general&HQS=Other+OT+tina-ti

I also have attached my original OPA548 simulation circuit so you can evaluate the summing amplifier idea.

Regards, Thomas

PA - Linear Applications Engineering

Hi Xiao,

I was discussing your application with one of my colleagues who has a good deal of knowledge about power line communications (PLC). He thinks a possible solution to your application would be to couple, via a capacitor, the ac signal to the dc line. The impedance of the dc line may be very low at low ac frequencies but likely increases as the frequency is increased. You would need to amplify the ac signal from your DDS so you would still need a fast, high current operational amplifier - but maybe not as high of current as my summing circuit approach.

The OPA564 has a rated output current of 1.5 Amps, a supply voltage range of 7 to 26 V and a gain-bandwidth product of 17 MHz. This amplifier may be able to deliver enough current to the dc line well enough to simulate the ac noise condition you are after.

Regards, Thomas

PA - Linear Applications Engineering

Hi Thomas,

Do you think OPA564 will deliver enough power for capacitor coupled ripple injection? Why I am asking this is because one of our U.S. design team designed a tester board with such method for ripple/noise injection about 15 years ago. They was exactly using the transformer and capacitor coupling method to inject sinusoidal waveform. This design was using very bulky audio amplifier (4KW) for low frequency injection and RF amplifier (75W) for high frequency injection and the design declare that able to inject sinusoidal waveform up to 100Mhz. However, we have done some calculation and analysis and conclude that even with 75W RF amplifier, the frequency can only inject below 1Mhz. (I have attached our analysis document for you. And this setup is very bulky and old. That's why I was signed up to design a compact Power tester with modern electronics which capable for Voltage margin control, 20MSPS high speed voltage/current sampling and ripple/noise injection. The ripple/noise is the most challenge part for me.

One more concern on capacitor coupled method is at low frequency For example DC - 1khz, However if OPA564 method can prove works for frequency higher than 1Khz, that's will be great already. Because I have fully control on power margining (1V - 14.5V with 10mV step), I am thinking modulate DC/DC converter pwm output up and down to simulate low frequency ripple condition (less than 1khz). But higher frequency especially higher than PWM frequency. This method is no achievable. 

I am still in research on suitable method. You have helped a lot already. Appreciated.

Regards,

Wang Xiao

Hi Xiao,

Your feedback is helpful in understanding the challenges of your application. The impedance looking into the dc line is what will dictate how much current the ac amplifier would have to provide. It does sound like that impedance can be very low based on the description of the AF/RF power amplifier solution and the high amounts of power it had to deliver when driving the line. I question if the OPA564 rated current (1.5 Amps) would provide deliver enough current for your application.

There is another company that TI has no association with that specializes in high-power operational amplifier products, Apex Microtechnology. Take a look at what they have to offer; some of their amplifiers provide both high current and have wide bandwidth. They may have a suitable power operational amplifier that could be applied in the non-inverting summing amplifier circuit I provided, or be used to inject the ac current directly into the dc line via a coupling capacitor

Regards, Thomas

PA - Linear Applications Engineering

Hi Thomas,

Thanks for the help. I am contacting Apex look for suitable power amplifier

Regards,

Wang Xiao

Hi Thomas.

May I know how did you observer AM1 and Vout together, I can only observer voltage by oscilloscope.

By the way, I used Tina-TI.

Thank you.

Hi Hamilton,

I understand your question to mean AM1 and Vout in the TINA simulation results I had posted. AM1 is one of TINA's types of ammeters. It appears as a horizontal line with an arrowhead pointing to the right. TINA refers to it as the "current arrow" and it is listed within TINA's "Meters" pull-down. I use the "open circuit" symbol in the "Meters" pull-down for the voltage measurement.

When both of these meters are used in the circuit TINA will show their measurement results whenever you perform an dc, ac, or transient analysis.

Regards, Thomas

PA - Linear Applications Engineering

Hi Thomas.

That's very kind of you, after your explanation, I used transient analysis to get the AM1 value I want.

Thanks.